Medicare Program; Merit-Based Incentive Payment System (MIPS) and Alternative Payment Model (APM) Incentive Under the Physician Fee Schedule, and Criteria for Physician-Focused Payment Models

AGENCY:

Centers for Medicare & Medicaid Services (CMS), HHS.

ACTION:

Proposed rule.

SUMMARY:

Medicare Access and CHIP Reauthorization Act of 2015 (MACRA) repeals the Medicare sustainable growth rate (SGR) methodology for updates to the physician fee schedule (PFS) and replaces it with a new Merit-based Incentive Payment System (MIPS) for MIPS eligible clinicians or groups under the PFS. This proposed rule would establish the MIPS, a new program for certain Medicare-enrolled practitioners. MIPS would consolidate components of three existing programs, the Physician Quality Reporting System (PQRS), the Physician Value-based Payment Modifier (VM), and the Medicare Electronic Health Record (EHR) Incentive Program for Eligible Professionals (EPs), and would continue the focus on quality, resource use, and use of certified EHR technology (CEHRT) in a cohesive program that avoids redundancies. This proposed rule also would establish incentives for participation in certain alternative payment models (APMs) and includes proposed criteria for use by the Physician-Focused Payment Model Technical Advisory Committee (PTAC) in making comments and recommendations on physician-focused payment models. In this proposed rule we have rebranded key terminology based on feedback from stakeholders, with the goal of selecting terms that would be more easily identified and understood by our stakeholders.

DATES:

To be assured consideration, comments must be received at one of the addresses provided below, no later than 5 p.m. on June 27, 2016.

ADDRESSES:

In commenting, please refer to file code CMS-5517-P. Because of staff and resource limitations, we cannot accept comments by facsimile (FAX) transmission. You may submit comments in one of four ways (please choose only one of the ways listed):

1. Electronically. You may submit electronic comments on this regulation to http://www.regulations.gov . Follow the “Submit a comment” instructions.

2. By regular mail. You may mail written comments to the following address ONLY: Centers for Medicare & Medicaid Services, Department of Health and Human Services, Attention: CMS-5517-P, P.O. Box 8013, Baltimore, MD 21244-8013.

Please allow sufficient time for mailed comments to be received before the close of the comment period.

3. By express or overnight mail. You may send written comments to the following address ONLY: Centers for Medicare & Medicaid Services, Department of Health and Human Services, Attention: CMS-5517-P, Mail Stop C4-26-05, 7500 Security Boulevard, Baltimore, MD 21244-1850.

4. By hand or courier. Alternatively, you may deliver (by hand or courier) your written comments ONLY to the following addresses prior to the close of the comment period:

a. For delivery in Washington, DC—Centers for Medicare & Medicaid Services, Department of Health and Human Services, Room 445-G, Hubert H. Humphrey Building, 200 Independence Avenue SW., Washington, DC 20201.

(Because access to the interior of the Hubert H. Humphrey Building is not readily available to persons without Federal government identification, commenters are encouraged to leave their comments in the CMS drop slots located in the main lobby of the building. A stamp-in clock is available for persons wishing to retain a proof of filing by stamping in and retaining an extra copy of the comments being filed.)

b. For delivery in Baltimore, MD—Centers for Medicare & Medicaid Services, Department of Health and Human Services, 7500 Security Boulevard, Baltimore, MD 21244-1850.

If you intend to deliver your comments to the Baltimore address, call telephone number (410) 786 7195 in advance to schedule your arrival with one of our staff members. Comments erroneously mailed to the addresses indicated as appropriate for hand or courier delivery may be delayed and received after the comment period.

For information on viewing public comments, see the beginning of the SUPPLEMENTARY INFORMATION section.

FOR FURTHER INFORMATION CONTACT:

Molly MacHarris, (410) 786-4461, for inquiries related to MIPS.

James P. Sharp, (410) 786-7388, for inquiries related to APMs.

SUPPLEMENTARY INFORMATION:

Inspection of Public Comments: All comments received before the close of the comment period are available for viewing by the public, including any personally identifiable or confidential business information that is included in a comment. We post all comments received before the close of the comment period on the following Web site as soon as possible after they have been received: http://www.regulations.gov . Follow the search instructions on that Web site to view public comments.

Comments received timely will also be available for public inspection as they are received, generally beginning approximately 3 weeks after publication of a document, at the headquarters of the Centers for Medicare & Medicaid Services, 7500 Security Boulevard, Baltimore, Maryland 21244, Monday through Friday of each week from 8:30 a.m. to 4 p.m. To schedule an appointment to view public comments, phone 1-800-743-3951.

Acronyms

Because of the many terms to which we refer by acronym in this proposed rule, we are listing the acronyms used and their corresponding meanings in alphabetical order below:

ABC^TM Achievable Benchmark of Care

ACA The Patient Protection and Affordable Care Act

ACO Accountable Care Organization

APM Alternative Payment Model

BPCI Bundled Payments for Care Improvement

CAH Critical Access Hospital

CAHPS Consumer Assessment of Healthcare Providers and Systems

CEHRT Certified EHR technology

CFR Code of Federal Regulations

CHIP Children's Health Insurance Program

CJR Comprehensive Care for Joint Replacement

CMMI Center for Medicare & Medicaid Innovation (Innovation Center)

CPIA Clinical Practice Improvement Activity

CPR Customary, Prevailing, and Reasonable

CPS Composite Performance Score

CPT Current Procedural Terminology

CQM Clinical Quality Measure

EHR Electronic heath record

EP Eligible professional

FFS Fee-for-Service

FQHC Federally Qualified Health Center

HIE Health Information Exchange

HIPAA Health Insurance Portability and Accountability Act of 1996

HITECH Health Information Technology for Economic and Clinical Health

HPSA Health Professional Shortage Area

HHS Department of Health & Human Services

HRSA Health Resources and Services Administration

IT Information technology

MACRA Medicare Access and CHIP Reauthorization Act of 2015

MEI Medicare Economic Index

MIPAA Medicare Improvements for Patients and Providers Act of 2008

MIPS Merit-Based Incentive Payment System

MLR Minimum Loss Rate

MSPB Medicare Spending per Beneficiary

MSR Minimum Savings Rate

MUA Medically Underserved Area

NPI National Provider Identifier

OCM Oncology Care Model

ONC Office of the National Coordinator for Health Information Technology

PECOS Medicare Provider Enrollment, Chain, and Ownership System

PFPMs Physician Focused Payment Models

PFS Physician Fee Schedule

PHS Public Health Service

PQRS Physician Quality Reporting System

QCDRs Qualified Clinical Data Registries

QP Qualifying APM Professional

QRDA Quality Reporting Document Architecture

QRUR Quality and Resource Use Reports

RBRVS Resource-Based Relative Value Scale

RHC Rural Health Clinic

RVU Relative Value Unit

SGR Sustainable Growth Rate

TCPI Transforming Clinical Practice Initiative

TIN Tax Identification Number

VM Value-based Payment Modifier

VPS Volume Performance Standard

Table of Contents

Executive Summary

I. Background

A. Physician and Practitioner Payment Under Medicare

B. Current Reporting Programs and Regulations (Overview)

C. Overview of Section 101 of the MACRA

D. Stakeholder Input

II. Provisions of the Proposed Regulations

A. Establishing MIPS and the APMs Incentive

B. Program Principles and Goals

C. Changes to Existing Programs

D. Definitions

E. MIPS Program Details

F. Incentive Payments for Participating in Advanced APMs

III. Collection of Information Requirements

IV. Response to Comments

V. Regulatory Impact Analysis

A. Statement of Need

B. Overall Impact

C. Changes in Medicare Payments

D. Impact on Beneficiaries

E. Impact on Other Health Care Programs and Providers

F. Alternatives Considered

G. Assumptions and Limitations

H. Accounting Statement

Executive Summary

1. Purpose

The Medicare Access and CHIP Reauthorization Act of 2015 (MACRA) (Pub. L. 114-10, enacted April 16, 2015), amended title XVIII of the Social Security Act (the Act) to repeal the Medicare sustainable growth rate and strengthen Medicare access by improving physician payments and making other improvements, to reauthorize the Children's Health Insurance Program (CHIP), and for other purposes. This rule is needed to propose policies to improve physician payments by changing the way Medicare incorporates quality measurement into payments and by developing new policies to address and incentivize participation in alternative payment models.

This proposed rule would establish the Merit-Based Incentive Payment System (MIPS), a new program for certain Medicare-participating practitioners. MIPS would consolidate components of three existing programs, the Physician Quality Reporting System (PQRS), the Physician Value-based Payment Modifier (VM), and the Medicare Electronic Health Record (EHR) Incentive Program for eligible professionals (EPs), and would continue the focus on quality, resource use, and use of certified EHR technology in a cohesive program that avoids redundancies. This proposed rule also would establish incentives for participation in certain alternative payment models (APMs), supporting the Administration's goals of moving more fee-for-service payments into APMs that focus on better care, smarter spending, and healthier people. This proposed rule also includes proposed criteria for use by the Physician-Focused Payment Model Technical Advisory Committee (PTAC) in making comments and recommendations to the Secretary on physician-focused payment models (PFPMs).

In this proposed rule we have rebranded key terminology based on feedback from stakeholders, with the goal of selecting terms that would be more easily identified and understood by our stakeholders. We discuss these terminology changes in greater detail in the following sections of this proposed rule.

2. Summary of the Major Provisions

This proposed rule would sunset payment adjustments under the current PQRS, VM, and the Medicare EHR Incentive Program for EPs. Components of these three programs would be carried forward into the new MIPS program.

This proposed rule would establish a new subpart O of our regulations at 42 CFR 414.1300 to implement the new MIPS program as required by the MACRA.

(a) MIPS

In establishing MIPS, this rule would define MIPS program participants as “MIPS eligible clinicians” rather than “MIPS EPs” as that term is defined at section 1848(q)(1)(C) and used throughout section 1848(q) of the Act. MIPS eligible clinicians will include physicians, physician assistants, nurse practitioners, clinical nurse specialists, certified registered nurse anesthetists, and groups that include such clinicians. The rule proposes definitions and requirements for groups. In addition to proposing definitions for MIPS eligible clinicians, the rule also proposes rules for the specific Medicare-enrolled practitioners that would be excluded from MIPS, including newly Medicare-enrolled eligible clinicians, Qualifying APM Participants (QPs), certain Partial Qualifying APM Participants (Partial QPs), and clinicians that fall under the proposed low-volume threshold.

This rule proposes MIPS performance standards and a MIPS performance period of 1 calendar year (January 1 through December 31) for all measures and activities applicable to the four performance categories. Further, we propose to use 2017 as the performance period for the 2019 payment adjustment. Therefore, the first performance period would start in 2017 for payments adjusted in 2019. This time frame is needed to allow data and claims to be submitted and data analysis to occur. In addition, it would allow for a full year of measurement and sufficient time to base adjustments on complete and accurate information.

As directed by the MACRA, this rule proposes measures, activities, reporting, and data submission standards across four performance categories: Quality, resource use, clinical practice improvement activities (CPIAs), and meaningful use of certified EHR technology (referred to in this proposed rule as “advancing care information”). Measures and activities would vary by category and include outcome measures, performance measures, and global and population-based measures. Consideration would be given to the application of measures to non-patient facing MIPS eligible clinicians.

Quality measures would be selected annually through a call for quality measures process. Selection of these measures is proposed to be based on certain criteria that align with CMS priorities, and a final list of quality measures will be published in the Federal Register by November 1 of each year. Under the standards proposed in this rule, there would be options for reporting as an individual MIPS eligible clinician or as part of a group. Some data could be submitted via relevant third party data submission entities, such as qualified clinical data registries (QCDRs), health IT vendors, qualified registries, and CMS-approved survey vendors.

Within each performance category, we propose some specific standards, including:

• Quality: For most MIPS eligible clinicians, we propose to include a minimum of six measures with at least one cross-cutting measure (for patient-facing MIPS eligible clinicians) and an outcome measure if available; if an outcome measure is not available, then the eligible clinician would report one other high priority measure (appropriate use, patient safety, efficiency, patient experience, and care coordination measures) in lieu of an outcome measure. MIPS eligible clinicians can meet this criterion by selecting measures either individually or from a specialty-specific measure set.
• Resource Use: Continuation of two measures from the VM: Total per costs capita for all attributed beneficiaries and Medicare Spending per Beneficiaries (MSPB) with minor technical adjustments. In addition, episode-based measures, as applicable to the MIPS eligible clinician.
• CPIA: We generally encourage but are not requiring a minimum number of CPIAs.
• Advancing Care Information: Assessment based on advancing care information measures and objectives.

We propose standards for measures, scoring, and reporting for MIPS eligible clinicians across all four performance categories outlined in this section. We propose that MIPS eligible clinicians who participate in certain types of APMs will be scored using an APM scoring standard instead of the generally applicable MIPS scoring standard.

The U.S. Department of Health & Human Services' (HHS) Office of the Assistant Secretary for Planning and Evaluation (ASPE) is conducting studies and making recommendations on the issue of risk adjustment for socioeconomic status on quality measures and resource use as required by section 2(d) of the Improving Medicare Post-Acute Care Transformation Act of 2014 (the IMPACT Act) and expects to issue a report to Congress by October 2016. We will closely examine the recommendations issued by ASPE and incorporate them, as feasible and appropriate, in future rulemaking.

We are proposing MIPS eligible clinicians have the flexibility to submit information individually or via a group or an APM Entity group; however, the MIPS eligible clinician would use the same identifier for all performance categories. The proposed scoring methodology has a unified approach across all performance categories, would allow MIPS eligible clinicians to know in advance what they need to do to perform well in MIPS, and eliminates the need for an “all or nothing” scoring as has been the case under some other CMS programs. The four performance category scores (quality, resource use, CPIA, and advancing care information) would be aggregated into a MIPS composite performance score (CPS). The MIPS CPS would be compared against a MIPS performance threshold. The CPS would be used to determine whether a MIPS eligible clinician receives an upward payment adjustment, no payment adjustment, or a downward payment adjustment as appropriate. Payment adjustments would be scaled for budget neutrality, as required by statute. The CPS would also be used to determine whether a MIPS eligible clinician qualifies for an additional positive adjustment factor for exceptional performance.

To ensure that MIPS results are useful and accurate, we propose a process for providing performance feedback to MIPS eligible clinicians. Beginning July 1, 2017, we propose to include information on the quality and resource use performance categories in the performance feedback. Initially, we propose to provide performance feedback on an annual basis. In future years, we may consider providing performance feedback on a more frequent basis as well as adding feedback on the performance categories of CPIA and advancing care information. We propose to make performance feedback available using a CMS designated system. Further, we propose to leverage additional mechanisms such as health IT vendors, registries, and QCDRs to help disseminate data/information contained in the performance feedback to eligible clinicians where applicable.

We propose to adopt a targeted review process under MIPS wherein a MIPS eligible clinician may request that we review the calculation of the MIPS adjustment factor and, as applicable, the calculation of the additional MIPS adjustment factor applicable to such MIPS eligible clinician for a year. We further propose a general process by which a MIPS eligible clinician could request targeted review.

We propose requirements for third-party data submission to MIPS. Specifically, qualified registries, QCDRs, health IT vendors, and CMS-approved survey vendors would have the ability to act as intermediaries on behalf of MIPS eligible clinicians and groups for submission of data to us across the quality, CPIA, and advancing care information performance categories.

We also propose a process for public reporting of MIPS information through the Physician Compare Web site. We propose public reporting of a MIPS eligible clinician's data; in that for each program year, we will post on a public Web site (for example, Physician Compare), in an easily understandable format, information regarding the performance of MIPS eligible clinicians or groups under the MIPS.

(b) APMs

In this rule, we propose standards we would use for the purposes of the Alternative Payment Model (APM) incentive. The MACRA defines APM for the purposes of the incentive as a model under section 1115A of the Social Security Act (the Act) (excluding a health care innovation award), the Shared Savings Program under section 1899 of the Act, a demonstration under section 1866C of the Act, or a demonstration required by federal law. We propose to define the term “Other Payer APMs” to refer to arrangements in which eligible clinicians may participate through other payers. We also propose to define the term APM Entity as an entity that participates in an APM through a contract with a payer.

APMs that meet the criteria to be Advanced APMs provide the pathway through which eligible clinicians can become QPs and earn incentive payments for participation in APMs as specified under the MACRA. This rule proposes two types of Advanced APMs: Advanced APMs and Other Payer Advanced APMs. To be an Advanced APM, an APM must meet three requirements: (1) Require participants to use certified EHR technology; (2) provide payment for covered professional services based on quality measures comparable to those used in the quality performance category of MIPS; and (3) be either a Medical Home Model expanded under section 1115A of the Act or bear more than a nominal amount of risk for monetary loses. In this rule, we propose criteria for each of the requirements to be an Advanced APM.

To be an Other Payer Advanced APM, a commercial or Medicaid APM must meet three requirements similar to the CMS Advanced APM requirements: (1) Require participants to use certified EHR technology; (2) provide payment based on quality measures comparable to those used in the quality performance category of MIPS; and (3) be either a Medicaid Medical Home Model that is comparable to Medical Home Models expanded under section 1115A of the Act or bear more than a nominal amount of risk for monetary loses.

We propose that we would notify the public of which APMs will be Advanced APMs prior to each QP Performance Period, starting no later than January 1, 2017. This information will be posted on our Web site.

We propose that professional services furnished at Critical Access Hospitals (CAHs), Rural Health Clinics (RHCs) and Federally Qualified Health Centers (FQHCs) that meet certain criteria be counted towards the QP determination.

The MACRA sets a Medicare threshold for the level of participation in Advanced APMs required for an eligible clinician to become a QP for a year. The Medicare Option, based on Part B payments for covered professional services or counts of patients furnished covered professional services under Part B, is applicable beginning with CY 2019. The All-Payer Combination Option, based on the Medicare Option, as well as an eligible clinician's participation in Other Payer Advanced APMs, is applicable beginning with CY 2021. For eligible clinicians to become QPs through the All-Payer Combination Option, an Advanced APM Entity or eligible clinician must submit information to us so that we can determine whether an Other Payer APM is an Other Payer Advanced APM and whether an eligible clinician meets the requisite QP threshold of participation. We propose a methodology and criteria to evaluate eligible clinicians using the All-Payer Combination Option. For purposes of evaluating Other Payer APMs, we also propose criteria for the definition of Medicaid Medical Homes and Medical Home Model.

We propose to identify individual eligible clinicians by a unique APM participant identifier using the individuals' TIN/NPI combinations, and to assess as an APM Entity group all individual eligible clinicians listed as participating in an Advanced APM Entity to determine QP status for a year. We also propose that if an individual eligible clinician who participates in multiple Advanced APM Entities does not achieve QP status through participation in any single APM Entity, we would assess the eligible clinician individually to determine QP status based on combined participation in Advanced APMs.

We propose the method that CMS would use to calculate and disburse the APM Incentive Payments to QPs. We propose specific rules for calculating the APM Incentive Payment when a QP also receives non-fee-for-service payments or payment adjustments through the Medicare EHR Incentive Program, PQRS, VM, MIPS, or other payment adjustment programs.

We propose a process for eligible clinicians to choose whether or not to be subject to the MIPS payment adjustment in the event that they are determined to be Partial QPs.

We propose that we would perform monitoring and compliance around APM Incentive Payments.

We propose a definition for Physician-Focused Payment Models (PFPMs), criteria that would be used by the PFPM Technical Advisory Committee (PTAC), the Secretary, and CMS to evaluate proposals for PFPMs, and the process by which PFPMs would be considered for testing and implementation by CMS after review by the PTAC.

We propose to require MIPS eligible clinicians, as well as EPs, eligible hospitals, and Critical Access Hospitals (CAHs) under the existing EHR Incentive Programs to make a demonstration related to the provisions concerning blocking the sharing of information under section 106(b)(2) of the MACRA and, separately, to demonstrate cooperation with authorized ONC surveillance of certified EHR technology.

3. Summary of Costs & Benefits

Under the MACRA's requirements, MIPS would distribute payment adjustments to between approximately 687,000 and 746,000 eligible clinicians in 2019. Payment adjustments would be based on MIPS eligible clinicians' performance on specified measures and activities within the four performance categories. We estimate that MIPS payment adjustments would be approximately equally distributed between negative adjustments ($833 million) and positive adjustments ($833 million) to MIPS eligible clinicians, to ensure budget neutrality. Additionally, MIPS would distribute approximately $500 million in exceptional performance payments to MIPS eligible clinicians whose performance exceeds a specified threshold. These payment adjustments are expected to drive quality improvement in the provision of MIPS eligible clinicians' care to Medicare beneficiaries and to all patients in the health care system. However, the distribution could change based on the final population of MIPS eligible clinicians for CY 2019 and the distribution of scores under the program.

We estimate that between approximately 30,658 and 90,000 eligible clinicians would become QPs through participation in Advanced APMs, and are estimated to receive between $146 million and $429 million in APM Incentive Payments for CY 2019. As with MIPS, we expect that APM participation would drive quality improvement for clinical care provided to Medicare beneficiaries and to all patients in the health care system.

I. Background

In January 2015, the Administration announced new goals for transforming Medicare by moving away from traditional fee-for-service payments in Medicare towards a payment system focused on linking physician reimbursements to quality care through APMs ( http://www.hhs.gov/about/news/2015/01/26/better-smarter-healthier-in-historic-announcement-hhs-sets-clear-goals-and-timeline-for-shifting-medicare-reimbursements-from-volume-to-value.html#) and other value-based purchasing arrangements. This is part of an overarching Administration strategy to transform how health care is delivered in America, changing payment structures to improve quality and patient outcomes.

The Medicare Access and CHIP Reauthorization Act of 2015 (MACRA) of 2015 (Pub. L. 114-10, enacted April 16, 2015, and hereafter referred to as the MACRA), landmark bipartisan legislation, advances a forward-looking, coordinated framework for health care providers to successfully take part in the CMS Quality Payment Program that rewards value and outcomes in one of two ways:

• Merit-Based Incentive Payment System (MIPS).
• Advanced Alternative Payment Models (Advanced APMs).

The MACRA marks a milestone in efforts to improve and reform the health care system. Building off of the successful coverage expansions and improvements to access under the Affordable Care Act, the MACRA puts an increased focus on the quality and value of care delivered. By incentivizing participation in certain APMs, such as Accountable Care Organizations (ACOs), Medical Home Models, and episode payment models, and by incentivizing quality and value for eligible clinicians under the MIPS, we support the nation's progress toward achieving a patient-centered health care system that delivers better care, smarter spending, and healthier people and communities.

The Department is focused on three core strategies to drive continued progress and improvement, and MACRA provides new tools to that end, which build upon existing efforts, such as the CMS Quality Strategy . First, we are focused on improving the way clinicians are paid to incentivize quality and value of care over simply quantity of services. The Quality Payment Program replaces the SGR update formula with Medicare PFS updates ultimately linked to participation in Advanced APMs and also creates a new, sustainable mechanism for calculating payment adjustments for clinicians' services that links payments to quality and value: The Merit-based Incentive Payment System (MIPS), with the ultimate goal of paying for value and better care. By rewarding eligible clinicians based on their performance, MIPS consolidates key components of the PQRS, the VM and the Medicare EHR Incentive Program for EPs into one single, streamlined program based on performance in the following:

• Quality.
• Resource use.
• CPIA.
• Advancing care information.

Second, we are focused on improving the way care is delivered by providing clinical practice support, data and feedback reports to guide improvement and better decision-making. Allowing for stronger, real-time, easy-to-understand feedback and actionable data on eligible clinician performance on clinical quality measures (CQMs), utilization of resources and cost can lead to stronger care coordination, help facilitate and enhance team-based approaches, and support greater integration within practices, improved patient communication, a stronger focus on population health, and continuous learning and rapid-cycle improvement.

Third, we are focused on making data more available and enabling the use of certified EHR technology to support care delivery. Consistent use of certified EHR technology and clinical quality measurement in managing patient populations would help lead to substantial improvements in our health care system, by allowing clinicians to track and take care of their patients throughout the care continuum and to easily and securely access electronic health information to support care when and where it is needed.

By driving significant changes in how care is delivered and changes in the health care system to make it more responsive to patients and families, we believe the Quality Payment Programs would encourage eligible clinicians to be accountable for the health of their patient population and support interested eligible clinicians in their successful transition into APMs. To implement this vision, we propose a program that allows for stronger alignment across requirements while minimizing burden on eligible clinicians. Further, we propose a program that is meaningful, understandable and flexible with a critical focus on transparency, effective communication with stakeholders and operational feasibility. To aid in this process, we have sought feedback from the health care community through various public avenues and will seek comment through this proposed rule. As we establish policies for effective implementation of the MACRA, we are also focused on improving the health system by ensuring that our policies can scale in future years. As we drive change through this proposed rule, we will begin by laying the groundwork for expansion towards an innovative, outcome-focused, patient-centered, resource-effective health system. Through a staged approach we can develop our policies are operationally feasible and made in consideration of system capabilities and of our core strategies to drive progress and reform efforts.

A. Physician and Practitioner Payment Under Medicare

1. History

Medicare payment systems have undergone significant changes since the Act established the Medicare program in 1965. Originally, Medicare was modeled on the existing health insurance marketplace (See 1965 Medicare Amendment to SSA, Pub. L. 89-97). Medicare payments to physicians and hospitals were based on the amounts that had been historically charged by physicians and hospitals for various health care services. Medicare initially paid for physicians' services using a “customary, prevailing, and reasonable” charge (CPR) payment system. (1965 Medicare Amendment to SSA, Pub. L. 89-97). Congress later changed the CPR system in part to counter increased charges to physicians, leading to rapid increases in program payments.

In 1984, Medicare changed the way it paid hospitals to a prospective payment system (Social Security Amendments of 1983, Pub. L. 98-21) that moved away from a charge-based per diem rate and introduced the Medicare Economic Index (MEI) to modify physician payment. The MEI was used to measure the annual increase in practice costs for updating payment for physicians' services.

Beginning in 1992 following the passage of the Omnibus Budget Reconciliation Act of 1989 (OBRA 89) (Pub. L. 101-239, enacted on December 19, 1989), the historical charge-based fee schedule was replaced with a fee schedule that used a Resource-Based Relative Value Scale, developed at Harvard University, which attempted to assess for each service the relative value of a physician's work effort, as well as the practice expenses and malpractice liability expenses involved.

Under OBRA 89, the resource-based Medicare PFS aimed to establish a rational basis for valuing payments for physicians' services. Therefore, under the current resource-based approach, payment for a service depends on the value of the resources involved in performing a particular service.

Following the implementation of the resource-based PFS over several years, the fee schedule has specified Medicare payments for physicians' services. Each medical, surgical and diagnostic service, described by a current procedural terminology (CPT) code is assigned relative value units (RVUs) for three resource categories: Work, practice expense, and malpractice expense. These three RVU values are summed, geographically adjusted, and multiplied by a fixed-dollar conversion factor for the payment year to determine the payment amount for each service or procedure. Over time, we have reviewed and revised the RVU values using our own methodologies and other information.

After the adoption of the resource-based PFS, further amendments to the Act have led to the imposition of spending targets for physicians' services. Initially, the spending limit was set by a Volume Performance Standard (VPS) that tied the annual update to a target that was based on historical trends in physician costs. Because of the way the adjustment was calculated, it produced very unstable updates, with swings that were much greater than the changes in the underlying MEI.

The Balanced Budget Act of 1997 (BBA) (Pub. L. 105-33, enacted on August 5, 1997) replaced the VPS with the SGR formula to update the PFS each year. Under BBA, the SGR made several changes including a much more aggressive measure to control spending, tying the allowable increases in physician spending to the growth rate in real GDP per capita. In general, under the SGR formula, if cumulative expenditures from the current period going back to 1996 (the base year) were less than the cumulative spending target over that same period, the annual update was increased according to a statutory formula. However, if spending exceeded the cumulative spending target over the same period, the SGR methodology requires reductions in the fee schedule update to bring spending back in line with the targeted growth rate.

In the initial years of implementation, actual expenditures did not exceed allowed targets. But beginning in 2002, cumulative actual expenditures began to exceed allowed targets for the year, resulting in SGR-mandated reductions in the fee schedule update adjustment factor. The Congress enacted a series of laws to override these reductions. The SGR-based update adjustment factor had not been allowed to take effect since 2003 due to consistent intervention by the Congress to avert payment reductions.

Currently, payments under the Medicare PFS include several payment adjustments that increase or decrease payments to practitioners based on performance. The Tax Relief and Health Care Act of 2006 required the establishment of the PQRS that would include an incentive payment to EPs who satisfactorily report data on quality measures. The Medicare Improvements for Patients and Provider Act of 2008 (MIPPA) (Pub. L. 110-275, enacted on July 15, 2008) made the PQRS program permanent. The HITECH Act of 2009, part of the American Recovery and Reinvestment Act (ARRA), established incentive payments to EPs to promote the adoption and meaningful use of certified EHR technology. HITECH provided the statutory basis for the Medicare incentive payments made to meaningful EHR users and also established downward payment adjustments, under Medicare, beginning with calendar year 2015, for EPs that are not meaningful users of certified EHR technology for certain associated reporting periods.

The Affordable Care Act (Pub. L. 111-148) required the establishment of a value-based payment modifier that provides for differential payment to a physician or group of physicians under the Medicare PFS based upon the quality of care furnished compared to cost, that is implemented in a budget-neutral manner. Beginning in 2015, the VM applies to payments for items and services furnished by physicians in groups of 100 or more, and will apply to all physicians and certain types of non-physician practitioners in later years. The VM is being phased in and will apply to all physicians in groups and individual physicians in 2017.

2. Payment Models and Innovation

The policies proposed in this rule are intended to continue to move Medicare away from a primarily volume based fee-for-service (FFS) payment system for physicians and other professionals. As described in this section of the proposed rule, for many years Medicare was primarily a FFS payment system that paid health care providers based on the volume of services they delivered, rather than the value of those services. This contributed to increased costs without incentivizing improvement in the quality of care. Over time, the Congress and CMS have taken progressive steps to move toward paying for value, as demonstrated by Medicare's long history of testing alternative payment methods.

Medicare has been testing alternative payment methods since waiver authority for Medicare demonstrations was granted through section 402 of the Social Security Amendments of 1967. Demonstrations and pilot programs, (also called “research studies”) are special projects that test improvements in Medicare coverage, payment, and quality of care ( https://www.medicare.gov/sign-up-change-plans/medicare-health-plans/other-health-plans/other-medicare-health-plans.html ). Demonstrations have examined whether alternative payment methods increase the efficiency of Medicare and Medicaid and whether payment for services not otherwise covered increases the effectiveness of care. Medicare's demonstration authority has allowed it to test the effect of policy changes on Medicare on a small scale in order to inform broader policy.

The Affordable Care Act includes a number of provisions, for example, the Medicare Shared Savings Program, designed to improve the quality of Medicare services, support innovation and the establishment of new payment models, better align Medicare payments with health care provider costs, strengthen Medicare program integrity, and put Medicare on a firmer financial footing.

The Affordable Care Act created the Center for Medicare and Medicaid Innovation (Innovation Center). The Innovation Center was established by section 1115A of the Act (as added by section 3021 of the Affordable Care Act). The Innovation Center's mandate gives it flexibility within the parameters of section 1115A of the Act to select and test promising innovative payment and service delivery models. Congress created the Innovation Center for the purpose of testing innovative payment and service delivery models to reduce program expenditures while preserving or enhancing the quality of care provided to those individuals who receive Medicare, Medicaid, or CHIP benefits. See https://innovation.cms.gov/about/index.html . Models that have met those expectations may be expanded in scope through rulemaking up to a national scale.

To better coordinate these models and demonstration projects and to avoid duplicative efforts and expenses, the former Office of Research, Development and Information, which oversaw statutory demonstrations and those under section 402 etc., was merged with the Innovation Center in early 2011. As a result, the Innovation Center oversees not only initiatives that are authorized under section 1115A of the Act, but also activities under several other authorities, including other provisions of the Affordable Care Act, and other laws and projects authorized by section 402 of the Social Security Amendments of 1967, as amended.

The Innovation Center's portfolio of models has attracted participation from a broad array of health care providers, states, payers, and other stakeholders, and serves Medicare, Medicaid, and CHIP beneficiaries in all 50 states, the District of Columbia, and Puerto Rico. We estimate that over 4.7 million Medicare, Medicaid, and CHIP beneficiaries are or soon will be receiving care furnished by the more than 61,000 eligible clinicians participating in APMs tested by the CMS Innovation Center.

Beyond the care improvements for these beneficiaries, Innovation Center models are affecting millions of additional Americans by engaging thousands of other health care providers, payers, and states in model tests and through quality improvement efforts across the country. Many payers other than CMS have implemented alternative payment arrangements or models, or have collaborated in Innovation Center models. The participation of multiple payers in alternative delivery and payment models increases momentum for delivery system transformation and encourages efficiency for health care organizations.

The Innovation Center works directly with other CMS components and colleagues throughout the federal government in developing and testing new payment and service delivery models. Other federal agencies with which the Innovation Center has collaborated include the Centers for Disease Control and Prevention (CDC), Health Resources and Services Administration (HRSA), Agency for Healthcare Research and Quality (AHRQ), Office of the National Coordinator for Health Information Technology (ONC), Administration for Community Living (ACL), Department of Housing and Urban Development (HUD), Administration for Children and Families (ACF), and the Substance Abuse and Mental Health Services Administration (SAMHSA). These collaborations help the Innovation Center effectively test new models and execute mandated demonstrations.

B. Current Reporting Programs and Regulations (Overview)

The MACRA's passage has led to several changes with the existing Medicare PFS, various Medicare payment programs that tie payment to value, and the testing of alternative payment models. Specifically, the MACRA's enactment consolidated aspects of certain quality reporting and performance programs into the new MIPS, including the meaningful use of certified EHR technology (section 1848(o) of the Act), the PQRS (section 1848(k) and (m) of the Act, and the VM (section 1848(p) of the Act). The following section provides an overview of existing programs and the extent of their programs before and after the MACRA.

Currently, the Medicare EHR Incentive Program has been divided into three progressive stages of meaningful use with certain specified requirements that EPs must meet in order to qualify for Medicare EHR incentive payments and avoid downward payment adjustments. Full achievement of these requirements designated an EP as a “meaningful EHR user” and made that EP eligible for incentive payments and not subject to downward payment adjustments. The MACRA's enactment altered the EHR Incentive Programs such that the existing Medicare payment adjustment for an EP under 1848(a)(7)(A) of the Act ends after CY 2018. Using certified EHR technology is included in MIPS as part of the advancing care information component of the overall performance score. Generally, the MACRA did not change hospital participation in the Medicare EHR Incentive Program or participation for EPs in the Medicaid EHR Incentive Program.

PQRS, as set forth in sections 1848(a), (k), and (m) of the Act, is a quality reporting program that provides for incentive payments (which ended in 2014) and payment adjustments (which began in 2015) to EPs and group practices based on whether they satisfactorily report data on quality measures for covered professional services furnished during a specified reporting period or to EPs and group practices based on whether they satisfactorily participate in a qualified clinical data registry (QCDR). The MACRA ends the PQRS adjustment after CY 2018 and provides for the inclusion of various aspects of PQRS in MIPS as part of the quality component of the overall performance score.

Section 1848(p) of the Act, as amended by the Affordable Care Act, required that we establish a VM that provides for differential payment under the Medicare PFS based upon the quality of care furnished compared to cost and apply it to specific physicians and groups of physicians as determined appropriate by the Secretary starting in 2015 and to all physicians by 2017. In the CY 2013 PFS final rule with comment period (77 FR 69307), we discussed the goals of the VM and also established the specific principles that should govern the implementation of the VM. The MACRA sunsets the VM, ending it after CY 2018 and establishing certain aspects of the VM as part of the resource use component of MIPS in CY 2019.

C. Overview of Section 101 of the MACRA

Section 101 of the MACRA amended sections 1848(d) and (f) of the Act to repeal the SGR formula for updating Medicare PFS payment rates and substituted a series of specified annual update percentages. Section 101 goes on to establish a new methodology that ties annual PFS payment adjustments to value for MIPS eligible clinicians. Section 101 also creates an incentive program to encourage participation by eligible clinicians in Advanced APMs.

Section 1848(q) of the Act, as added by section 101(c) of the MACRA, requires establishment of the MIPS, applicable beginning with payments for items and services furnished on or after January 1, 2019, under which the Secretary is required to: (1) Develop a methodology for assessing the total performance of each MIPS eligible clinician according to performance standards for a performance period for a year; (2) using the methodology, provide a CPS for each MIPS eligible clinician for each performance period; and (3) use the CPS of the MIPS eligible clinician for a performance period for a year to determine and apply a MIPS adjustment factor (and, as applicable, an additional MIPS adjustment factor) to the MIPS eligible clinician for the year. Under section 1848(q)(2)(A) of the Act, a MIPS eligible clinician's CPS is determined using four performance categories: (1) Quality; (2) resource use; (3) CPIA; and (4) advancing care information. Section 1848(q)(10) of the Act requires the Secretary to consult with stakeholders (through a request for information (RFI) or other appropriate means) in carrying out the MIPS, including for the identification of measures and activities for each of the four performance categories under the MIPS, the methodology to assess each MIPS eligible clinician's total performance to determine their MIPS CPS, the methodology to specify the MIPS adjustment factor for each MIPS eligible clinician for a year, and the use of QCDRs for purposes of the MIPS.

Section 1848(q)(11) of the Act, as added by section 101(c) of the MACRA, provides for technical assistance to MIPS eligible clinicians in small practices, rural areas, and practices located in geographic health professional shortage areas (HPSAs). In general, the section requires the Secretary to enter into contracts or agreements with appropriate entities (such as quality improvement organizations, regional extension centers (as described in section 3012(c) of the Public Health Service (PHS) Act), or regional health collaboratives) (such as those identified in section 1115A of the Act) to offer guidance and assistance to MIPS eligible clinicians in practices of 15 or fewer eligible clinicians. Priority is to be given to such practices located in rural areas which we propose to define at § 414.1305 to include clinicians in counties designated as Micropolitan or Non-Core Based Statistical Areas (CBSAs), using HRSA's 2014-2015 Area Health Resource File ( http://datawarehouse.hrsa.gov/data/datadownload/ahrfdownload.aspx ), HPSAs (as designated under section 332(a)(1)(A) of the PHS Act), medically underserved areas (MUAs), and practices with low composite scores, for the MIPS performance categories or in transitioning to the implementation of, and participation in, an APM. Details regarding the technical assistance program are outside the scope of this proposed rule, and will be addressed in separate guidance.

Section 101(e) of the MACRA encourages participation in APMs by eligible clinicians and other eligible clinicians, and promotes the development of PFPMs by creating the PTAC. Specifically, this section: (1) Creates a payment incentive that applies to eligible clinicians from 2019 through 2024 who are Qualifying APM Participants (QPs) during the respective performance years, and provides for a higher fee schedule update for eligible clinicians who are QPs for a year beginning in 2026; (2) requires the establishment of a process for stakeholders to propose PFPMs to an independent PTAC that will review, comment on, and provide recommendations to the Secretary on the proposed PFPMs; and (3) requires CMS to establish criteria for PFPMs for use by the PTAC in making comments and recommendations to the Secretary. Additionally, section 101(c)(1) of the MACRA exempts QPs from payment adjustments under MIPS.

D. Stakeholder Input

In developing this proposed rule, in accordance with the law, we have sought feedback from stakeholders throughout the process such as in the 2016 Medicare PFS Proposed Rule; the Request for Information Regarding Implementation of the Merit-Based Incentive Payment System, Promotion of Alternative Payment Models, and Incentive Payments for Participation in Eligible Alternative Payment Models (hereafter referred to as the MIPS and APMs RFI); listening sessions; conversations with a wide number of stakeholders; and conversations with tribes and tribal officials through CMS' Tribal Technical Advisory Group. In addition, we note that the National Indian Health Board has requested an opportunity for consultation with CMS, as well as that we coordinate its standards with the Indian Health Service. Through the MIPS and APMs RFI published in the Federal Register on October 1, 2015 (80 FR 59102, 59102-59113), the Secretary of Health and Human Services (the Secretary) solicited comments regarding implementation of certain aspects of the MIPS and broadly sought public comments on the topics in section 101 of the MACRA, including the incentive payments for participation in APMs and increasing transparency of PFPMs. We received a high number of public comments in response to the MIPS and APMs RFI from a broad range of sources including professional associations and societies, physician practices, hospitals, patient groups, and health IT vendors.

We appreciate the high level of interest expressed by commenters and acknowledge their valued input throughout this proposed rule, providing summaries of RFI comments in relevant sections of this rule. In general, commenters supported the passage of regulations implementing the MACRA and maintain optimism as we move from fee-for-service Medicare payment towards an enhanced focus on the quality and value of care. Public support for the MACRA focuses on the potential of a value-based program to provide enough flexibility to be applied meaningfully to physician practices and patient quality of care. Commenters cautioned us to avoid elements of prior reporting programs that have been perceived as too focused on the volume of measures reported rather than measure relevance and impact on treatment. Commenters also requested that we avoid implementing additional requirements on top of the fee-for-service system, which would increase the reporting and compliance burden for eligible clinicians. Commenters believe the underlying goal in establishing the MACRA should be to create a new program that combines a limited (yet meaningful) set of requirements with choices for health care providers on how to meet those requirements. Commenters requested that there be broad opportunities to participate in APMs and the development of new Advanced APMs, and that resources be made available to assist them in moving towards participation in APMs if they do not already participate. Commenters expressed eagerness to participate in Advanced APMs and to be a part of transforming care.

Once again, we thank stakeholders for their considered responses through various venues including comments to the MIPS and APMs RFI. We intend to continue open communication with stakeholders (including consultation with tribes and tribal officials) on an ongoing basis, and we look forward to comments on the policies proposed in this rule.

II. Provisions of the Proposed Regulations

A. Establishing MIPS and the APM Incentive

Section 1848(q) of the Act, as added by section 101(c) of the MACRA, requires establishment of the MIPS (see section I.C. of this proposed rule for additional background information). Section 101(e) of the MACRA promotes the development of, and participation in, APMs for eligible clinicians (see section I.C. of this proposed rule for additional background information). Further information will be provided in future rulemaking.

B. Program Principles and Goals

Through the MACRA amendments, we believe the Congress sets broad goals to be accomplished intended to improve care and health outcomes for every American. More specifically, our goal with the Quality Payment Program is to continue to support health care quality, efficiency, and patient safety. MIPS promotes better care, healthier people, and smarter spending by evaluating MIPS eligible clinicians using a CPS that incorporates MIPS eligible clinicians' performance on quality, resource use, clinical practice improvement activities, and advancing care information. Under the incentives for participation in Advanced APMs, our goals, described in greater detail in section II.F. of this proposed rule, are to expand the opportunities for participation in APMs, maximize participation in current and future Advanced APMs, create clear and attainable standards for incentives, promote the continued flexibility in the design of APMs, and support multi-payer initiatives across the health care market. The Quality Payment Program will encourage more MIPS eligible clinicians to participate in Advanced APMs, which link quality and value to payment. The APM Incentive Payment for eligible clinicians who qualify as QPs will only be available through Advanced APMs, but it is a powerful incentive to increase participation in those APMs. MIPS eligible clinicians participating in APMs (who do not qualify as QPs) will receive favorable scoring under certain MIPS categories.

Our strategic goals in developing the Quality Payment Program include: (1) Design a patient-centered approach to program development that leads to better, smarter, and healthier care; (2) develop a program that is meaningful, understandable, and flexible for participating clinicians; (3) design incentives that drive delivery system reform principles and participation in APMs; and (4) ensure close attention to CMS' excellence in implementation, effective communication with stakeholders and operational feasibility.

C. Changes to Existing Programs

1. Sunsetting of Current Payment Adjustment Programs

Section 101(b) of the MACRA calls for the sunsetting of payment adjustments under three existing programs for Medicare enrolled physicians and other practitioners:

• The PQRS that incentivizes EPs to report on quality measures;
• The VM that provides for budget neutral, differential payment adjustment for EPs in physician groups and solo practices based on quality of care compared to cost; and
• The Medicare EHR Incentive Program for EPs that entails meeting certain requirements for the use of certified EHR technology.

Accordingly, we propose to revise certain regulations associated with these programs. We are not proposing to delete these regulations entirely, as the final payment adjustments under these programs will not occur until the end of 2018. For PQRS, we propose to revise § 414.90(e) introductory text and § 414.90(e)(1)(ii) to continue payment adjustments through 2018.

Similarly, we are proposing to amend the regulation text at § 495.102(d) to remove references to the payment adjustment percentage for years after the 2018 payment adjustment year and add a terminal limit of the 2018 payment adjustment year.

We are not proposing changes to 42 CFR part 414 subpart N—Value-Based Payment Modifier Under the PFS (§ 414.1200-1285), at this time. These regulations are already limited to certain years.

We invite comments on these proposed regulatory changes.

2. Meaningful Use Prevention of Information Blocking and Surveillance Demonstrations for MIPS Eligible Clinicians, EPs, Eligible Hospitals, and CAHs

a. Cooperation With Surveillance and Direct Review of Certified EHR Technology

We are proposing to require EPs, eligible hospitals, and CAHs to attest (as part of their demonstration of meaningful use under the Medicare and Medicaid EHR Incentive Programs) that they have cooperated with the surveillance of certified EHR technology under the ONC Health IT Certification Program, as authorized by 45 CFR part 170, subpart E. Similarly, we are proposing to require such an attestation from all eligible clinicians under the advancing care information performance category of MIPS, including eligible clinicians who report on the advancing care information performance category as part of an APM Entity group under the APM Scoring Standard, as discussed in section II.E.5.h of this proposed rule.

On October 16, 2015, ONC published the 2015 Edition Health Information Technology (Health IT) Certification Criteria, 2015 Edition Base Electronic Health Record (EHR) Definition, and ONC Health IT Certification Program Modifications final rule (“2015 Edition final rule”). The final rule made changes to the ONC Health IT Certification Program that strengthen the testing, certification, and surveillance of health IT. In addition, the final rule clarified and expanded the responsibilities of ONC-Authorized Certification Bodies (ONC-ACBs) with respect to the surveillance of certified EHR technology and other health IT certified under the ONC Health IT Certification Program, including requirements for ONC-ACBs to conduct more frequent and more rigorous surveillance of certified technology and capabilities “in the field” (80 FR 62707). The purpose of in-the-field surveillance is to provide greater assurance that health IT meets certification requirements not only in a controlled testing environment but also when used by health care providers in actual production environments (80 FR 62707).

In addition to these changes, on March 2, 2016, ONC published the ONC Health IT Certification Program: Enhanced Oversight and Accountability proposed rule, which would expand ONC's role to strengthen oversight under the ONC Health IT Certification Program by providing a means for ONC to directly review and evaluate the performance of certified health IT in certain circumstances, such as in response to potential systemic or widespread issues, or in response to problems or issues that could pose a risk to public health or safety, compromise the security or privacy of patients' health information, or give rise to other exigencies (81 FR 11055).

These efforts to strengthen surveillance and other oversight of certified health IT, including through expanded in-the-field surveillance and ONC direct review of technology and capabilities, are critical to the success of HHS programs and initiatives that require the use of certified health IT to improve health care quality and the efficient delivery of care. With respect to the use of certified EHR technology under the Medicare and Medicaid EHR Incentive Programs and the MIPS Program, effective surveillance and oversight is fundamental to providing basic confidence that such technology consistently meets applicable standards, implementation specifications, and certification criteria adopted by the Secretary when it is used by eligible clinicians, EPs, eligible hospitals, and CAHs, as well as by other persons with whom eligible clinicians, EPs, eligible hospitals, and CAHs need to exchange electronic health information to comply with program requirements. The need to ensure that technology consistently meets applicable standards, implementation specifications, and certification criteria is important both at the time it is certified and on an ongoing basis when it is implemented and used in the field by eligible clinicians, EPs, eligible hospitals, and CAHs in order to meet objectives and measures under the Medicare and Medicaid EHR Incentive Program or MIPS. Efforts to strengthen surveillance and oversight of certified EHR technology in the field will become even more important as the types and capabilities of certified EHR technology continue to evolve and with the onset of Stage 3 of the Medicare and Medicaid EHR Incentive Programs and MIPS, which include heightened requirements for sharing electronic health information with other providers and with patients using a broad range of certified EHR technology and other health IT. Finally, we note that effective surveillance and oversight of certified EHR technology is necessary if eligible clinicians, EPs, eligible hospitals, and CAHs are to be able to rely on certifications issued under the ONC Health IT Certification Program as the basis for selecting appropriate technologies and capabilities that support the use of certified EHR technology while avoiding potential implementation and performance issues.

For all of these reasons, the effective surveillance and oversight of certified health IT, and certified EHR technology in particular, is necessary to enable eligible clinicians, EPs, eligible hospitals, and CAHs to demonstrate that they are using certified EHR technology in a meaningful manner as required by sections 1848(o)(2)(A)(i) and 1886(n)(3)(A)(i) of the Act. Yet as ONC observed in the 2015 Edition final rule, such surveillance and oversight will not be effective unless EPs, eligible hospitals, and CAHs are actively engaged and cooperate with the authorized surveillance and oversight of their technology, including by granting access to and assisting ONC and ONC-ACBs to observe the performance of production systems (80 FR 62716).

Accordingly, we are proposing that as part of demonstrating that it is using certified EHR technology in a meaningful manner, an eligible clinician, EP, eligible hospital, or CAH must demonstrate its cooperation with these authorized surveillance and oversight activities. We are proposing to revise the definition of a meaningful EHR user at § 495.4, as well as the attestation requirements at § 495.40(a)(2)(i)(H) and § 495.40(b)(2)(i)(H) to require EPs, eligible hospitals, and CAHs to attest their cooperation with certain authorized health IT surveillance and direct review activities, described in more detail in this section of the rule, as part of demonstrating meaningful use under the Medicare and Medicaid EHR Incentive Programs. Similarly, we are proposing to include an identical attestation requirement in the submission requirements for eligible clinicians under the advancing care information performance category proposed at § 414.1375.

We propose that eligible clinicians, EPs, eligible hospitals, and CAHs would be required to attest that they have cooperated in good faith with the surveillance and ONC direct review of their health IT certified under the ONC Health IT Certification Program, as authorized by 45 CFR part 170, subpart E, to the extent that such technology meets (or can be used to meet) the definition of CEHRT. Under the terms of the attestation, such cooperation would include responding in a timely manner and in good faith to requests for information (for example, telephone inquiries, written surveys) about the performance of the certified EHR technology capabilities in use by the provider in the field. The provider's cooperation would also include accommodating requests (from ONC-Authorized Certification Bodies or from ONC) for access to the provider's certified EHR technology (and data stored in such certified EHR technology) as deployed by the provider in its production environment, for the purpose of carrying out authorized surveillance or direct review, and to demonstrate capabilities and other aspects of the technology that are the focus of such efforts, to the extent that doing so would not compromise patient care or be unduly burdensome for the eligible clinician, EP, eligible hospital, or CAH.

We understand that cooperating with in-the-field surveillance may require prioritizing limited time and other resources. We note that ONC has established safeguards to minimize the burden of surveillance on eligible clinicians, EPs, eligible hospitals, and CAHs. In conducting randomized surveillance, ONC-ACBs must use consistent, objective, valid, and reliable methods to select the locations at which the surveillance will be performed (80 FR 62715). ONC-ACBs may also use appropriate sampling methodologies to minimize disruption to any individual provider or class of providers and to maximize the value and impact of surveillance activities for all providers and stakeholders (80 FR 62715). Moreover, if an ONC-ACB makes a good faith effort but is unable to complete in-the-field surveillance at a particular location, it may exclude the location and substitute a different location for surveillance (80 FR 62716).

In addition, we note that ONC has clarified, in consultation with the Office for Civil Rights, that ONC-ACBs engaging in authorized surveillance of certified EHR technology under the ONC Health IT Certification Program meet the definition of a “health oversight agency” in the HIPAA Privacy Rule (45 CFR 164.501), and as such a health care provider is permitted to disclose protected health information (PHI) (without patient authorization and without a business associate agreement) to an ONC-ACB during the limited time and as necessary for the ONC-ACB to perform the required on-site surveillance of the certified EHR technology (45 CFR 164.512(d)(1)(iii)) (80 FR 62716).

For the foregoing reasons, we believe this proposal will support the surveillance and oversight of certified health IT, as necessary to support meaningful use of CEHRT for all eligible clinicians under the MIPS program, as well as EPs, eligible hospitals and CAHs under the Medicare and Medicaid EHR Incentive Programs, while ensuring that such surveillance or review does not create unnecessary or unreasonable burdens for health care providers or patients. We request public comment on this proposal.

b. Support for Health Information Exchange and the Prevention of Information Blocking

To prevent actions that block the exchange of information, section 106(b)(2)(A) of the MACRA amended section 1848(o)(2)(A)(ii) of the Act to require that, to be a meaningful EHR user, an EP must demonstrate that he or she has not knowingly and willfully taken action (such as to disable functionality) to limit or restrict the compatibility or interoperability of certified EHR technology. Section 106(b)(2)(B) of MACRA made corresponding amendments to section 1886(n)(3)(A)(ii) of the Act for eligible hospitals and, by extension, under section 1814(l)(3) of the Act for CAHs. Sections 106(b)(2)(A) and (B) of the MACRA provide that the manner of this demonstration is to be through a process specified by the Secretary, such as the use of an attestation. Section 106(b)(2)(C) of the MACRA states that the demonstration requirements in these amendments shall apply to meaningful EHR users as of the date that is 1 year after the date of enactment, which would be April 16, 2016.

On December 16, 2014, in an explanatory statement accompanying the Consolidated and Further Continuing Appropriations Act, Congress urged ONC to take steps to decertify products that proactively block the sharing of information because those practices frustrate congressional intent, devalue taxpayer investments in certified EHR technology, and make certified EHR technology less valuable and more burdensome for eligible hospitals and eligible health care providers to use. Congress also asked for a detailed report on health information blocking, which ONC delivered on April 10, 2015. In the report, and based on the available evidence and its own experience, ONC found that some persons and entities—including some health care providers—are knowingly and unreasonably interfering with the exchange or use of electronic health information in ways that limit its availability and use to improve health and health care.

Following these activities, on April 16, 2015, the MACRA was enacted, including section 106(b)(2), which amended sections 1848(o)(2)(A)(ii) and 1886(n)(3)(A)(ii) of the Act, as discussed in this section of the rule. Prior to these amendments, to be treated as a meaningful EHR user, an EP, eligible hospital, or CAH had to demonstrate to the satisfaction of the Secretary that its certified EHR technology was connected during the relevant EHR reporting period in a manner that provided, in accordance with law and standards applicable to the exchange of information, for the electronic exchange of health information to improve the quality of health care, such as promoting care coordination. As amended, respectively, by sections 106(b)(2)(A) and (B) of the MACRA, sections 1848(o)(2)(A)(ii) and 1886(n)(3)(A)(ii) of the Act now require that, in addition to demonstrating such connectivity, an eligible clinician, EP, eligible hospital, or CAH must also demonstrate that it did not knowingly and willfully take action to limit or restrict the compatibility or interoperability of the certified EHR technology.

We believe that, at a minimum, such a demonstration would need to provide substantial assurance not only that the certified EHR technology was connected in accordance with applicable standards during the relevant EHR reporting period, but that the eligible clinician, EP, eligible hospital, or CAH acted in good faith to implement and use the certified EHR technology in a manner that supported and did not interfere with the electronic exchange of health information among health care providers and with patients to improve quality and promote care coordination. Accordingly, we are proposing that such a demonstration be made through an attestation comprising three statements related to health information exchange and information blocking, which are set forth in our proposal in this rule. We are proposing to revise the definition of a meaningful EHR user at § 495.4 and the attestation requirements at § 495.40(a)(2)(i)(I) and § 495.40(b)(2)(i)(I) to provide that, for attestations submitted on or after April 16, 2016, an EP, eligible hospital, or CAH under the Medicare and Medicaid EHR Incentive Programs must attest to this three-part attestation. For the same reasons stated in this section of the rule, we are also proposing to require such an attestation from all eligible clinicians under the advancing care information performance category of MIPS, including eligible clinicians who report on the advancing care information performance category as part of an APM Entity group under the APM Scoring Standard, as discussed in section II.E.5.h of this proposed rule. As noted in this section, the attestation we are proposing would consist of three statements related to health information exchange and information blocking. First, the eligible clinician, EP, eligible hospital, or CAH would be required to attest that it did not knowingly and willfully take action (such as to disable functionality) to limit or restrict the compatibility or interoperability of certified EHR technology.

Second, the eligible clinician, EP, eligible hospital, or CAH would be required to attest that it implemented technologies, standards, policies, practices, and agreements reasonably calculated to ensure, to the greatest extent practicable and permitted by law, that the certified EHR technology was, at all relevant times: connected in accordance with applicable law; compliant with all standards applicable to the exchange of information, including the standards, implementation specifications, and certification criteria adopted at 45 CFR part 170; implemented in a manner that allowed for timely access by patients to their electronic health information; (including the ability to view, download, and transmit this information) and implemented in a manner that allowed for the timely, secure, and trusted bi-directional exchange of structured electronic health information with other health care providers (as defined by 42 U.S.C. 300jj(3)), including unaffiliated providers, and with disparate certified EHR technology and vendors.

Third, the eligible clinician, EP, eligible hospital, or CAH would be required to attest that it responded in good faith and in a timely manner to requests to retrieve or exchange electronic health information, including from patients, health care providers (as defined by 42 U.S.C. 300jj(3)), and other persons, regardless of the requestor's affiliation or technology vendor. We invite public comment on this proposal, including whether the foregoing statements could provide the Secretary with adequate assurances that an eligible clinician, EP, eligible hospital, or CAH has complied with the statutory requirements for information exchange. We also encourage public comment on whether there are additional facts or circumstances to which eligible clinicians, EPs, eligible hospitals, or CAHs should be required to attest, or whether there is additional information that they should be required to report.

D. Definitions

At § 414.1305, subpart O, we are proposing definitions for the following terms:

• Additional performance threshold.
• Advanced Alternative Payment Model (Advanced APM).
• Advanced APM Entity.
• Affiliated practitioner.
• Alternative Payment Model (APM).
• APM Entity.
• APM Entity group.
• APM Incentive Payment.
• Attestation.
• Attributed beneficiary.
• Attribution-eligible beneficiary.
• Certified Electronic Health Record Technology (CEHRT).
• Clinical Practice Improvement Activity (CPIA).
• CMS-approved survey vendor.
• CMS Web Interface.
• Composite performance score (CPS).
• Covered professional services.
• Eligible clinician.
• Episode payment model.
• Estimated aggregate payment amounts.
• Group.
• Health professional shortage areas (HPSA).
• High priority measure.
• Hospital-based MIPS eligible clinician.
• Incentive payment base period.
• Low-volume threshold.
• Meaningful EHR user for MIPS.
• Measure benchmark.
• Medicaid APM.
• Medical Home Model.
• Medicaid Medical Home Model.
• Merit-Based Incentive Payment System (MIPS).
• MIPS APM.
• MIPS Payment Year.

MIPS eligible clinician.

• MIPS payment year.
• New Medicare-Enrolled MIPS eligible clinician.
• Non-patient-facing MIPS eligible clinician.
• Other Payer Advanced APM.
• Partial Qualifying APM Participant (Partial QP).
• Partial QP patient count threshold.
• Partial QP payment amount threshold.
• Participation List.
• Performance category score.
• Performance standards.
• Performance threshold.
• Qualified Clinical Data Registry (QCDR).
• Qualified registry.
• QP patient count threshold.
• QP payment amount threshold.
• QP Performance Period.
• Qualifying APM Participant (QP).
• Rural areas.
• Small practices.
• Threshold Score.
• Topped out measure.

Some of these terms are new in conjunction with MIPS and APMs, while others are used in existing CMS programs. For the new proposed terms and definitions, we note that some of them have been developed alongside proposed policies of this regulation while others are defined by statute. Specifically, the following terms and definitions were established by the MACRA: APM, CPIA, Eligible Alternative Payment Entity (which we have termed Advanced APM Entity), Eligible professional or EP (which we have termed eligible clinician), MIPS Eligible professional or MIPS EP (which we have termed MIPS eligible clinicians), Qualifying APM Participant, and Partial Qualifying APM Participant.

We invite public comments on all of these proposed terms and definitions, and discuss most of them in detail in relevant sections of this preamble.

E. MIPS Program Details

1. MIPS Eligible Clinicians

We believe a successful MIPS program fully equips clinicians identified as MIPS eligible clinicians with the tools and incentives to focus on improving health care quality, efficiency, and patient safety for all their patients. Under MIPS, MIPS eligible clinicians are incentivized to engage in proven improvement measures and activities that impact patient health and safety and are relevant for their patient population. One of our strategic goals in developing the MIPS program is to advance a program that is meaningful, understandable, and flexible for participating MIPS eligible clinicians. One way we believe this will be accomplished is by minimizing MIPS eligible clinicians' burden. We have made an effort to focus on policies that remove as much administrative burden as possible from MIPS eligible clinicians and their practices while still providing meaningful incentives for high-quality, efficient care. In addition, we hope to balance practice diversity with flexibility to address varied MIPS eligible clinicians' practices. Examples of this flexibility include special consideration for non-patient-facing MIPS eligible clinicians, an exclusion from MIPS for eligible clinicians who do not exceed the low-volume threshold, and other proposals discussed below.

a. Definition of a MIPS Eligible Clinician

Section 1848(q)(1)(C)(i) of the Act, as added by section 101(c)(1) of the MACRA, outlines the general definition of a MIPS eligible clinician for the MIPS program. Specifically, for the first and second year for which MIPS applies to payments (and the performance period for such years) a MIPS eligible clinician is defined as a physician (as defined in section 1861(r) of the Act), a physician assistant, nurse practitioner, and clinical nurse specialist (as such terms are defined in section 1861(aa)(5) of the Act), a certified registered nurse anesthetist (as defined in section 1861(bb)(2) of the Act), and a group that includes such professionals. The statute also provides flexibility to specify additional eligible clinicians (as defined in section 1848(k)(3)(B) of the Act) as MIPS eligible clinicians in the third and subsequent years of MIPS. As discussed in section II.E.3. of this proposed rule, section 1848(q)(1)(C)(ii) and (v) of the Act specifies several exclusions from the definition of a MIPS eligible clinician. In addition, section 1848(q)(1)(A) of the Act requires the Secretary to permit any eligible clinician (as defined in section 1848(k)(3)(B) of the Act) who is not a MIPS eligible clinician the option to volunteer to report on applicable measures and activities under MIPS. Section 1848(q)(1)(C)(vi) of the Act clarifies that a MIPS adjustment factor (or additional MIPS adjustment factor) will not be applied to an individual who is not a MIPS eligible clinician for a year, even if such individual voluntarily reports measures under MIPS.

To implement the MIPS program we must first establish and define a MIPS eligible clinician in accordance with the statutory definition. We propose to define a MIPS eligible clinician at § 414.1305 as a physician (as defined in section 1861(r) of the Act), a physician assistant, nurse practitioner, and clinical nurse specialist (as such terms are defined in section 1861(aa)(5) of the Act), a certified registered nurse anesthetist (as defined in section 1861(bb)(2) of the Act), and a group that includes such professionals. In addition, we propose that Qualifying APM Participants, Partial Qualifying APM Participants who do not report data under MIPS, low-volume threshold eligible clinicians, and new Medicare-enrolled eligible clinicians as defined at § 414.1305 would be excluded from this definition per the statutory exclusions defined in section 1848(q)(1)(C)(ii) and (v) of the Act. We intend to consider using our authority under section 1848(q)(1)(C)(i)(II) of the Act to expand the definition of MIPS eligible clinician to include additional eligible clinicians (as defined in section 1848(k)(3)(B) of the Act) through rulemaking in future years.

In addition, in accordance with section 1848(q)(1)(A) and (q)(1)(C)(vi) of the Act, we propose to allow eligible clinicians who are not MIPS eligible clinicians as defined at proposed § 414.1305 the option to voluntarily report measures and activities for MIPS. We propose at § 414.1310(d) that those eligible clinicians who are not MIPS eligible clinicians, but who voluntarily report on applicable measures and activities specified under MIPS, would not receive an adjustment under MIPS; however, they will have the opportunity to gain experience in the MIPS program. We are particularly interested in public comment regarding the feasibility and advisability of voluntary reporting in the MIPS program for entities such as Rural Health Clinics (RHCs) and/or Federally Qualified Health Centers (FQHCs), including comments regarding the specific technical issues associated with reporting that are unique to these health care providers. We anticipate some eligible clinicians that will not be MIPS eligible clinicians during the first 2 years of MIPS, such as physical and occupational therapists, clinical social workers, and others that have been reporting quality measures under the PQRS for a number of years, will want to have the ability to continue to report and gain experience under MIPS. We request comments on these proposals.

b. Non-Patient-Facing MIPS Eligible Clinicians

Section 1848(q)(2)(C)(iv) of the Act requires the Secretary, in specifying measures and activities for a performance category, to give consideration to the circumstances of professional types (or subcategories of those types determined by practice characteristics) who typically furnish services that do not involve face-to-face interaction with a patient. To the extent feasible and appropriate, the Secretary may take those circumstances into account and apply alternative measures or activities that fulfill the goals of the applicable performance category to such non-patient-facing MIPS eligible clinicians. In carrying out these provisions, we are required to consult with non-patient-facing MIPS eligible clinicians.

In addition, section 1848(q)(5)(F) of the Act allows the Secretary to re-weight MIPS performance categories if there are not sufficient measures and activities applicable and available to each type of MIPS eligible clinician. We assume many non-patient-facing MIPS eligible clinicians will not have sufficient measures and activities applicable and available to report under the performance categories under MIPS. We refer readers to section II.E.6. of this proposed rule to discuss how we address performance categories weighting for MIPS eligible clinicians for whom no measures exist in a given category.

To establish policies surrounding non-patient-facing MIPS eligible clinicians, we must first define the term “non-patient-facing.” Currently, the PQRS, VM, and Medicare EHR Incentive Program include two existing policies for considering whether an EP is providing patient-facing services. To determine, for purposes of PQRS, whether an EP had a “face-to-face” encounter with Medicare patients, we assess whether the EP billed for services under the PFS that are associated with face-to-face encounters, such as whether an EP billed general office visit codes, outpatient visits, and surgical procedures. Under PQRS, if an EP bills for at least one service under the PFS during the performance period that is associated with face-to-face encounters and reports quality measures via claims or registries, then the EP is required to report at least one “cross-cutting” measure. EPs who do not meet these criteria are not required to report a cross-cutting measure. For the purposes of PQRS, telehealth services have not historically been included in the definition of face-to-face encounters. For more information, please see the CY 2016 PFS final rule for these discussions (80 FR 71140).

In the Stage 2 final rule (77 FR 54098 through 54099), the Medicare EHR Incentive Program established a significant hardship exception from the meaningful use payment adjustment under section 1848(a)(7)(A) of the Act for EPs that lack face-to-face interactions with patients and those who lack the need to follow-up with patients. EPs with a primary specialty of anesthesiology, pathology or radiology listed in the Provider Enrollment, Chain, and Ownership System (PECOS) as of 6 months prior to the first day of the payment adjustment year automatically receive this hardship exemption (77 FR 54100). Codes associated with these specialties include 05 Anesthesiology, 22 Pathology, 30 Diagnostic Radiology, 36 Nuclear Medicine, 94 Interventional Radiology. EPs with a different specialty are also able to request this hardship exception through the hardship application process. However, telehealth services could be counted by EPs who choose to include these services within the definition of “seen by the EP” for the purposes of calculating patient encounters with the EHR Incentive Program (77 FR 53982).

In the MIPS and APMs RFI, we sought comments on MIPS eligible clinicians that should be considered non-patient-facing MIPS eligible clinicians and the criteria we should use to identify these MIPS eligible clinicians. Commenters were split when it came to defining and identifying non-patient-facing MIPS eligible clinicians. Many took a specialty-driven approach. Commenters generally did not support use of enrollment specialty codes alone, which is the approach used by the Medicare EHR Incentive Program. Commenters indicated that these codes do not necessarily delineate between the same specialists who may or may not have patient-facing interaction. One example is cardiologists who specialize in cardiovascular imaging which is also coded as cardiology. On the other hand, as one commenter mentioned, physicians with enrollment specialty codes other than “cardiology” (for example, internal medicine) may perform cardiovascular imaging services. Therefore, using the enrollment specialty code for cardiology to identify clinicians who typically do not provide patient-facing services would be both over-inclusive and under-inclusive. Other commenters identified specialty types that they believe should be considered non-patient-facing MIPS eligible clinicians. Specific specialty types included radiologists, anesthesiologists, nuclear cardiology or nuclear medicine physicians, and pathologists. Others pointed out that certain MIPS eligible clinicians may be primarily non-patient-facing MIPS eligible clinicians even though they practice within a traditionally patient-facing specialty. The MIPS and APMs RFI comments and listening sessions with medical societies representing non-patient-facing MIPS eligible clinicians specified radiology/imaging, anesthesiology, nuclear cardiology and oncology, and pathology as inclusive of non-patient-facing MIPS eligible clinicians. Commenters noted that roles within specific types of specialties may need to be further delineated between patient-facing and non-patient-facing MIPS eligible clinicians. An illustrative list of specific types of clinicians within the non-patient-facing spectrum include:

• Pathologists who may be primarily dedicated to working with local hospitals to identify early indicators related to evolving infectious diseases;
• Radiologists who primarily provide consultative support back to a referring physician or provide image interpretation and diagnosis versus therapy;
• Nuclear medicine physicians who play an indirect role in patient care, for example as a consultant to another physician in proper dose administration; or
• Anesthesiologists who are primarily providing supervision oversight to Certified Registered Nurse Anesthetists.

Some commenters believed that MIPS eligible clinicians should be defined as non-patient-facing MIPS eligible clinicians based on whether their billing indicates they provide face-to-face services. Commenters indicated that the use of specific HCPCS codes in combination with enrollment specialty codes, may be a more appropriate way to identify MIPS eligible clinicians that have no patient interaction.

After reviewing current policies, we propose to define a non-patient-facing MIPS eligible clinicians for MIPS at § 414.1305 as an individual MIPS eligible clinician or group that bills 25 or fewer patient-facing encounters during a performance period. We consider a patient-facing encounter as an instance in which the MIPS eligible clinician or group billed for services such as general office visits, outpatient visits, and surgical procedure codes under the PFS. We intend to publish the proposed list of patient-facing encounter codes on a CMS Web site similar to the way we currently publish the list of face-to-face encounter codes for PQRS. This proposal differs from the current PQRS policy in two ways. First, it creates a minimum threshold for the quantity of patient-facing encounters that MIPS eligible clinicians or groups would need to furnish to be considered patient-facing, rather than classifying MIPS eligible clinicians as patient-facing based on a single patient-facing encounter. Second, this proposal includes telehealth services in the definition of patient-facing encounters.

We believe that setting the non-patient-facing MIPS eligible clinician threshold for individual MIPS eligible clinician or group at 25 or fewer billed patient-facing encounters during a performance period is appropriate. We selected this threshold based on an analysis of non-patient-facing HCPCS codes billed by MIPS eligible clinicians. Using these codes and this threshold we identified approximately one quarter of MIPS eligible clinicians as non-patient-facing before MIPS exclusions, such as low-volume and newly-enrolled eligible clinician policies, were applied. The majority of clinicians enrolled in Medicare with specialties such as anesthesiology, nuclear medicine, and pathology were identified as non-patient-facing in this analysis. The addition of telemedicine to the analysis did not affect the outcome, as it created a less than 0.01 percent change in MIPS eligible clinicians categorized as non-patient-facing.

Therefore, this proposed approach allows the definition of non-patient-facing MIPS eligible clinicians, to include both MIPS eligible clinicians who practice within specialties traditionally considered non-patient-facing, as well as MIPS eligible clinicians who provide occasional patient-facing services that do not represent the bulk of their practices. This definition is also consistent with the statutory requirement that refers to professional types who typically furnish services that do not involve patient-facing interaction with a patient.

We also propose to include telehealth services in the definition of patient-facing encounters. Various MIPS eligible clinicians use telehealth services as an innovative way to deliver care to beneficiaries and we believe these services, while not furnished in-person, should be recognized as patient-facing. In addition, Medicare eligible telehealth services substitute for an in-person encounter and meet other site requirements under the PFS as defined at § 410.78.

The proposed addition of the encounter threshold for patient-facing MIPS eligible clinicians should minimize concerns that a MIPS eligible clinician could be misclassified as patient-facing as a result of providing occasional telehealth services that do not represent the bulk of their practice. Finally, this proposed definition of a non-patient-facing MIPS eligible clinician for MIPS can be consistently used throughout the MIPS program to identify those MIPS eligible clinicians for whom certain proposed requirements for patient-facing MIPS eligible clinicians (such as reporting cross-cutting measures) may not be meaningful.

We weighed several options when considering the appropriate definition of non-patient-facing MIPS eligible clinicians for MIPS; and some options were similar to those we considered in implementing the Medicare EHR Incentive Program. One option we considered was basing the non-patient-facing MIPS eligible clinician's definition on a set percentage of patient-facing encounters, such as 5 to 10 percent, that is tied to the same list of patient-facing encounter codes discussed in this section of the proposed rule. Another option we considered was the identification of non-patient-facing MIPS eligible clinicians for MIPS only by specialty, which might be a simpler approach. However, we do not consider this approach sufficient for identifying all the possible non-patient-facing MIPS eligible clinicians, as some patient-facing MIPS eligible clinicians practice in multi-specialty practices with non-patient-facing MIPS eligible clinician's practices with different specialties. We would likely have had to develop a separate process to identify non-patient-facing MIPS eligible clinicians in other specialties, whereas maintaining a single definition that is aligned across performance categories is simpler. Many comments from the MIPS and APMs RFI discouraged use of enrollment specialty alone. Additionally, we believe our proposal would allow us to more accurately identify MIPS eligible clinicians who are non-patient-facing by applying a threshold to recognize that a MIPS eligible clinician who furnishes almost exclusively non-patient-facing services should be treated as a non-patient-facing MIPS eligible clinicians despite furnishing a small number of patient-facing services. We seek comment on these alternative approaches.

In the MIPS and APMs RFI, we also requested comments on what types of measures and/or CPIAs (new or from other payment systems) we should use to assess non-patient-facing MIPS eligible clinicians' performance and how we should apply the MIPS performance categories to non-patient-facing MIPS eligible clinicians. Commenters were split on these subjects. A number of commenters stated that non-patient-facing MIPS eligible clinicians should be exempt from specific performance categories under MIPS or should be exempt from MIPS as a whole. Commenters who did not favor exemptions generally suggested that we focus on process measures and work with specialty societies to develop new, more clinically relevant measures for non-patient-facing MIPS eligible clinicians.

We took these stakeholder comments into consideration. We note that section 1848(q)(2)(C)(iv) of the Act does not grant the Secretary discretion to exempt non-patient-facing MIPS eligible clinicians from a performance category entirely, but rather to apply to the extent feasible and appropriate alternative measures or activities that fulfill the goals of the applicable performance category. However, we have placed safeguards to ensure that MIPS eligible clinicians, including non-patient facing, that do not have sufficient alternative measures that are applicable and available in a performance category are scored appropriately. We propose to apply the Secretary's authority under section 1848(q)(5)(F) of the Act to reweight such performance categories score to zero if there is no performance category score or to lower the weight of the quality performance category score if there are not at least three scored measures. Please refer to section II.E.6.b.(2)(b) in this proposed rule for details on the reweighting proposals. Accordingly, we have proposed alternative requirements for non-patient-facing MIPS eligible clinicians across this proposed rule (see sections II.E.5.b. II.E.5.e. and II.E.5.f. of this proposed rule for more details). While non-patient-facing MIPS eligible clinicians will not be exempt from any performance category under MIPS, we believe these alternative requirements fulfill the goals of the applicable performance categories and are in line with the commenters' desire to ensure that non-patient-facing MIPS eligible clinicians are not placed at an unfair disadvantage under the new program. The requirements also build on prior program components in meaningful ways and are meant to help us appropriately assess and incentivize non-patient-facing MIPS eligible clinicians. We request comments on these proposals.

c. MIPS Eligible Clinicians Who Practice in Critical Access Hospitals Billing Under Method II (Method II CAHs)

Section 1848(q)(6)(E) of the Act provides that the MIPS adjustment is applied to the amount otherwise paid under Part B for the items and services furnished by a MIPS eligible clinician during a year (beginning with 2019). In the case of MIPS eligible clinicians who practice in CAHs that bill under Method I (“Method I CAHs”), the MIPS adjustment would apply to payments made for items and services billed by MIPS eligible clinicians under the PFS, but it would not apply to the facility payment to the CAH itself. In the case of MIPS eligible clinicians who practice in Method II CAHs and have not assigned their billing rights to the CAH, the MIPS adjustment would apply in the same manner as for MIPS eligible clinicians who bill for items and services in Method I CAHs.

Under section 1834(g)(2) of the Act, a Method II CAH bills and is paid for facility services at 101 percent of its reasonable costs and for professional services at 115 percent of such amounts as would otherwise be paid under this part if such services were not included in outpatient critical access hospital services. In the case of MIPS eligible clinicians who practice in Method II CAHs and have assigned their billing rights to the CAHs, those professional services would constitute “covered professional services” under section 1848(k)(3)(A) of the Act because they are furnished by an eligible clinician and payment is “based on” the PFS. Moreover, this is consistent with the precedent CMS has established by applying the PQRS and EHR-MU adjustments to Method II CAH payments. Therefore, we propose the MIPS adjustment does apply to Method II CAH payments under section 1834(g)(2)(B) of the Act when MIPS eligible clinicians who practice in Method II CAHs have assigned their billing rights to the CAH. We request comments on this proposal.

d. MIPS Eligible Clinicians Who Practice in Rural Health Clinics (RHCs) and/or Federally Qualified Health Centers (FQHCs)

As noted previously in this proposed rule, section 1848(q)(6)(E) of the Act provides that the MIPS adjustment is applied to the amount otherwise paid under Part B with respect to the items and services furnished by a MIPS eligible clinician during a year. Some eligible clinician s may not receive MIPS adjustments due to their billing methodologies. If a MIPS eligible clinician furnishes items and services in an RHC and/or FQHC and the RHC and/or FQHC bills for those items and services under the RHC's or FQHC's all-inclusive payment methodology, the MIPS adjustment would not apply to the facility payment to the RHC or FQHC itself. However, if a MIPS eligible clinician furnishes other items and services in an RHC and/or FQHC and bills for those items and services under the PFS, the MIPS adjustment would apply to payments made for items and services. Accordingly, the MIPS eligible clinician would need to meet the applicable MIPS reporting requirements to avoid a downward MIPS adjustment to payments made for items and services billed by the MIPS eligible clinician under the PFS. Therefore, we propose services rendered by an eligible clinician that are payable under the RHC or FQHC methodology would not be subject to the MIPS payments adjustments. However, these eligible clinicians have the option to voluntarily report on applicable measures and activities for MIPS and the data received would not be used to assess their performance for the purpose of the MIPS adjustment. We request comments on this proposal.

e. Group Practice (Group)

Section 1848(q)(1)(D) of the Act, requires the Secretary to establish and apply a process that includes features of the PQRS group practice reporting option (GPRO) established under section 1848(m)(3)(C) of the Act for MIPS eligible clinicians in a group for purposes of assessing performance in the quality performance category. In addition, it gives the Secretary the discretion to do so for the other three performance categories. Additionally, we will assess performance either for individual MIPS eligible clinicians or for groups. As discussed in section II.E.2.b of this proposed rule, we propose to define a group at § 414.1305 as a single Taxpayer Identification Number (TIN) with two or more MIPS eligible clinicians, as identified by their individual National Provider Identifier (NPI), who have reassigned their Medicare billing rights to the TIN. Also, as outlined in section II.E.2.c. of this proposed rule, we propose to define an APM Entity group at § 414.1305 identified by a unique APM participant identifier.

2. MIPS Eligible Clinician Identifier

To support MIPS eligible clinicians reporting to a single comprehensive and cohesive MIPS program, we need to align the technical reporting requirements from PQRS, VM, and EHR-MU into one program. This requires an appropriate MIPS eligible clinician identifier. We currently use a variety of identifiers to assess an individual eligible clinician or group under different programs. For example, under the PQRS for individual reporting, CMS uses a combination of TIN and NPI to assess eligibility and participation, where each unique TIN and NPI combination is treated as a distinct eligible clinician and is separately assessed for purposes of the program. Under the PQRS GPRO, eligibility and participation are assessed at the TIN level. Under the Medicare EHR Incentive Program, we utilize the NPI to assess eligibility and participation. And under the VM, performance and payment adjustments are assessed at the TIN level. Additionally, for APMs such as the Pioneer Accountable Care Organization (ACO) Model, we also assign a program-specific identifier (in the case of the Pioneer ACO Model, an ACO ID) to the organization(s), and associate that identifier with individual eligible clinicians who are, in turn, identified through a combination of a TIN and an NPI.

In the MIPS and APMs RFI, we sought comments on which specific identifier(s) should be used to identify a MIPS eligible clinician for purposes of determining eligibility, participation, and performance under the MIPS performance categories. In addition, we requested comments pertaining to what safeguards should be in place to ensure that MIPS eligible clinicians do not switch identifiers to avoid being considered “poor-performing” and comments on what safeguards should be in place to address any unintended consequences, if the MIPS eligible clinician identifier were a unique TIN/NPI combination, to ensure an appropriate assessment of the MIPS eligible clinician's performance. In the MIPS and APMs RFI, we sought comment on using a MIPS eligible clinician's TIN, NPI, or TIN/NPI combination as potential MIPS eligible clinician identifiers, or creating a unique MIPS eligible clinician identifier. The commenters did not demonstrate a consensus on a single best identifier.

Commenters favoring the use of the MIPS eligible clinician's TIN recommended that MIPS eligible clinicians should be associated with the TIN used for receiving payment from CMS claims. They further commented that this approach will deter MIPS eligible clinicians from “gaming” the system by switching to a higher performing group. Under this approach, commenters suggest that MIPS eligible clinicians who bill under more than one TIN can be assigned the performance and payment adjustment for the primary practice based upon majority of dollar amount of claims or encounters from the prior year.

Other commenters supported using unique TIN and NPI combinations to identify MIPS eligible clinicians. Commenters suggested many eligible clinicians are familiar with using TIN and NPI together from PQRS and other CMS programs. Commenters also noted this approach can calculate performance for multiple unique TIN/NPI combinations for those MIPS eligible clinicians who practice under more than one TIN. Commenters who supported the TIN/NPI also believe this approach enables greater accountability for individual MIPS eligible clinicians beyond what might be achieved when using TIN as an identifier and would provide a safeguard from MIPS eligible clinicians changing their identifier to avoid payment penalties.

Some commenters supported the use of only the NPI as the MIPS identifier. They believe this approach would best provide for individual accountability for quality in MIPS while minimizing potential confusion because providers do not generally change their NPI over time. Supporters of using the NPI only as the MIPS identifier also commented that this approach would be simplest for administrative purposes. These commenters also note the continuity inherent with the NPI would address the safeguard issue of providers attempting to change their identifier for MIPS performance purposes.

In the MIPS and APMs RFI, we also solicited feedback on the potential for creating a new MIPS identifier for the purposes of identifying MIPS eligible clinicians within the MIPS program. In response, many commenters indicated they would not support a new MIPS identifier. Commenters generally expressed concern that a new identifier for MIPS would only add to administrative burden, create confusion for MIPS eligible clinicians and increase reporting errors.

After reviewing the comments, we are not proposing to create a new MIPS eligible clinician identifier. However, we appreciate the various ways a MIPS eligible clinician may engage with MIPS, either individually or through a group. Therefore, we are proposing to use multiple identifiers that allow MIPS eligible clinicians to be measured as an individual or collectively through a group's performance. We also propose that the same identifier be used for all four performance categories; for example, if a group is submitting information collectively, then it must be measured collectively for all four MIPS performance categories: Quality, resource use, CPIA, and advancing care information. As discussed later in the CPS methodology section II.E.6. of this proposed rule, while we have multiple identifiers for participation and performance, we proposed to use a single identifier, TIN/NPI, for applying the payment adjustment, regardless of how the MIPS eligible clinician is assessed. Specifically, if the MIPS eligible clinician is identified for performance only using the TIN, when applying the payment adjustment we propose to use the TIN/NPI. We request comments on these proposals.

a. Individual Identifiers

We propose to use a combination of billing TIN/NPI as the identifier to assess performance of an individual MIPS eligible clinician. Similar to PQRS, each unique TIN/NPI combination would be considered a different MIPS eligible clinician, and MIPS performance would be assessed separately for each TIN under which an individual bills. While we considered using the NPI only, we believe TIN/NPI is a better approach for MIPS. Both TIN and NPI are needed for payment purposes and using a combination of billing TIN/NPI as the MIPS eligible clinician identifier allows us to match MIPS performance and payment adjustments with the appropriate practice, particularly for MIPS eligible clinicians that bill under more than one TIN. In addition, using TIN/NPI also provides the flexibility to allow individual MIPS eligible clinician and group reporting, as the group identifiers being proposed also include TIN as part of the identifier. We recognize that TIN/NPI is not a static identifier and can change if an individual MIPS eligible clinician changes practices and/or if a group merges with another between the performance period and payment adjustment period. Section II.E.5.h. of this proposed rule describes in more detail how we propose to match performance in cases where the TIN/NPI changes. We request comments on this proposal.

b. Group Identifiers for Performance

We propose the following way a MIPS eligible clinician may have their performance assessed as part of a group under MIPS. We propose to use a group's billing TIN to identify a group. This approach has been used as a group identifier for both PQRS and VM. The use of the TIN would significantly reduce the participation burden that could be experienced by large groups. Additionally, the utilization of the TIN benefits large and small practices by allowing such entities to submit performance data one time for their group and develop systems to improve performance. Groups that report on quality performance measures through certain data submission methods must register in order to participate in MIPS as described in section II.E.5.b. of this proposed rule.

We are proposing to codify the definition of a group at § 414.1305 as a group that would consist of a single TIN with two or more MIPS eligible clinicians (as identified by their individual NPI) who have reassigned their billing rights to the TIN. We request comments on this proposal.

c. APM Entity Group Identifier for Performance

We propose the following way to identify a group to support APMs (see section II.F.5.b. of this proposed rule). To ensure we have accurately captured all of the eligible clinicians identified as participants that are participating in the APM Entity, we propose that each eligible clinician who is a participant of an APM Entity would be identified by a unique APM participant identifier. The unique APM participant identifier would be a combination of four identifiers: (1) APM Identifier (established by CMS; for example, XXXXXX); (2) APM Entity identifier (established under the APM by CMS; for example, AA00001111); (3) TIN(s) (9 numeric characters; for example, XXXXXXXXX); (4) EP NPI (10 numeric characters; for example, 1111111111). For example, an APM participant identifier could be APM XXXXXX, APM Entity AA00001111, TIN-XXXXXXXXX, NPI-11111111111.

We are proposing to codify the definition of an APM Entity group at § 414.1305 as an APM Entity identified by a unique APM participant identifier. We request comments on these proposals. See section II.E.5.h. of this rule for proposed policies regarding requirements for APM Entity groups under MIPS.

3. Exclusions

a. New Medicare-Enrolled Eligible Clinician

Section 1848(q)(1)(C)(v) of the Act provides that in the case of a professional who first becomes a Medicare-enrolled eligible clinician during the performance period for a year (and had not previously submitted claims under Medicare either as an individual, an entity, or a part of a physician group or under a different billing number or tax identifier), that the eligible clinician will not be treated as a MIPS eligible clinician until the subsequent year and performance period for that year. In addition, section 1848(q)(1)(C)(vi) of the Act clarifies that individuals who are not deemed MIPS eligible clinicians for a year will not receive a MIPS adjustment factor (or additional MIPS adjustment factor). Accordingly, we propose at § 414.1305 that a new Medicare-enrolled eligible clinician be defined as a professional who first becomes a Medicare-enrolled eligible clinician within the PECOS during the performance period for a year and who has not previously submitted claims as a Medicare-enrolled eligible clinician either as an individual, an entity, or a part of a physician group or under a different billing number or tax identifier. These eligible clinicians will not be treated as a MIPS eligible clinician until the subsequent year and the performance period for such subsequent year. As discussed in section II.E.4. of this proposed rule, we are proposing that the MIPS performance period would be the calendar year (January 1 through December 31) 2 years prior to the year in which the MIPS adjustment is applied. For example, an eligible clinician who newly enrolls in Medicare within PECOS in 2017 would not be required to participate in MIPS in 2017, and he or she would not receive a MIPS adjustment in 2019. The same eligible clinician would be required to participate in MIPS in 2018 and would receive a MIPS adjustment in 2020, and so forth. In addition, in the case of items and services furnished during a year by an individual who is not an MIPS eligible clinician, there will not be a MIPS adjustment factor (or additional MIPS adjustment factor) applied for that year. We also propose at § 414.1310(d) that in no case would a MIPS adjustment factor (or additional MIPS adjustment factor) apply to the items and services furnished by new Medicare-enrolled eligible clinicians.

We request comments on these proposals.

b. Qualifying APM Participants (QP) and Partial Qualifying APM Participant (Partial QP)

Sections 1848(q)(1)(C)(ii)(I) and (II) of the Act provide that the definition of a MIPS eligible clinician does not include, for a year, an eligible clinician who is a Qualifying APM Participant (QP) (as defined in section 1833(z)(2) of the Act) or a Partial Qualifying APM Participant (Partial QP) (as defined in section 1848(q)(1)(C)(iii) of the Act) who does not report on the applicable measures and activities that are required under MIPS. Section II.F.5. of this proposed rule provides detailed information on the determination of QPs and Partial QPs.

We propose that the definition of a MIPS eligible clinician at § 414.1310 does not include qualifying APM participants (defined at § 414.1305) and Partial QPs defined at § 414.1305 who do not report on applicable measures and activities that are required to be reported under MIPS for any given performance period. Partial QPs will have the option to elect whether or not to report under MIPS, which determines whether or not they will be subject to MIPS adjustments. Please refer to the section II.F.5.c. of this proposed rule where this election is discussed in greater detail. We request comments on this proposal.

c. Low-Volume Threshold

Section 1848(q)(1)(C)(ii)(III) of the Act provides that the definition of a MIPS eligible clinician does not include MIPS eligible clinicians who are below the low-volume threshold selected by the Secretary under section 1848(q)(1)(C)(iv) of the Act for a given year. Section 1848(q)(1)(C)(iv) of the Act requires the Secretary to select a low-volume threshold to apply for the purposes of this exclusion which may include one or more of the following: (1) The minimum number, as determined by the Secretary, of Part B-enrolled individuals who are treated by the MIPS eligible clinician for a particular performance period; (2) the minimum number, as determined by the Secretary, of items and services furnish to Part B-enrolled individuals by the MIPS eligible clinician for a particular performance period; and (3) the minimum amount, as determined by the Secretary, of allowed charges billed by the MIPS eligible clinician for a particular performance period.

We propose at § 414.1305 to define MIPS eligible clinicians or groups who do not exceed the low-volume threshold as an individual MIPS eligible clinician or group who, during the performance period, have Medicare billing charges less than or equal to $10,000 and provides care for 100 or fewer Part B-enrolled Medicare beneficiaries. We believe this strategy is value-oriented as it retains as MIPS eligible clinicians those MIPS eligible clinicians who are treating relatively few beneficiaries, but engage in resource intensive specialties, or those treating many beneficiaries with relatively low-priced services. By requiring both criteria be met, we can meaningfully measure the performance and drive quality improvement across the broadest range of MIPS eligible clinician types and specialties. Conversely, it excludes MIPS eligible clinicians who do not have a substantial quantity of interactions with Medicare beneficiaries or furnish high cost services.

In developing this proposal we considered using items and services furnished to Part B-enrolled individuals by the MIPS eligible clinician for a particular performance period rather than patients but a review of the data reflected there were nominal differences between the two methods. We plan to monitor the proposed requirement and anticipate that the specific thresholds will evolve over time. We request comments on this proposal including alternative patient threshold, case thresholds, and dollar values.

d. Group Reporting

(1) Background

As noted above, section 1848(q)(1)(D) of the Act, requires the Secretary to establish and apply a process that includes features of the PQRS group practice reporting option (GPRO) established under section 1848(m)(3)(C) of the Act for MIPS eligible clinicians in a group for the purpose of assessing performance in the quality category and give the Secretary the discretion to do so for the other performance categories. The process established for purposes of MIPS must, to the extent practicable, reflect the range of items and services furnished by the MIPS eligible clinicians in the group. We believe this means that the process established for purposes of MIPS should, to the extent practicable, encompass elements that enable MIPS eligible clinicians in a group to meet reporting requirements that reflect the range of items and services furnished by the MIPS eligible clinicians in the group. At § 414.1310(e) we propose requirements for groups. For purposes of section 1848(q)(1)(D) of the Act, at § 414.1310(e)(1) we propose the following way for individual MIPS eligible clinicians to have their performance assessed as a group: As part of a single TIN associated with two or more MIPS eligible clinicians, as identified by a NPI, that have their Medicare billing rights reassigned to the TIN (as discussed further in section II.E.1.f. of this proposed rule).

In order to have its performance assessed as a group, at § 414.1310(e)(2) we propose a group must meet the proposed definition of a group at all times during the performance period for the MIPS payment year. Additionally, at § 414.1310(e)(3) we propose in order to have their performance assessed as a group, individual MIPS eligible clinicians within a group must aggregate their performance data across the TIN. At § 414.1310(e)(3), we propose a group that elects to have its performance assessed as a group would be assessed as a group across all four MIPS performance categories. For example, if a group submits data for the quality performance category as a group, CMS would assess them as a group for the remaining three performance categories. We solicit public comments on the proposal regarding how groups will be assessed under MIPS.

(2) Registration

Under the PQRS, groups are required to complete a registration process to participate in PQRS as a group. During the implementation and administration of PQRS, we received feedback from stakeholders regarding the registration process for the various methods available for data submission. Stakeholders indicated that the registration process was burdensome and confusing. Additionally, we discovered that during the registration process when groups are required to select their group submission mechanism, groups sometimes selected the option not applicable to their group, which has created issues surrounding the mismatch of data. Unreconciled data mismatching can impact the quality of data. In order to address this issue, we are proposing to eliminate a registration process for groups submitting data using third party entities. When groups submit data utilizing third party entities, such as a qualified registry, health IT vendor, or QCDR, we are able to obtain group information from the third party entity and discern whether the data submitted represents group submission or individual submission once the data is submitted.

At § 414.1310(e)(5), we propose that a group must adhere to an election process established and required by CMS, as described below. We do not propose to require groups to register to have their performance assessed as a group except for groups submitting data on performance measures via participation in the CMS Web Interface or groups electing to report the Consumer Assessment of Healthcare Providers and Systems (CAHPS) for MIPS survey for the quality performance category as described further in section II.E.5.b. of this proposed rule. For all other data submission methods, groups must work with appropriate third party entities to ensure the data submitted clearly indicates that the data represent a group submission rather than an individual submission. In order for groups to elect participation via the CMS Web Interface or administration of the CAHPS for MIPS survey, we propose that such groups must register by June 30 of the applicable 12-month performance period (that is, June 30, 2017, for performance periods occurring in 2017). For the criteria regarding group reporting applicable to the four MIPS performance categories, see section II.E.5.a. of this proposed rule.

e. Virtual Groups

(1) Implementation

Section 1848(q)(5)(I) of the Act establishes the use of voluntary virtual groups for certain assessment purposes. The statute requires the establishment and implementation of a process that allows an individual MIPS eligible clinician or a group consisting of not more than 10 MIPS eligible clinicians to elect to form a virtual group with at least one other such individual MIPS eligible clinician or group of not more than 10 MIPS eligible clinicians for a performance period of a year. As determined in statute, individual MIPS eligible clinicians and groups forming virtual groups are required to make such election prior to the start of the applicable performance period under MIPS and cannot change their election during the performance period. As discussed in section II.E.4. of this proposed rule, we are proposing that the performance period would be based on a calendar year.

As we assessed the timeline for the establishment and implementation of virtual groups and applicable election process and requirements for the first performance period under MIPS, we identified significant barriers regarding the development of a technological infrastructure required for successful implementation and the operationalization of such provisions that would negatively impact the execution of virtual groups as a conducive option for MIPS eligible clinicians or groups. The development of an electronic system before policies are finalized poses several risks, particularly relating to the impediments of completing and adequately testing the system before execution and assuring that any change in policy made during the rulemaking process are reflected in the system and operationalized accordingly. We believe that it would be exceedingly difficult to make a successful system to support the implementation of virtual groups and given these factors, such implementation would compromise not only the integrity of the system, but the intent of the policies.

Additionally, we recognize that it would be impossible for us to develop an entire infrastructure for electronic transactions pertaining to an election process, reporting of data, and performance measurement before the start of the performance period beginning on January 1, 2017. Moreover, the actual implementation timeframe would be more condensed given that the development, testing, and execution of such a system would need to be completed months in advance of the beginning of the performance period in order to provide MIPS eligible clinicians and groups with an election period.

During the implementation and ongoing functionality of other programs such as PQRS, Medicare EHR Incentive Program, and VM, we received feedback from stakeholders regarding issues they encountered when submitting reportable data for these programs. With virtual groups as a new option, we want to minimize potential issues for end-users and implement a system that encourages and enables MIPS eligible clinicians and groups to participate in a virtual group. A web-based registration process, which would simplify and streamline the process for participation, is our preferred approach. Given the aforementioned dynamics discussed in this section, implementation for the calendar year 2017 performance period is infeasible as a result of the insufficient timeframe to develop a web-based registration process. We have assessed alternative approaches for the first year only, such as an email registration process, but believe that there are limitations and potential risks for numerous errors, such as submitted information being incomplete or not in the required format. A manual verification process would cause a significant delay in verifying registration due to the lack of an automated system to ensure the accuracy of the type of information submitted that is required for registration. We believe that an email registration process could become cumbersome and a burden for groups to pursue participation in a virtual group. Implementation of a web-based registration system for calendar year 2018 would provide the necessary time to establish and implement an election process and requirements applicable to virtual groups, and enable proper system development and operations. We intend to implement virtual groups for the 2018 calendar year performance period and we intend to address all of the requirements pertaining to virtual groups in future rulemaking. We request comments on factors we should consider regarding the establishment and implementation of virtual groups.

(2) Election Process

Section 1848(q)(5)(I)(iii)(I) of the Act provides that the election process must occur prior to the performance period and may not be changed during the performance period. We propose to establish an election process that would end on June 30 of a calendar year preceding the applicable performance period. During the election process, we propose that individual MIPS eligible clinicians and groups electing to be a virtual group would be required to register in order to submit reportable data. Virtual groups would be assessed across all four MIPS performance categories. In future rulemaking, we intend to address all elements relating to the election process. We solicit public comments on this proposal. Future rulemaking will outline the criteria and requirements regarding the formation of virtual groups.

4. MIPS Performance Period

MIPS incorporates many of the requirements of several programs into a single, comprehensive program. This consolidation includes key policy goals as common themes across multiple categories such as quality improvement, patient and family engagement, and care coordination through interoperable health information exchange. However, each of these legacy programs included different eligibility requirements, reporting periods, and systems for providers seeking to participate. This means that we must balance potential impacts of changes to systems and technical requirements in order to successfully synchronize reporting, as noted in the discussion regarding the definition of a MIPS eligible clinician in section II.E.1.a. of this proposed rule. We must take operational feasibility, systems impacts, and education and outreach on participation requirements into account in developing technical requirements for participation. One area where this is particularly important is in the definition of a performance period.

MIPS applies to payments for items and services furnished on or after January 1, 2019. Section 1848(q)(4) of the Act requires the Secretary to establish a performance period (or periods) for a year (beginning with 2019). Such performance period (or periods) must begin and end prior to such year and be as close as possible to such year. In addition, section 1848(q)(7) of the Act provides that, not later than 30 days prior to January 1 of the applicable year, the Secretary must make available to each MIPS eligible clinician the MIPS adjustment (and, as applicable, the additional MIPS adjustment) applicable to the MIPS eligible clinician for items and services furnished by the MIPS eligible clinician during the year.

We considered various factors when developing the policy for the MIPS performance period. Stakeholders have stated that having a performance period as close to when payments are adjusted is beneficial, even if such period would be less than a year. We have also received feedback from stakeholders that they prefer having a 1 year performance period and have further suggested that the performance period start during the calendar year. For example, having the performance period occurring from July 1 through June 30. We additionally considered operational factors, such as that a 1 year performance period may be beneficial for all four performance categories because many measures and activities cannot be reported in a shorter time frame. We also considered that data submission activities and claims for items and services furnished during the 1 year performance period (which could be used for claims- or administrative claims-based quality or resource use measures) may not be fully processed until the following year.

These circumstances will require adequate lead time to collect performance data, assess performance, and compute the MIPS adjustment so the applicable MIPS adjustment can be made available to each MIPS eligible clinician at least 30 days prior to when the payment adjustment is applied each year. For 2019, these actions will occur during 2018. In other payment systems, we have used claims that are processed within a specified time period after the end of the performance period, such as 60 or 90 days, for assessment of performance and application of the payment adjustment. For MIPS, we propose at § 414.1325(g)(2) to use claims that are processed within 90 days, if operationally feasible, after the end of the performance period for purposes of assessing performance and computing the MIPS payment adjustment. If we determine that it is not operationally feasible to have a claims data run-out for the 90-day timeframe, then we would utilize a 60-day duration.

This proposal does not affect the performance period per se, but rather the deadline by which claims for items and services furnished during the performance period need to be processed for those items and services to be included in our calculation. To the extent that claims are used for submitting data on MIPS measures and activities to us, such claims would have to be processed by no later than 90 days after the end of the applicable performance period, in order for information on the claims to be included in our calculations. As noted above, if we determine that it is not operationally feasible to have a claims data run-out for the 90-day timeframe, then we will utilize a 60-day duration. As an alternative to the above proposal, we also considered using claims that are paid within 60 days after 2017, for assessment of performance and application of the MIPS payment adjustment for 2019. We are seeking comment on both approaches.

Given the need to collect and process information, we propose at § 414.1320 that for 2019 and subsequent years, the performance period under MIPS would be the calendar year (January 1 through December 31) 2 years prior to the year in which the MIPS adjustment is applied. For example, the performance period for the 2019 MIPS adjustment would be the full calendar year 2017, that is, January 1, 2017 through December 31, 2017. We propose to use the 2017 performance year for the 2019 payment adjustment consistent with other CMS programs. This approach allows for a full year of measurement and sufficient time to base adjustments on complete and accurate information.

For individual MIPS eligible clinicians and group practices with less than 12 months of performance data to report, such as when a MIPS eligible clinician switches practices during the performance period or when a MIPS eligible clinician may have stopped practicing for some portion of the performance period (for example, a MIPS eligible clinician who is on maternity leave or has an illness), we propose that the individual MIPS eligible clinician or group would be required to report all performance data available from the performance period. Specifically, if a MIPS eligible clinician is reporting as an individual, they would report all partial year performance data. Alternatively, if the MIPS eligible clinician is reporting with a group, then the group would report all performance data available from the performance period, including partial year performance data available for the individual MIPS eligible clinician.

Under this approach, MIPS eligible clinicians with partial year performance data could achieve a positive, neutral, or negative MIPS adjustment based on their performance data. We propose this approach in order to incentivize accountability for all performance during the performance period. Two policies will help minimize the impact of partial year data. First, MIPS eligible clinicians with volume below the low-volume threshold would be excluded from any payment adjustments. Second, MIPS eligible clinicians who report measures, yet have insufficient sample size, would not be scored on those measures and activities refer to section II.E.6. of this proposed rule for further details.

To potentially refine this proposal in future years, we seek comment on methods to identify accurately MIPS eligible clinicians with less than 12-month reporting periods, notwithstanding common and expected absences due to illness, vacation, or holiday leave. Reliable identification of these MIPS eligible clinicians will allow us to analyze the characteristics of this MIPS eligible clinicians' patient population and better understand how a reduced reporting period impacts performance.

We also seek public comment on an alternative approach for future years for assessment of individual MIPS eligible clinicians with less than 12 months of performance data in the performance year. For example, if we can identify such MIPS eligible clinician's and confirm there are data issues that led to invalid performance calculations, then we could score the MIPS eligible clinician with a CPS equal to the performance threshold, which would result in a zero payment adjustment. We note this approach would not assess a MIPS eligible clinicians' performance for partial-year performance data. We do not believe that consideration of partial year performance is necessary for assessment of groups, which should have adequate coverage across MIPS eligible clinicians to provide valid performance calculations.

We also seek comment on reasonable thresholds for considering performance to be less than 12 months. For example, we expect that some MIPS eligible clinicians will take leave related to illness, vacation, and holidays. We would not anticipate applying special policies for lack of performance related to these common and expected absences assuming MIPS eligible clinicians' quality reporting includes measures with sufficient sample size to generate valid and reliable scores. We seek comment on how to account for MIPS eligible clinicians with extended leave that may affect measure sample size.

We request comments on these proposals and approaches.

5. MIPS Category Measures and Activities

a. Performance Category Measures and Reporting

(1) Statutory Requirements

Section 1848(q)(2)(A) of the Act requires the Secretary to use four performance categories in determining each MIPS eligible clinician's CPS under the MIPS: Quality; resource use; CPIA; and advancing care information. Section 1848(q)(2)(B) of the Act, subject to section 1848(q)(2)(C) of the Act, describes the measures and activities that, for purposes of the MIPS performance standards, must be specified under each performance category for a performance period.

Section 1848(q)(2)(B)(i) of the Act describes the measures and activities that must be specified under the MIPS quality performance category as the quality measures included in the annual final list of quality measures published under section 1848(q)(2)(D)(i) of the Act and the list of quality measures described in section 1848(q)(2)(D)(vi) of the Act used by QCDRs under section 1848(m)(3)(E) of the Act. Under section 1848(q)(2)(C)(i) of the Act, the Secretary must, as feasible, emphasize the application of outcome-based measures in applying section 1848(q)(2)(B)(i) of the Act. Under section 1848(q)(2)(C)(iii) of the Act, the Secretary may also use global measures, such as global outcome measures and population-based measures, for purposes of the quality performance category. Section 1848(q)(2)(B)(ii) of the Act describes the measures and activities that must be specified under the resource use performance category as the measurement of resource use for the performance period under section 1848(p)(3) of the Act, using the methodology under section 1848(r) of the Act as appropriate, and, as feasible and applicable, accounting for the cost of drugs under Part D.

Section 1848(q)(2)(C)(ii) of the Act allows the Secretary to use measures from other CMS payment systems, such as measures for inpatient hospitals, for purposes of the quality and resource use performance categories, except that the Secretary may not use measures for hospital outpatient departments, other than in the case of items and services furnished by emergency physicians, radiologists, and anesthesiologists. This proposed rule seeks comment on how it might be feasible and when it might be appropriate to incorporate measures from other systems into MIPS for clinicians that work in facilities such as inpatient hospitals. For example, it may be appropriate to use such measures when other applicable measures are not available for individual MIPS eligible clinicians or when strong payment incentives are tied to measure performance, either at the facility level or with employed or affiliated MIPS eligible clinicians.

Section 1848(q)(2)(B)(iii) of the Act describes the measures and activities that must be specified under the CPIA performance category as CPIAs under subcategories specified by the Secretary for the performance period, which must include at least the subcategories specified in section 1848(q)(2)(B)(iii)(I) through (VI) of the Act. Section 1848(q)(2)(C)(v)(III) of the Act defines a CPIA as an activity that relevant eligible clinician organizations and other relevant stakeholders identify as improving clinical practice or care delivery and that the Secretary determines, when effectively executed, is likely to result in improved outcomes. Section 1848(q)(2)(B)(iii) of the Act requires the Secretary to give consideration to the circumstances of small practices (consisting of 15 or fewer professionals) and practices located in rural areas and geographic HPSAs in establishing CPIAs.

Section 1848(q)(2)(B)(iv) of the Act describes the measures and activities that must be specified under the advancing care information performance category as the requirements established for the performance period under section 1848(o)(2) for determining whether an eligible clinician is a meaningful EHR user.

As discussed in section II.E.1.b. of this proposed rule, section 1848(q)(2)(C)(iv) of the Act requires the Secretary to give consideration to the circumstances of non-patient facing MIPS eligible clinicians in specifying measures and activities under the MIPS performance categories and allows the Secretary, to the extent feasible and appropriate, to take those circumstances into account and apply alternative measures or activities that fulfill the goals of the applicable performance category. In doing so, the Secretary is required to consult with non-patient facing professionals.

Section 101(b) of MACRA amends certain provisions of section 1848(k), (m), (o), and (p) of the Act to generally provide that the Secretary will carry out such provisions in accordance with section 1848(q)(1)(F) of the Act for purposes of MIPS. Section 1848(q)(1)(F) of the Act provides that, in applying a provision of section 1848(k), (m), (o), and (p) of the Act for purposes of MIPS, the Secretary must adjust the application of the provision to ensure that it is consistent with the MIPS requirements and must not apply the provision to the extent that it is duplicative with a MIPS provision.

(2) Submission Mechanisms

We propose at § 414.1325(a) that individual MIPS eligible clinicians and groups would be required to submit data on measures and activities for the quality, CPIA and advancing care information performance categories. As proposed at § 414.1325(f), we do not propose any data submission requirements for the resource use performance category and for certain quality measures used to assess performance on the quality performance category and for certain activities in the CPIA performance category. For the resource use performance category, we propose that each individual MIPS eligible clinician's and group's resource use performance would be calculated using administrative claims data. As a result, individual MIPS eligible clinicians and groups would not be required to submit any additional information for the resource use performance category. In addition, we would be using administrative claims data to calculate performance on a subset of the MIPS quality measures and the CPIA performance category. For this subset of quality measures and CPIAs, MIPS eligible clinicians and groups would not be required to submit additional information. For individual clinicians and groups that are not MIPS eligible clinicians, such as physical therapists, but elect to report to MIPS, we would calculate administrative claims resource use measures and quality measures, if data is available. We are proposing multiple data submission mechanisms for MIPS as outlined in Tables 1 and 2 to provide MIPS eligible clinicians with flexibility to submit their MIPS measures and activities in a manner that best accommodates the characteristics of their practice. We note that other terms have been used for these submission mechanisms in earlier programs and in industry. As a result, the terms used for the submission mechanisms may be refined in the final rule for clarity.

We propose at § 414.1325(d) that MIPS eligible clinicians and groups may elect to submit information via multiple mechanisms; however, they must use the same identifier for all performance categories and they may only use one submission mechanism per category. For example, a MIPS eligible clinician could use one submission mechanism for sending quality measures and another for sending CPIA data, but a MIPS eligible clinician could not use two submission mechanisms for a single category such as submitting three quality measures via claims and three quality measures via registry. We believe the proposal to allow multiple mechanisms, while restricting the number of mechanisms per category, offers flexibility without adding undue complexity.

For individual MIPS eligible clinicians, we propose at § 414.1325(b), that an individual MIPS eligible clinician may choose to submit their quality, CPIA, and advancing care information data using qualified registry, QCDR, or EHR submission mechanisms. Furthermore, we propose at § 414.1400 that a qualified registry, health IT vendor, or QCDR could submit data on behalf of the MIPS eligible clinician for the three performance categories: Quality, CPIA, and advancing care information. As described in section II.E.9. of this proposed rule, these third party intermediaries would have to be qualified to submit for each of the performance categories. Additionally, we propose at § 414.1325(b)(4) and (5) that individual MIPS eligible clinicians may elect to report quality information via Medicare Part B claims and their CPIA and advancing care information performance category data through attestation.

For groups that are not reporting through the APM scoring standard, we propose at § 414.1325(c) that these groups may choose to submit their MIPS quality, CPIA, and advancing care information data using qualified registry, QCDR, EHR, or CMS Web Interface (for groups of 25+ MIPS eligible clinicians) submission mechanisms. Furthermore, we propose at § 414.1400 that a qualified registry, health IT vendor that obtains data from a MIPS eligible clinician's CEHRT, or QCDR could submit data on behalf of the group for the three performance categories: Quality, CPIA, and advancing care information. Additionally, groups may elect to submit their CPIA or advancing care information performance category data through attestation.

For those MIPS eligible clinicians participating in an APM that uses the APM scoring standard, we refer readers to section II.E.5.h. of this proposed rule, which describes how certain APM Entities submit data to MIPS, including separate approaches to the quality and resource use performance categories for APMs.

We propose one exception to the requirement for one reporting mechanism per category. Groups consisting of two or more eligible clinicians that elect to include CAHPS for MIPS as a quality measure must use a CMS-approved survey vendor. Their other quality information may be reported by any single one of the other proposed submission mechanisms.

While we allow MIPS eligible clinicians and groups to submit data for different performance categories via multiple submission mechanisms, we encourage MIPS eligible clinicians to submit MIPS information for the CPIA and advancing care information performance categories through the same reporting mechanism that is used for quality reporting. We believe it would reduce administrative burden and would simplify the data submission process for MIPS eligible clinicians by having a single reporting mechanism for all three performance categories for which MIPS eligible clinicians would be required to submit data: Quality, CPIA and advancing care information. However, we were concerned that not all third party entities would be able to implement the changes necessary to support reporting on all categories in the first year. We seek comments for future rulemaking on whether we should propose requiring health IT vendors, QCDRs and qualified registries to have the capability to submit data for all MIPS performance categories.

As we noted in this section of the proposed rule, we propose that MIPS eligible clinicians may report measures and activities using different submission methods across the performance categories. As we gain experience under MIPS, we anticipate that in future years it may be beneficial and reduce burden on MIPS eligible clinicians to require data for multiple performance categories to come through a single submission mechanism.

Further, we will be flexible in implementing MIPS. For example, if a MIPS eligible clinician submits data via multiple submission mechanisms (for example, registry and QCDR), we would score all the options and use the highest performance score for the eligible clinician or group as described in section II.E.6.a.(1)(b). However, we encourage eligible clinicians to report data for a given performance category using a single submission mechanism.

Finally, section 1848(q)(1)(E) of the Act requires the Secretary to encourage the use of QCDRs under section 1848(m)(3)(E) of the Act in carrying out MIPS. Section 1848(q)(5)(B)(ii)(I) of the Act requires the Secretary, under the CPS methodology, to encourage MIPS eligible clinicians to report on applicable measures with respect to the quality performance category through the use of CEHRT and QCDRs. We note that this proposed rule uses the term CEHRT and certified health IT in different contexts. For an explanation of these terms and contextual use within this proposed rule, we refer readers to section II.E.5.g. of this proposed rule.

We have multiple policies to encourage the usage of QCDRs and CEHRT. In part, we are promoting the use of CEHRT by awarding bonus points in the quality scoring section for measures gathered and reported electronically via the QCDR, qualified registry, Web Interface, or CEHRT submission mechanisms (see II.E.6.b). By promoting use of CEHRT through various submission mechanisms, we believe MIPS eligible clinicians have flexibility in implementing electronic measure reporting in a manner which best suits their practice.

To encourage the use of QCDRs, we have created opportunities for QCDRs to report new and innovative quality measures. In addition, several CPIAs emphasize QCDR participation. Finally, we allow for QCDRs to report data on all MIPS performance categories that require data submission and hope this will become a viable option for MIPS eligible clinicians. We believe these flexible options will allow MIPS eligible clinicians to more easily meet the submission criteria for MIPS, which in turn will positively affect their CPS.

We request comments on these proposals.

(3) Submission Deadlines

For the submission mechanisms described in section II.E.5.a.(2) of this proposed rule, we propose a submission deadline whereby all associated data for all performance categories must be submitted. In establishing the submission deadlines, we have taken into account multiple considerations, including the type of submission mechanism, the MIPS performance period, and stakeholder input and our experiences under the submission deadlines for the PQRS, VM, and Medicare EHR Incentive Programs.

Historically, under the PQRS, VM or Medicare EHR Incentive Programs, the submission of data occurred after the close of the performance periods. Our experience has shown that allowing for the submission of data after the close of the performance period provides either the eligible clinician or the third party intermediary time to ensure the data they submit to us is valid, accurate and has undergone necessary data quality checks. Stakeholders have also stated that they would appreciate the ability to submit data to us on a more frequent basis so they can receive feedback more frequently throughout the performance period. We also note that, as described in section II.E.4. of this proposed rule, the MIPS performance period for payments adjusted in 2019 is calendar year 2017 (January 1 through December 31).

Based on the factors noted, we propose at § 414.1325(e) the data submission deadline for the qualified registry, QCDR, EHR, and attestation submission mechanisms would be March 31 following the close of the performance period. We anticipate that the submission period would begin January 2 following the close of the performance period. For example, for the first MIPS performance period, the data submission period would occur from January 2, 2018, through March 31, 2018. We note that this submission period is the same time frame as what is currently available to eligible professionals and group practices under PQRS. We are interested in receiving feedback on whether it is advantageous to either (1) have a shorter time frame following the close of the performance period, or (2) have a submission period that would occur throughout the performance period, such as bi-annual or quarterly submissions; and (3) whether January 1 should also be included in the submission period. We welcome comments on these items.

We further propose that for the Medicare Part B claims submission mechanism, the submission deadline would occur during the performance period with claims required to be processed no later than 90 days following the close of the performance period. Lastly, for the CMS Web Interface submission mechanism, the submission deadline will occur during an eight-week period following the close of the performance period that will begin no earlier than January 1 and end no later than March 31. For example, the CMS Web Interface submission period could span an 8 week timeframe beginning January 16 and ending March 13. The specific deadline during this timeframe will be published on the CMS Web site.

We request comments on these proposals.

b. Quality Performance Category

(1) Background

(a) General Overview and Strategy

The MIPS program is one piece of the broader health care infrastructure needed to reform the health care system and improve health care quality, efficiency, and patient safety for all Americans. We seek to balance the sometimes competing considerations of the health system and minimize burdens on health care providers given the short timeframe available under the MACRA for implementation. Ultimately, MIPS should, in concert with other provisions of the Act, support health care that is patient-centered, evidence-based, prevention-oriented, outcome driven, efficient, and equitable.

Under MIPS, clinicians are incentivized to engage in improvement measures and activities that have a proven impact on patient health and safety and are relevant to their patient population. We envision a future state where MIPS eligible clinicians will be seamlessly using their certified health IT to leverage advanced clinical quality measurement to manage patient population with the least amount of workflow disruption and reporting burden. Ensuring clinicians are held accountable for patients' transitions across the continuum of care is imperative. For example, when a patient is discharged from an emergency department to a primary care physician office, the emergency department clinicians should have a shared incentive for a seamless transition. Clinicians may also be working with a QCDR to abstract and report quality measures to CMS and commercial payers and to track patients longitudinally over time for quality improvement.

Ideally, clinicians in the MIPS program will have accountability for quality and resource use measures that are related to one another and will be engaged in CPIAs that directly help them improve in both specialty-specific clinical practice and more holistic areas (for example, patient experience, prevention, population health). Finally, MIPS eligible clinicians will be using CEHRT and other tools which leverage interoperable standards for data capture, usage, and exchange in order to facilitate and enhance patient and family engagement, care coordination among diverse care team members, and, in continuous learning and rapid-cycle improvement leveraging advanced quality measurement and safety initiatives.

One of our goals in the MIPS program is to use a patient-centered approach to program development that will lead to better, smarter, and healthier care. Part of that goal includes meaningful measurement which we hope to achieve through:

• Measuring performance on measures that are relevant and meaningful.
• Maximizing the benefits of CEHRT.
• Flexible scoring that recognizes all of a MIPS eligible clinician's efforts above a minimum level of effort and rewards performance that goes above and beyond the norm.
• Measures that are built around real clinical workflows and data captured in the course of patient care activities.
• Measures and scoring that can discern meaningful differences in performance in each performance category and collectively between low and high performers.

(b) The MACRA Requirements

Sections 1848(q)(1)(A)(i) and (ii) of the Act require the Secretary to develop a methodology for assessing the total performance of each MIPS eligible clinician according to performance standards and, using that methodology, to provide for a CPS for each MIPS eligible clinician. Section 1848(q)(2)(A)(i) of the Act requires us to use the quality performance category in determining each MIPS eligible clinician's CPS, and section 1848(q)(2)(B)(i) of the Act describes the measures and activities that must be specified under the quality performance category.

The statute does not specify the number of quality measures on which a MIPS eligible clinician must report, nor does it specify the amount or type of information that a MIPS eligible clinician must report on each quality measure. However, section 1848(q)(2)(C)(i) of the Act requires the Secretary, as feasible, to emphasize the application of outcomes-based measures.

Sections 1848(q)(1)(E) of the Act requires the Secretary to encourage the use of QCDRs, and section 1848(q)(5)(B)(ii)(I) of the Act requires the Secretary to encourage the use of CEHRT and QCDRs for reporting measures under the quality performance category under the CPS methodology, but the statute does not limit the Secretary's discretion to establish other reporting mechanisms.

Section 1848(q)(2)(C)(iv) of the Act generally requires the Secretary to give consideration to the circumstances of non-patient-facing MIPS eligible clinicians and allows the Secretary, to the extent feasible and appropriate, to apply alternative measures or activities to such clinicians.

(c) Relationship to the PQRS and VM

Previously, the PQRS, which is a pay-for-reporting program, defined standards for satisfactory reporting and satisfactory participation to earn payment incentives or to avoid a payment adjustment EPs could choose from a number of reporting mechanisms and options. Based on the reporting option, the EP had to report on a certain number of measures for a certain portion of their patients. In addition, the measures had to span a set number of National Quality Strategy (NQS) domains, information related to the NQS can be found at http://www.ahrq.gov/workingforquality/about.htm . The VM built its policies off the PQRS criteria for avoiding the PQRS payment adjustment. Groups that did not meet the criteria as a group to avoid the PQRS payment adjustment or groups that did not have at least 50 percent of the EPs that did not meet the criteria as individuals to avoid the PQRS payment adjustment automatically received the maximum negative adjustment established under the VM and are not measured on their quality performance.

MIPS, in contrast to PQRS, is not a pay-for-reporting program, and we propose that it would not have a “satisfactory reporting” requirement. However, in order to develop an appropriate methodology for scoring the quality performance category, we believe that MIPS needs to define the expected data submission criteria and that the measures need to meet a data completeness standard. In this section we propose the minimum data submission criteria and data completeness standard for the MIPS quality performance category for the submission mechanisms that were proposed earlier in section II.E.5.a. The scoring methodology described in section II.E.6. of this proposed rule would adjust the quality performance category scores based on whether or not an individual MIPS eligible clinician or group met these criteria.

In the MIPS and APMs RFI, we requested feedback on numerous provisions related to data submission criteria including: How many measures should be required? Should we maintain the policy that measures cover a specified number of NQS domains? How do we apply the quality performance category to MIPS eligible clinicians that are in specialties that may not have enough measures to meet our defined criteria? Several themes emerged from the comments. Commenters expressed concern that the general PQRS satisfactory reporting requirement to report nine measures across three NQS domains is too high and forces eligible clinicians to report measures that are not relevant to their practices. The commenters requested a more meaningful set of requirements that focused on patient care, with some expressing the opinion that NQS domain requirements are arbitrary and make reporting more difficult. Some commenters asked that we align measures across payers and consider using core measure sets. Other commenters expressed the need for flexibility and different reporting options for different types of practices.

In response to the comments, and based on our desire to simplify the MIPS reporting system and make the measurement more meaningful, we are proposing MIPS quality criteria that focus on measures that are important to beneficiaries and maintain some of the flexibility from PQRS, while addressing several of the issues that concerned commenters.

• To encourage meaningful measurement, we are proposing to allow individual MIPS eligible clinicians and groups the flexibility to determine the most meaningful measures and reporting mechanisms for their practice.
• To simplify the reporting criteria, we are aligning the submission criteria for several of the reporting mechanisms.
• To reduce administrative burden and focus on measures that matter, we are lowering the expected number of the measures for several of the reporting mechanisms, yet are still requiring that certain types of measures be reported.
• To create alignment with other payers and reduce burden on MIPS eligible clinicians, we are incorporating measures that align with other national payers.
• To create a more comprehensive picture of the practice performance, we are also proposing to use all-payer data where possible.

As beneficiary health is always our top priority, we propose criteria to continue encouraging the reporting of certain measures such as outcome, appropriate use, patient safety, efficiency, care coordination, or patient experience measures. However, we are proposing to remove the requirement for measures to span across multiple domains of the NQS. We continue to believe the NQS domains to be extremely important and we encourage MIPS eligible clinicians to continue to strive to provide care that focuses on: Effective clinical care, communication, efficiency and cost reduction, person and caregiver-centered experience and outcomes, community and population health, and patient safety. While we will not require that a certain number of measures must span multiple domains, we strongly encourage MIPS eligible clinicians to select measures that cross multiple domains. In addition, we believe the MIPS program overall, with the focus on resource use, CPIAs, and advancing care information performance categories will naturally cover many elements in the NQS.

(2) Contribution to Composite Performance Score (CPS)

For the 2019 MIPS adjustment year, the quality performance category will account for 50 percent of the CPS, subject to the Secretary's authority to assign different scoring weights under section 1848(q)(5)(F) of the Act. Section 1848(q)(2)(E)(i)(I)(aa) of the Act states the quality performance category will account for 30 percent of the CPS for MIPS. However, section 1848(q)(2)(E)(i)(I)(bb) of the Act stipulates that for the first and second years for which MIPS applies to payments, the percentage of the CPS applicable for the quality performance category will be increased so that the total percentage points of the increase equals the total number of percentage points by which the percentage applied for the resource use performance category is less than 30 percent. Section 1848(q)(2)(E)(i)(II)(bb) of the Act requires that, for the first year for which MIPS applies to payments, not more than 10 percent of the of CPS shall be based on performance to the resource use performance category. Furthermore, section 1848(q)(2)(E)(i)(II)(bb) of the Act states that, for the second year for which MIPS applies to payments, not more than 15 percent of the CPS shall be based on performance to the resource use performance category. We propose at § 414.1330 for payment years 2019 and 2020, 50 percent and 45 percent, respectively, of the MIPS CPS will be based on performance on the quality performance category. For the third and future years, 30 percent of the MIPS CPS will be based on performance on the quality performance category.

Section 1848(q)(5)(B)(i) of the Act requires the Secretary to treat any MIPS eligible clinician who fails to report on a required measure or activity as achieving the lowest potential score applicable to the measure or activity. Specifically, under our proposed scoring policies, a MIPS eligible clinician or group that reports on all required measures and activities could potentially obtain the highest score possible within the performance category, presuming they performed well on the measures and activities they reported. A MIPS eligible clinician or group who does not meet the reporting threshold would receive a zero score for the unreported items in the category (in accordance with section 1848(q)(5)(B)(i) of the Act). The MIPS eligible clinician or group could still obtain a relatively good score by performing very well on the remaining items, but a zero score would prevent the MIPS eligible clinician or group from obtaining the highest possible score.

(3) Quality Data Submission Criteria

(a) Submission Criteria

The following are the proposed criteria for the various proposed MIPS data submission mechanisms described above in section II.E.5.a. of this proposed rule for the quality performance category.

(i) Submission Criteria for Quality Measures Excluding CMS Web Interface and CAHPS for MIPS

We propose at § 414.1335 that individual MIPS eligible clinicians submitting data via claims and individual MIPS eligible clinicians and groups submitting via all mechanisms (excluding CMS Web Interface, and for CAHPS for MIPS survey, CMS-approved survey vendors) would be required to meet the following submission criteria. We propose that for the applicable 12-month performance period, the MIPS eligible clinician or group would report at least six measures including one cross-cutting measure (if patient-facing) found in Table C and including at least one outcome measure. If an applicable outcome measure is not available, we propose that the MIPS eligible clinician or group would be required to report one other high priority measure (appropriate use, patient safety, efficiency, patient experience, and care coordination measures) in lieu of an outcome measure. If fewer than six measures apply to the individual MIPS eligible clinician or group, then we propose the MIPS eligible clinician or group would be required to report on each measure that is applicable.

MIPS eligible clinicians and groups will have to select their measures from either the list of all MIPS measures in Table A or a set of specialty-specific measure set in Table E. Note that some specialty-specific measure sets include measures grouped by subspecialty; in these cases, the measure set is defined at the subspecialty level.

We designed the specialty-specific measure sets to address feedback we have received in the past that the quality measure selection process can be confusing. A common complaint about PQRS was that EPs were asked to review close to 300 measures to find applicable measures for their specialty. The specialty measure sets in Table E are the same measures that are within Table A, however these are sorted consistent with the American Board of Medical Specialties (ABMS) specialties. Please note that these specialty-specific measure sets are not all inclusive of every specialty or subspecialty. We request comments on the measures proposed under each of the specialty-specific measure sets. Specifically, we seek comments on whether or not the measures proposed for inclusion in the specialty-specific measure sets are appropriate for the designated specialty or sub-specialty and whether there are additional proposed measures that should be included in a particular specialty-specific measure set.

Furthermore, we note that there are some special scenarios for those MIPS eligible clinicians who select their measures from a specialty-specific measure set at either the specialty or subspecialty level (Table E). For example, some of the specialty-specific measure sets have less than six measures, in these instances MIPS eligible clinicians would report on all of the available measures including an outcome measure or, if an outcome measure is unavailable, report another high priority measure (appropriate use, patient safety, efficiency, patient experience, and care coordination measures), within the set and a cross-cutting measure if they are a patient-facing MIPS eligible clinician. To illustrate, the subspecialty-level the electrophysiology cardiac specialist specialty-specific measure set only has three measures within the set, all of which are outcome measures. MIPS eligible clinicians and groups reporting on the electrophysiology cardiac specialist specialty-specific measure set would report on all three measures and since these MIPS eligible clinicians are patient-facing they must also report on a cross-cutting measure which is defined in Table C. In other scenarios, the specialty-specific measure sets may have six or more measures, in these instances MIPS eligible clinicians would report on at least six measures including at least one cross-cutting measure and at least one outcome measure or, if an outcome measure is unavailable, report another high priority measure (appropriate use, patient safety, efficiency, patient experience, and care coordination measure). Specifically, the general surgery specialty-specific measure set has eight measures within the set, including four outcome measures, three other high priority measures and one process measure. MIPS eligible clinicians and groups reporting on the general surgery specialty-specific measure set would either have the option to report on all measures within the set or could select six measures from the set and since these MIPS eligible clinicians are patient-facing one of their six measures must be a cross-cutting measure which is defined in Table C.

As noted above, the submission criteria for each specialty-specific measure set, or in the measure set defined at the subspecialty level, if applicable. Regardless of the number of measures that are contained in a specialty-specific measure set, MIPS eligible clinicians reporting on a measure set would be required to report at least one cross-cutting measure and either at least one outcome measure or, if no outcome measures are available in that specialty-specific measure set, report another high priority measure. MIPS eligible clinicians or groups that report on a specialty-specific measure set that includes more than six measures can report on as many measures as they wish as long as they meet the minimum requirement to report at least six measures, including one cross-cutting measure and one outcome measure, or if an outcome measure is not available another high priority measure. We seek comment on our proposal to allow reporting of specialty-specific measure sets to meet the submission criteria for the quality performance category, including whether it is appropriate to allow reporting of a measure set at the subspecialty level to meet such criteria, since reporting at the subspecialty level would require reporting on fewer measures. Alternatively, we seek comment on whether we should only consider reporting up to six measures at the higher overall specialty level to satisfy the submission criteria. We note that our proposal to allow reporting of specialty-specific measure sets at the subspecialty level was intended to address the fact that very specialized clinicians who may be represented by our subspecialty categories may only have one or two applicable measures. Further, we note that we will continue to work with specialty societies and other measure developers to increase the availability of applicable measures for specialists across the board.

We propose to define a high priority measure at § 414.1305 as an outcome, appropriate use, patient safety, efficiency, patient experience, or care coordination quality measures. These measures are identified in Table A. We further note that measure types listed as an “intermediate outcome” are considered outcome measures for the purposes of scoring; see section II.E.6.

As an alternative to the above proposals, we also considered requiring individual MIPS eligible clinicians submitting via claims and individual MIPS eligible clinicians and groups submitting via all mechanisms (excluding the CMS Web Interface and, for CAHPS for MIPS survey, CMS-approved survey vendors) to meet the following submission criteria. For the applicable 12-month performance period, the MIPS eligible clinician or group would report at least six measures including one cross-cutting measure (if patient-facing) found in Table C and one high priority measure (outcome, appropriate use, patient safety, efficiency, patient experience, and care coordination measures). If fewer than six measures apply to the individual MIPS eligible clinician or group, then the MIPS eligible clinician or group must report on each measure that is applicable. MIPS eligible clinicians and groups will have to select their measures from either the list of all MIPS Measures in Table A or a set of specialty-specific measure set in Table E.

As discussed in section II.E.1.b. of this proposed rule, MIPS eligible clinicians who are non-patient-facing MIPS eligible clinicians would not be required to report any cross-cutting measures.

We intend to develop a validation process to review and validate a MIPS eligible clinician's or group's ability to report on at least six quality measures, or a specialty-specific measure set, with a sufficient sample size, including at least one cross-cutting measure (if the MIPS eligible clinician is patient-facing) and either an outcome measure if one is available or another high priority measure. If a MIPS eligible clinician or group had the ability to report on the minimum required measures with sufficient sample size and elects to report on fewer than the minimum required measures, then, as described in the proposed scoring algorithm in section II.E.6., the missing measures would be scored with a zero performance score.

Our proposal is a decrease from the 2016 PQRS requirement to report at least nine measures. In addition, as previously noted, we propose to no longer require reporting across multiple NQS domains. We believe these proposals are the best approach for the quality performance category because it decreases the MIPS eligible clinician's reporting burden while focusing on more meaningful types of measures.

We also note that we believe that outcome measures are more valuable than clinical process measures and are instrumental to improving the quality of care patients receive. To keep the emphasis on such measures in the statute, we plan to increase the requirements for reporting outcome measures over the next several years through future rulemaking, as more outcome measures become available. For example, we may increase the required number of outcome measures to two or three. We also believe that appropriate use, patient experience, safety, and care coordination measures are more relevant than clinical process measures for improving care of patients. Through future rulemaking, we plan to increase the requirements for reporting on these types of measures over time.

In consideration of which MIPS measures to identify as reasonably focused on appropriate use, we have selected measures which focus on minimizing overuse of services, treatments, or the related ancillary testing that may promote overuse of services and treatments. We have also included select measures of underuse of specific treatments or services that either (1) reflected overuse of alternative treatments and services that were are not evidence-based or supported by clinical guidelines; or (2) where the intent of the measure reflected overuse of alternative treatments and services that were not evidence-based or supported by clinical guidelines. We realize there are differing opinions on what constitutes appropriate use. Therefore, we are seeking comments on what specific measures of over or under use should be included as appropriate use measures.

We plan to continue developing care episode groups, patient condition groups, and patient relationship categories (and codes for such groups and categories). We plan to incorporate new measures as they become available and will give the public the opportunity to comment on these provisions through future notice and comment rulemaking. We also will closely examine the recommendations from HHS' Office of the Assistant Secretary for Planning and Evaluation (ASPE) study, once they are available, on the issue of risk adjustment for socioeconomic status on quality measures and resource use as required by section 2(d) of the IMPACT Act and incorporate them as feasible and appropriate through future rulemaking. In addition, we are seeking comments on ways to minimize potential gaming, for example, requiring MIPS eligible clinicians to report only on measures for which they have a sufficient sample size, to address concerns that MIPS eligible clinicians may solely report on measures that do not have a sufficient sample size to decrease the overall weight on their quality score. More information on the way we propose to score MIPS eligible clinicians in this scenario is in section II.E.6.a.2. We also seek comment on whether these proposals sufficiently encourage providers and measure developers to move away from clinical process measures and towards outcome measures and measures that reflect other NQS domains. We request comments on these proposals.

(ii) Submission Criteria for Quality Measures for Groups Reporting via the CMS Web Interface

We propose at § 414.1335 the following criteria for the submission of data on quality measures by registered groups of 25 or more MIPS eligible clinicians who want to report via the CMS Web Interface. For the applicable 12-month performance period, we propose that the group would be required to report on all measures included in the CMS Web Interface completely, accurately, and timely by populating data fields for the first 248 consecutively ranked and assigned Medicare beneficiaries in the order in which they appear in the group's sample for each module/measure. If the pool of eligible assigned beneficiaries is less than 248, then the group would report on 100 percent of assigned beneficiaries. A group would be required to report on at least one measure for which there is Medicare patient data. We do not propose any modifications to this reporting process. Groups reporting via the CMS Web Interface are required to report on all of the measures in the set. Any measures not reported would be considered zero performance for that measure in our scoring algorithm.

Lastly, from our experience with using the CMS Web Interface under prior Medicare programs we are aware groups may register for this mechanism and have zero Medicare patients assigned and sampled to them. We clarify that should a group have no assigned patients, then the group, or individual MIPS eligible clinicians within the group, would need to select another mechanism to submit data to MIPS. If a group does not typically see Medicare patients for which the CMS Web Interface measures are applicable, or if the group does not have adequate billing history for Medicare patients to be used for assignment and sampling of Medicare patients into the CMS Web Interface, we advise the group to participate in the MIPS via another reporting mechanism.

As discussed in the CY 2016 PFS final rule with comment period (80 FR 71144), beginning with the 2017 PQRS payment adjustment, the PQRS aligned with the VM's beneficiary attribution methodology for purposes of assigning patients for groups that registered to participate in the PQRS Group Reporting Option (GPRO) using the CMS Web Interface (formerly referred to as the GPRO Web Interface). For certain quality and cost measures, the VM uses a two-step attribution process to associate beneficiaries with TINs during the period in which performance is assessed. This process attributes a beneficiary to the TIN that bills the plurality of primary care services for that beneficiary (79 FR 67960-67964). We propose to continue to align the 2019 CMS Web Interface beneficiary assignment methodology with the measures that used to be in the VM: the population quality measures discussed below in this proposed rule and total per capita cost for all attributed beneficiaries discussed in section II.E.5.e. of this proposed rule. As MIPS is a different program, we propose to modify the attribution process to update the definition of primary care services and to adapt the attribution to different identifiers used in MIPS. These changes are discussed in section II.E.5.e. of this proposed rule. We request comments on these proposals.

(iii) Performance Criteria for Quality Measures for Groups Electing To Report Consumer Assessment of Healthcare Providers and Systems (CAHPS) for MIPS Survey

The CAHPS for MIPS survey (formerly known as the CAHPS for PQRS survey) consists of the core CAHPS Clinician & Group Survey developed by AHRQ, plus additional survey questions to meet CMS's information and program needs. For more information on the CAHPS for MIPS survey, please see the explanation of the CAHPS for PQRS survey in the CY 2016 PFS final rule with comment period (80 FR 71142 through 71143). While we anticipate that the CAHPS for MIPS survey will closely align with the CAHPS for PQRS survey, we may explore the possibility of updating the CAHPS for MIPS survey under MIPS, specifically we may not finalize all proposed Summary Survey Measures (SSM).

We propose to allow registered groups of two or more MIPS eligible clinicians to voluntarily elect to participate in the CAHPS for MIPS survey. Specifically, we propose at § 414.1335 the following criteria for the submission of data on the CAHPS for MIPS survey by registered groups via CMS-approved survey vendor: For the applicable 12-month performance period, the group must have the CAHPS for MIPS survey reported on its behalf by a CMS-approved survey vendor. In addition, the group will need to use another submission mechanism (that is, qualified registries, QCDRs, EHR etc.) to complete their quality data submission. The CAHPS for MIPS survey would count as one cross-cutting and/or a patient experience measure, and the group would be required to submit at least five other measures through one other data submission mechanisms. A group may report any five measures within MIPS plus the CAHPS for MIPS survey to achieve the six measures threshold.

The administration of the CAHPS for MIPS survey would contain a six-month look-back period. In previous years the CAHPS for PQRS survey was administered from November to February of the reporting year. We propose to retain the same survey administration period for the CAHPS for MIPS survey. Groups that voluntarily elect to participate in the CAHPS for MIPS survey would bear the cost of contracting with a CMS-approved survey vendor to administer the CAHPS for MIPS survey on the group's behalf, just as groups do now for the CAHPS for PQRS survey.

Under current provisions of PQRS, the CAHPS for PQRS survey is required for groups of 100 or more eligible clinicians. Although we are not requiring groups to participate in the CAHPS for MIPS survey, we do still believe patient experience is important and we are therefore proposing a scoring incentive for those groups who report via the CAHPS for MIPS survey. As described in section II.E.3.d. of this proposed rule, we propose that groups electing to report the CAHPS for MIPS survey, would be required to register for the reporting of data. Because we believe patients' experiences as they interact with the health care system is important, our proposed scoring methodology would give bonus points for reporting CAHPS data (or other patient experience measures). Please refer to section II.E.6. for further details. We are interested in receiving comments on whether the CAHPS for MIPS survey should be required for groups of 100 or more MIPS eligible clinicians or whether it should be voluntary.

Currently, the CAHPS for PQRS beneficiary sample is based on Medicare claims data. Therefore, only Medicare beneficiaries can be selected to participate in the CAHPS for PQRS survey. In future years of the MIPS program, we may consider expanding the potential patient experience measures to all payers, so that Medicare and non-Medicare patients can be included in the CAHPS for MIPS survey sample. We are seeking comments on criteria that would ensure comparable samples. We seek comments on these proposals.

(b) Data Completeness Criteria

We want to ensure that data submitted on quality measures are complete enough to accurately assess each MIPS eligible clinician's quality performance. Section 1848(q)(5)(H) of the Act provides that analysis of the quality performance category may include quality measure data from other payers, specifically, data submitted by MIPS eligible clinicians with respect to items and services furnished to individuals who are not individuals entitled to benefits under Part A or enrolled under Part B of Medicare.

To ensure completeness for the broadest group of patients, we propose at § 414.1340 the criteria below. MIPS eligible clinicians and groups who do not meet the proposed reporting criteria noted below would fail the quality component of MIPS.

• Individual MIPS eligible clinicians or groups submitting data on quality measures using QCDRs, qualified registries, or via EHR need to report on at least 90 percent of the MIPS eligible clinician or group's patients that meet the measure's denominator criteria, regardless of payer for the performance period. In other words, for these submission mechanisms, we would expect to receive quality data for both Medicare and non-Medicare patients.
• Individual MIPS eligible clinicians submitting data on quality measures data using Medicare Part B claims, would report on at least 80 percent of the Medicare Part B patients seen during the performance period to which the measure applies.
• Groups submitting quality measures data using the CMS Web Interface or a CMS-approved survey vendor to report the CAHPS for MIPS survey would need to meet the data submission requirements on the sample of the Medicare Part B patients CMS provides.

We propose to include all-payer data for the QCDR, qualified registry, and EHR submission mechanisms because we believe this approach provides a more complete picture of each MIPS eligible clinicians scope of practice and provides more access to data about specialties and subspecialties not currently captured in PQRS. In addition, we propose the QCDR, qualified registry, or EHR submission must contain a minimum of one quality measure for at least one Medicare patient.

We desire all-payer data for all reporting mechanisms, yet certain reporting mechanisms are limited to Medicare Part B data. Specifically, the claims reporting mechanism relies on individual MIPS eligible clinicians attaching quality information on Medicare Part B claims; therefore only Medicare Part B patients can be reported by this mechanism. The CMS Web Interface and the CAHPS for MIPS survey currently rely on sampling protocols based on Medicare Part B billing; therefore, only Medicare Part B beneficiaries are sampled through that methodology. We welcome comments on ways to modify the methodology to assign and sample patients for these mechanisms using data from other payers.

The data completeness criteria we are proposing are an increase in the percentage of patients to be reported by each of the mechanisms when compared to PQRS. We believe the proposed thresholds are appropriate to ensure a more accurate assessment of a MIPS eligible clinician's performance on the quality measures and to avoid any selection bias that may exist under the current PQRS requirements. In addition, we would like to align all the reporting mechanisms as closely as possible with achievable data completeness criteria. We intend to continually assess the proposed data completeness criteria and will consider increasing these thresholds for future years of the program. We request comments on this proposal.

We are also interested in data that would indicate these data completeness criteria are inappropriate. For example, we could envision that reporting a cross-cutting measure would not always be appropriate for every telehealth service or for certain acute situations. We would not want a MIPS eligible clinician to fail reporting the measure in appropriate circumstances; therefore, we seek feedback data and circumstances where it would be appropriate to lower the data completeness criteria.

(c) Summary of Data Submission Criteria Proposals

Table 3 reflects our proposed Quality Data Submission Criteria for MIPS:

(4) Application of Quality Measures to Non-Patient-Facing MIPS Eligible Clinicians

Section 1848(q)(2)(C)(iv) of the Act provides that the Secretary must give consideration to the circumstances of non-patient-facing MIPS eligible clinicians and may, to the extent feasible and appropriate, take those circumstances into account and apply alternative measures or activities that fulfill the goals of the applicable performance category to such clinicians. In doing so, the Secretary must consult with non-patient-facing MIPS eligible clinicians.

In addition, section 1848(q)(5)(F) to the Act allows the Secretary to re-weight MIPS performance categories if there are not sufficient measures and activities applicable and available to each type of MIPS eligible clinician. We assume many non-patient-facing MIPS eligible clinician will not have sufficient measures and activities applicable and available to report and will not be scored on the quality performance category under MIPS. We refer readers to section II.E.6. of this proposed rule to discuss how we address performance categories weighting for MIPS eligible clinicians for whom no measures exist in a given category.

In the MIPS and APMs RFI, we solicited feedback on how we should apply the four MIPS performance categories to non-patient-facing MIPS eligible clinicians and what types of measures and/or CPIAs (new or from other payments systems) would be appropriate for these MIPS eligible clinicians. We also engaged with seven separate organizations representing non-patient-facing MIPS eligible clinicians in the areas of anesthesiology, radiology/imaging, pathology, and nuclear medicine, specifically cardiology. Organizations we spoke with representing several specialty areas indicated that Appropriate Use Criteria (AUC) can be incorporated into the CPIA performance category by including activities related to appropriate assessments and reducing unnecessary tests and procedures. AUC are distinct from clinical guidelines and specify when it is appropriate to use a diagnostic test or procedure—thus reducing unnecessary tests and procedures. Use of AUC is an important CPIA as it fosters appropriate utilization and is increasingly used to improve quality in cardiovascular medicine, radiology, imaging, and pathology. These groups also highlighted that many non-patient-facing MIPS eligible clinicians have multiple patient safety and practice assessment measures and activities that could be included, such as activities that are tied to their participation in the Maintenance of Certification (MOC) Part IV for improving the clinician's practice. One organization expressed concern that because their quality measures are specialized, some members could be negatively affected when comparing quality scores because they did not have the option to be compared on a broader, more common set of measures. The MIPS and APMs RFI commenters noted that the emphasis should be on measures and activities that are practical, attainable, and meaningful to individual circumstances and that measurement should be as outcomes-based to the extent possible. The MIPS and APMs RFI commenters emphasized that CPIAs should be selected from a very broad array of choices and that ideally non-patient-facing MIPS eligible clinicians should help develop those activities so that they provide value and are easy to document. For more details regarding the CPIA performance category refer to section II.E.5.f. of this proposed rule. The comments from these organizations were considered in developing these proposals.

We understand that non-patient-facing MIPS eligible clinicians may have a limited number of measures on which to report. Therefore, we propose at § 414.1335 that non-patient-facing MIPS eligible clinicians would be required to meet the otherwise applicable submission criteria, but would not be required to report a cross-cutting measure.

Thus we would employ the following strategy for the quality performance criteria to accommodate non-patient-facing MIPS eligible clinicians:

• Allow non-patient-facing MIPS eligible clinicians to report on specialty-specific measure set (which may have fewer than the required six measures).
• Allow non-patient-facing MIPS eligible clinicians to report through a QCDR that can report non-MIPS measures.
• Non-patient-facing MIPS eligible clinicians would be exempt from reporting a cross-cutting measure as proposed at § 414.1340.

We request comments on these proposals.

(5) Application of Additional System Measures

Section 1848(q)(2)(C)(ii) of the Act provides that the Secretary may use measures used for payment systems other than for physicians, such as measures used for inpatient hospitals, for purposes of the quality and resource use performance categories. The Secretary may not, however, use measures for hospital outpatient departments, except in the case of items and services furnished by emergency physicians, radiologists, and anesthesiologists.

In the MIPS and APMs RFI, we sought comment on how we could best use this authority. Some facility-based commenters requested a submission option that allows the MIPS eligible clinician to be scored based on the facility's measures. These commenters noted that the care they provide directly relates to and affects the facility's overall performance on quality measures and that using this score may be a more accurate reflection of the quality of care they provide than the quality measures in the PQRS or the VM program.

We will consider an option for facility-based MIPS eligible clinicians to elect to use their institution's performance rates as a proxy for the MIPS eligible clinician's quality score. We are not proposing an option for year 1 of MIPS because there are several operational considerations that must be addressed before this option can be implemented. We are requesting comment on the following issues: (1) Whether we should attribute a facility's performance to a MIPS eligible clinician for purposes of the quality and resource use performance categories and under what conditions such attribution would be appropriate and representative of the MIPS eligible clinician's performance; (2) possible criteria for attributing a facility's performance to a MIPS eligible clinician for purposes of the quality and resource use performance categories; and (3) the specific measures and settings for which we can use the facility's quality and resource use data as a proxy for the MIPS eligible clinician's quality and resource use performance categories; and (4) if attribution should be automatic or if a MIPS eligible clinician or group should elect for it to be done and choose the facilities through a registration process. We may also consider other options that would allow us to gain experience. We seek comments on these approaches.

(6) Global and Population-Based Measures

Section 1848(q)(2)(C)(iii) of the Act provides that the Secretary may use global measures, such as global outcome measures, and population-based measures for purposes of the quality performance category.

Under the current PQRS program and Medicare EHR Incentive Program quality measures are categorized by domains which include global and population-based measures. We identified population and community health measures as one of the quality domains related to the CMS Quality Strategy and the NQS priorities for health care quality improvement discussed in section II.E.5.c. of this proposed rule. Population-based measures are also used in the Medicare Shared Savings Program and for groups in the VM. For example, in 2015, clinicians were held accountable for a component of the Agency for Health Care Research (AHRQ) population-based, Ambulatory Care Sensitive Condition measures as part of a larger set of Prevention Quality Indicators (PQIs). Two broader composite measures of acute and chronic conditions are calculated using the respective individual measure rates for VM calculations. These PQIs assess the quality of the health care system as a whole, and especially the quality of ambulatory care, in preventing medical complications that lead to hospital admissions.

In the CY 2015 PFS final rule with comment period (79 FR 67909), Medicare Payment Advisory Commission (MedPAC) commented that we should move quality measurement for ACOs, Medicare Advantage (MA) plans, and FFS Medicare in the direction of a small set of population-based outcome measures, such as potentially preventable inpatient hospital admissions, emergency department visits, and readmissions. In the June 2014 MedPAC Report to the Congress: Medicare and the Health Care Delivery System MedPAC suggests considering an alternative quality measurement approach that would use population-based outcome measures to publicly report on quality of care across Medicare's three payment models, FFS, Medicare Advantage, and ACOs.

In creating policy for global and population-based measures for MIPS we considered a more broad-based approach to the use of “global” and “population-based” measures in the MIPS quality performance category. After considering the above we propose to use the acute and chronic composite measures of Agency for Healthcare Research and Quality (AHRQ) Prevention Quality Indicators (PQIs) that meet a minimum sample size in the calculation of the quality measure domain for the MIPS total performance score; see Table B. Eligible clinicians will be evaluated on their performance on these measures in addition to the six required quality measures discussed previously and summarized in Table A. Based on experience in the VM program, these measures have been determined to be reliable with a minimum case size of 20. Average reliabilities for the acute and chronic measures range from 0.64 to 0.79 for groups and individual MIPS eligible clinicians. We intend to incorporate a clinical risk adjustment as soon as feasible to the PQI composites and continue to research ways to develop and use other population-based measures for the MIPS program that could be applied to greater numbers of MIPS eligible clinicians going forward. In addition to the acute and chronic composite measure, we also propose to include the all-cause hospital readmissions measure from the VM as we believe this measure also encourages care coordination. In the CY 2016 Medicare PFS final rule (80 FR 71296), we did a reliability analysis that indicates this measure is not reliable for solo clinicians or practices with fewer than 10 clinicians; therefore, we propose to limit this measure to groups with 10 or more clinicians and to maintain the current VM requirement of 200 cases. Eligible clinicians in groups with 10 or more clinicians with sufficient cases will be evaluated on their performance on this measure in addition to the six required quality measures discussed previously and summarized in Table A.

Furthermore, the proposed claims-based population measures would rely on the same two-step attribution methodology that is currently used in the VM (79 FR 67961 through 67694). The attribution focuses on the delivery of primary care services (77 FR 69320) by both primary care physicians and specialists. This attribution logic aligns with the total per capita measure and is similar to, but not exactly the same, as the assignment methodology used for the Shared Savings Program. For example, the Shared Savings Program definition of primary care services can be found at § 425.20 and excludes claims for certain Skilled Nursing Facility (SNF) services that include the POS 31 modifier). In section II.E.5.e.3.a.i. of this proposed rule, we propose to exclude the POS 31 modifier from the definition of primary care services. As described in section II.E.2. of this proposed rule, the attribution would be modified slightly to account for the MIPS eligible clinician identifiers. We are seeking comments on additional measures or measure topics for future years of MIPS and attribution methodology. We request comments on these proposals.

c. Selection of Quality Measures for Individual MIPS Eligible Clinicians and Groups

(1) Annual List of Quality Measures Available for MIPS Assessment

Under section 1848(q)(2)(D)(i) of the Act, the Secretary, through notice and comment rulemaking, must establish an annual list of quality measures from which MIPS eligible clinicians may choose for purposes of assessment for a performance period. The annual list of quality measures must be published in the Federal Register no later than November 1 of the year prior to the first day of a performance period. Updates to the annual list of quality measures must be published in the Federal Register not later than November 1 of the year prior to the first day of each subsequent performance period. Updates may include the removal of quality measures, the addition of new quality measures, and the inclusion of existing quality measures that the Secretary determines have undergone substantive changes. For example, a quality measure may be considered for removal if the Secretary determines that the measure is no longer meaningful, such as measures that are topped out. A measure may be considered topped out if measure performance is so high and unvarying that meaningful distinctions and improvement in performance can no longer be made. Additionally, we are not the measure steward for most of the proposed quality measures available for inclusion in the MIPS annual list of quality measures. We rely on outside measure stewards and developers to maintain these measures. Therefore, we also propose to give consideration in removing measures that measure stewards are no longer able to maintain.

Under section 1848(q)(2)(D)(ii) of the Act, the Secretary must solicit a “Call for Quality Measures” each year. Specifically, the Secretary must request that eligible clinician organizations and other relevant stakeholders identify and submit quality measures to be considered for selection in the annual list of quality measures, as well as updates to the measures. Although we will accept quality measures submissions at any time, only measures submitted before June 1 of each year will be considered for inclusion in the annual list of quality measures for the performance period beginning 2 years after the measure is submitted. For example, a measure submitted prior to June 1, 2016 would be considered for the 2018 performance period. Of those quality measures submitted before June 1, we will determine which quality measures will move forward as potential measures for use in MIPS. Prior to finalizing new measures for inclusion in the MIPS program, those measures that we determine will move forward must also go through notice-and-comment rulemaking and the new proposed measures must be submitted to a peer review journal. Finally, for quality measures that have undergone substantive changes, we propose to identify measures including but not limited to measures that have had measure specification, measure title, and domain changes. Through NQF's or the measure steward's measure maintenance process, NQF-endorsed measures are sometimes updated to incorporate changes that we believe do not substantively change the intent of the measure. Examples of such changes may include updated diagnosis or procedure codes or changes to exclusions to the patient population or definitions. While we address such changes on a case-by-case basis, we generally believe these types of maintenance changes are distinct from substantive changes to measures that result in what are considered new or different measures.

In the first year of MIPS, we propose to maintain a majority of previously implemented measures in PQRS (80 FR 70885-71386) for inclusion in the annual list of quality measures. These measures can be found in the appendix at Table A: Proposed Individual Quality Measures Available for MIPS Reporting in 2017. Also included in the appendix in Table B is a list of quality measures that do not require data submission, some of which were previously implemented in the VM (80 FR 71273-71300), that we propose to include in the annual list of MIPS quality measures. These measures can be calculated from administrative claims data and do not require data submission. We are also proposing measures that were not previously finalized for implementation in the PQRS program. These measures and their draft specifications are listed in Table D. The proposed specialty-specific measure sets are listed in Table E. As we continue to develop measures and specialty-specific measure sets, we recognize that there are many MIPS eligible clinicians who see both Medicaid and Medicare patients and seek to align our measures to utilize Medicaid measures in the MIPS quality performance category. We believe that aligning Medicaid and Medicare measures is in the interest of all providers and will help drive quality improvement for our beneficiaries. For future years, we seek comment about the addition of a “Medicaid measure set” based on the CMCS Adult Core Set ( https://www.medicaid.gov/medicaid-chip-program-information/by-topics/quality-of-care/adult-health-care-quality-measures.html ). Measures we are proposing for removal can be found in Table F and measures that will have substantive changes for the 2017 performance period can be found in Table G. In future years, the annual list of quality measures available for MIPS assessment will occur through rulemaking. We request comment on these proposals. In particular, we seek comment on whether there are any measures that commenters believe should be classified in a different NQS domain than what was proposed or that should be classified as a different measure type (e.g., process vs. outcome) than what was proposed.

(2) Call for Quality Measures

Each year, we have historically solicited a “Call for Quality Measures” from the public for possible quality measures for consideration for the PQRS. Under MIPS, we propose to continue the annual “Call for Quality Measures” as a way to engage eligible clinician organizations and other relevant stakeholders in the identification and submission of quality measures for consideration. Under section 1848(q)(2)(D)(ii) of the Act, eligible clinician organizations are professional organizations as defined by nationally recognized specialty boards of certification or equivalent certification boards. However, we do not believe there needs to be any special restrictions on the type or make-up of the organizations carrying out the process of development of quality measures. Any such restriction would limit the development of quality measures and the scope and utility of the quality measures that may be considered for endorsement. Submission of potential quality measures regardless of whether they were previously published in a proposed rule or endorsed by an entity with a contract under section 1890(a) of the Act, which is currently the National Quality Forum, is encouraged.

As previously noted, we encourage the submission of potential quality measures regardless of whether such measures were previously published in a proposed rule or endorsed by an entity with a contract under section 1890(a) of the Act. However, consistent with the expectations established under PQRS, we propose to request that stakeholders apply the following considerations when submitting quality measures for possible inclusion in MIPS:

• Measures that are not duplicative of an existing or proposed measure.
• Measures that are beyond the measure concept phase of development and have started testing, at a minimum.
• Measures that include a data submission method beyond claims-based data submission.
• Measures that are outcome-based rather than clinical process measures.
• Measures that address patient safety and adverse events.
• Measures that identify appropriate use of diagnosis and therapeutics.
• Measures that address the domain for care coordination.
• Measures that address the domain for patient and caregiver experience.
• Measures that address efficiency, cost and resource use.
• Measures that address a performance gap or measurement gap.

We request comment on these proposals.

(3) Requirements

Section 1848(q)(2)(D)(iii) of the Act provides that, in selecting quality measures for inclusion in the annual final list of quality measures, the Secretary must provide that, to the extent practicable, all quality domains (as defined in section 1848(s)(1)(B) of the Act) are addressed by such measures and must ensure that the measures are selected consistent with the process for selection of measures under section 1848(k), (m), and (p)(2) of the Act.

Section 1848(s)(1)(B) of the Act defines “quality domains” as at least the following domains: clinical care, safety, care coordination, patient and caregiver experience, and population health and prevention. We believe the five domains applicable to the quality measures under MIPS are included in the NQS's six priorities as follows:

• Patient Safety. These are measures that reflect the safe delivery of clinical services in all health care settings. These measures may address a structure or process that is designed to reduce risk in the delivery of health care or measure the occurrence of an untoward outcome such as adverse events and complications of procedures or other interventions. We believe this NQS priority corresponds to the domain of safety.
• Person and Caregiver-Centered Experience and Outcomes. These are measures that reflect the potential to improve patient-centered care and the quality of care delivered to patients. They emphasize the importance of collecting patient-reported data and the ability to impact care at the individual patient level, as well as the population level. These are measures of organizational structures or processes that foster both the inclusion of persons and family members as active members of the health care team and collaborative partnerships with health care providers and provider organizations or can be measures of patient-reported experiences and outcomes that reflect greater involvement of patients and families in decision making, self-care, activation, and understanding of their health condition and its effective management. We believe this NQS priority corresponds to the domain of patient and caregiver experience.
• Communication and Care Coordination. These are measures that demonstrate appropriate and timely sharing of information and coordination of clinical and preventive services among health professionals in the care team and with patients, caregivers, and families to improve appropriate and timely patient and care team communication. They may also be measures that reflect outcomes of successful coordination of care. We believe this NQS priority corresponds to the domain of care coordination.
• Effective Clinical Care. These are measures that reflect clinical care processes closely linked to outcomes based on evidence and practice guidelines or measures of patient-centered outcomes of disease states. We believe this NQS priority corresponds to the domain of clinical care.
• Community/Population Health. These are measures that reflect the use of clinical and preventive services and achieve improvements in the health of the population served. They may be measures of processes focused on primary prevention of disease or general screening for early detection of disease unrelated to a current or prior condition. We believe this NQS priority corresponds to the domain of population health and prevention.
• Efficiency and Cost Reduction. These are measures that reflect efforts to lower costs and to significantly improve outcomes and reduce errors. These are measures of cost, resource use and appropriate use of health care resources or inefficiencies in health care delivery.

Section 1848(q)(2)(D)(viii) of the Act provides that the pre-rulemaking process under section 1890A of the Act is not required to apply to the selection of MIPS quality measures. Although not required to go through the pre-rulemaking process, we have found the NQF convened Measure Application Partnership's (MAP) input valuable. We propose that we may consider the MAP's recommendations as part of the comprehensive assessment of each measure considered for inclusion under MIPS. Elements we propose to consider in addition to those listed in the “Call for Quality Measures” section of this rule include a measure's fit within MIPS, if a measure fills clinical gaps, changes or updates to performance guidelines, and other program needs. Further, we will continue to explore how global and population-based measures can be expanded and plan to add additional population-based measures through future rulemaking. We request comment on these proposals.

(4) Peer Review

Section 1848(q)(2)(D)(iv) of the Act, requires the Secretary to submit new measures for publication in applicable specialty-appropriate, peer-reviewed journals before including such measures in the final annual list of quality measures. The submission must include the method for developing and selecting such measures, including clinical and other data supporting such measures. We believe this opportunity for peer review helps ensure that new measures published in the final rule are meaningful and comprehensive. We propose to use the Call for Quality Measures process as an opportunity to gather the information necessary to draft the journal articles for submission from measure developers, measure owners and measure stewards since CMS does not always develop measures for the quality programs. Information from measure developers, measure owners and measure stewards will include but is not limited to: Background, clinical evidence and data that supports the intent of the measure; recommendation for the measure that may come from a study or the United States Preventive Task Force (USPTF) recommendations; and how this measure would align with the CMS Quality Strategy. The Call for Quality Measures is a yearlong process; however, to be aligned with the regulatory process, establishing the proposed measure set for the year generally begins in April and concludes in July. We will submit new measures for publication in applicable specialty-appropriate, peer-reviewed journals before including such measures in the final annual list of quality measures. We request comment on this proposal. Additionally, we seek comment on mechanisms that could be used, such as the CMS Web site, to notify the public that the requirement to submit new measures for publication in applicable specialty-appropriate, peer-reviewed journals is met. Additionally, we seek comment on the type of information that should be included in such notification.

(5) Measures for Inclusion

Under section 1848(q)(2)(D)(v) of the Act, the final annual list of quality measures must include, as applicable, measures from under section 1848(k), (m), and (p)(2) of the Act, including quality measures among: (1) Measures endorsed by a consensus-based entity; (2) measures developed under section 1848(s) of the Act; and (3) measures submitted in response to the “Call for Quality Measures” required under section 1848(q)(2)(D)(ii) of the Act. Any measure selected for inclusion that is not endorsed by a consensus-based entity must have an evidence-based focus. Further, under section 1848(q)(2)(D)(ix), the process under section 1890A of the Act is considered optional.

Section 1848(s)(1) of the Act, as added by section 102 of the MACRA, also requires the Secretary of Health and Human Services to develop a draft plan for the development of quality measures by January 1, 2016. We solicited comments from the public on the “Draft CMS Measure Development Plan” through March 1, 2016. The final CMS Measure Development Plan must be finalized and posted on the CMS Web site by May 1, 2016.

(6) Exception for QCDR Measures

Section 1848(q)(2)(D)(vi) of the Act provides that quality measures used by a QCDR under section 1848(m)(3)(E) of the Act are not required to be established through notice-and-comment rulemaking or published in the Federal Register; be submitted for publication in applicable specialty-appropriate, peer-reviewed journals, or meet the criteria described in section 1848(q)(2)(D)(v) of the Act. The Secretary must publish the list of quality measures used by such QCDRs on the CMS Web site. We propose to post the quality measures for use by qualified clinical data registries in the spring of 2017 for the initial performance period and no later than January 1 for future performance periods.

Quality measures that are owned or developed by the QCDR entity and proposed by the QCDR for inclusion in MIPS but are not a part of the MIPS quality measure set are considered non-MIPS measures. If a QCDR wants to use a non-MIPS measure for inclusion in the MIPS program for reporting, we propose that these measures go through a rigorous CMS approval process during the QCDR self-nomination period. Specific details on third party entity requirements can be found in section II.E.9 of this proposed rule. The measure specifications will be reviewed and each measure will be analyzed for its scientific rigor, technical feasibility, duplication to current MIPS measures, clinical performance gaps, as evidenced by background and/or literature review, and relevance to specialty practice quality improvement. Once the measures are analyzed, the QCDR will be notified of which measures are approved for implementation. Each non-MIPS measure will be assigned a unique ID that can only be used by the QCDR that proposed it. Although non-MIPS measures are not required to be NQF-endorsed, we encourage the use of NQF-endorsed measures and measures that have been in use prior to implementation in MIPS. Lastly, we note that MIPS eligible clinicians reporting via QCDR have the option of reporting MIPS measures included in Table A in the Appendix to the extent that such measures are appropriate for the specific QCDR and have been approved by CMS. We request comment on these proposals.

(7) Exception for Existing Quality Measures

Section 1848(q)(2)(D)(vii)(II) of the Act provides that any quality measure specified by the Secretary under section 1848(k) or (m) of the Act and any measure of quality of care established under section 1848(p)(2) of the Act for a performance or reporting period beginning before the first MIPS performance period (herein referred to collectively as “existing quality measures”) must be included in the annual list of MIPS quality measures unless removed by the Secretary. As discussed in section II.E.4 of this proposed rule, we are proposing that the performance period for the 2019 MIPS adjustment would be CY 2017, that is, January 1, 2017 through December 31, 2017. Therefore existing quality measures would consist of those that have been specified or established by the Secretary as part of the PQRS measure set or VM measure set for a performance or reporting period beginning before CY 2017.

Section 1848(q)(2)(D)(vii)(I) of the Act provides that existing quality measures are not required to be established through notice-and-comment rulemaking or published in the Federal Register (although they remain subject to the applicable requirements for removing measures and including measures that have undergone substantive changes), nor are existing quality measures required to be submitted for publication in applicable specialty-appropriate, peer-reviewed journals.

(8) Consultation With Relevant Eligible Clinician Organizations and Other Relevant Stakeholders

Section 1890A of the Act, as added by section 3014(b) of the ACA, requires that the Secretary establish a pre-rulemaking process under which certain steps occur for the selection of certain categories of quality and efficiency measures, one of which is that the entity with a contract with the Secretary under section 1890(a) of the Act (that is, the NQF) convenes multi-stakeholder groups to provide input to the Secretary on the selection of such measures. These categories are described in section 1890(b)(7)(B) of the Act and include the quality measures selected for the PQRS. In accordance with section 1890A(a)(1) of the Act, the NQF convened multi-stakeholder groups by creating the MAP. Section 1890A(a)(2) of the Act requires that the Secretary make publicly available by December 1 of each year a list of the quality and efficiency measures that the Secretary is considering under Medicare. The NQF must provide the Secretary with the MAP's input on the selection of measures by February 1 of each year. The lists of measures under consideration for selection are available at http://www.qualityforum.org/map/ .

Section 1848(q)(2)(D)(viii) of the Act provides that relevant eligible clinician organizations and other relevant stakeholders, including state and national medical societies, must be consulted in carrying out the annual list of quality measures available for MIPS assessment. Section 1848(q)(2)(D)(ii)(II) of the Act defines an eligible clinician organization as a professional organization as defined by nationally recognized specialty boards of certification or equivalent certification boards. Section 1848(q)(2)(D)(viii) of the Act further provides that the pre-rulemaking process under section 1890A of the Act is not required to apply to the selection of MIPS quality measures.

Although MIPS quality measures are not required to go through the pre-rulemaking process under section 1890A of the Act, we have found the MAP's input valuable. The MAP process enables us to consult with relevant eligible professional organizations and other stakeholders, including state and national medical societies in finalizing the annual list of quality measures. In addition to the MAP's input this year, we also received input from the Core Measure Collaborative, specifically the America's Health Insurance Plans (AHIP), on core quality measure sets. The Core Measure Collaborative was organized by CMS in coordination with AHIP in 2014. This stakeholder workgroup has developed several condition-specific core measure sets to help align reporting requirements for private and public health insurance providers. Sixteen of the newly proposed measures under MIPS were recommended by the Core Measure Collaborative.

(9) Cross-Cutting Measures for 2017 and Beyond

Under PQRS we realized the value in requiring EPs to report a cross-cutting measure and have proposed to continue the use of cross-cutting measures under MIPS. The cross-cutting measures help focus our efforts on population health improvement and they also allow for meaningful comparisons between MIPS eligible clinicians. Under MIPS, we are proposing fewer cross-cutting measures than those available under PQRS for 2016 reporting; however, we believe the list contains measures for which all patient-facing MIPS eligible clinicians should be able to report, as the measures proposed include commonplace health improvement activities such as checking blood pressure and medication management. We have eliminated some measures for which the reporting MIPS eligible clinician may not actually be providing the care, but are just reporting another MIPS eligible clinician's performance result. An example of this would be a MIPS eligible clinician who never manages a diabetic patient's glucose, yet previously could have reported a measure about hemoglobin A1c based on an encounter. This type of reporting will likely not help improve or confirm the quality of care the MIPS eligible clinician provides to his or her patients. Although there are fewer proposed cross-cutting measures under MIPS, in previous years some measures were too specialized and could not be reported on by all MIPS eligible clinicians. The proposed cross-cutting measures under MIPS are more broadly applicable and can be reported on by most specialties. The proposed MIPS cross-cutting measure set will be available on the CMS Web site. Non-patient-facing MIPS eligible clinicians do not have a cross-cutting measure requirement. The cross-cutting measures that were available under PQRS for 2016 reporting that are not being proposed as cross-cutting measures for 2017 reporting are:

• PQRS #001 (Diabetes: Hemoglobin A1c Poor Control).
• PQRS #046 (Medication Reconciliation Post Discharge).
• PQRS #110 (Preventive Care and Screening: Influenza Immunization).
• PQRS #111 (Pneumonia Vaccination Status for Older Adults).
• PQRS #112 (Breast Cancer Screening).
• PQRS #131 (Pain Assessment and Follow-Up).
• PQRS #134 (Preventive Care and Screening: Screening for Clinical Depression and Follow-Up Plan).
• PQRS #154 (Falls: Risk Assessment).
• PQRS #155 (Falls: Plan of Care).
• PQRS #182 (Functional Outcome Assessment).
• PQRS #240 (Childhood Immunization Status).
• PQRS #318 (Falls: Screening for Fall Risk).
• PQRS #400 (One-Time Screening for Hepatitis C Virus (HCV) for Patients at Risk).

While we are proposing to remove the above listed measures from the cross-cutting measure set, these measures are being proposed to be available as individual quality measures available for MIPS reporting, some of which have proposed substantive changes. The proposed MIPS cross-cutting measure set can be found in Table C of the appendix of this proposed rule and will be available on the CMS Web site.

e. Resource Use Performance Category

(1) Background

(a) General Overview and Strategy

Measuring resource use is an integral part of measuring value. We envision the measures in the MIPS resource use performance category would provide MIPS eligible clinicians with the information they need to provide appropriate care to their patients and enhance health outcomes. In implementing the resource use performance category, we propose to start with existing condition and episode-based measures, and the total per capita costs for all attributed beneficiaries measure (total per capita cost measure). All resource use measures would be adjusted for geographic payment rate adjustments and beneficiary risk factors. In addition, a specialty adjustment would be applied to the total per capita cost measure. As detailed in section II.E.6.a.3 of this proposed rule, all of the measures attributed to a MIPS eligible clinician or group would be weighted equally within the resource use performance category, and there would be no minimum number of measures required to receive a score under the resource use performance category. We plan to draw on standards for measure reliability, patient attribution, risk adjustment, and payment standardization from the Physician Value-based Payment Modifier (Value Modifier or VM) as well as the Physician Feedback Program, as we believe many of the same measurement principles for cost measurement in the VM are applicable for measurement in the resource use performance category in MIPS.

All measures used under the resource use performance category would be derived from Medicare administrative claims data and as a result, participation would not require use of a data submission mechanism.

We plan to continue developing care episode groups, patient condition groups, and patient relationship categories (and codes for such groups and categories). We plan to incorporate new measures as they become available and will give the public the opportunity to comment on these provisions through future notice and comment rulemaking. We also will closely examine the recommendations from the HHS' Office of the Assistant Secretary for Planning and Evaluation (ASPE) study, when they are available, on the issue of risk adjustment for socioeconomic status on quality measures and resource use as required by section 2(d) of the IMPACT Act and incorporate them as feasible and appropriate through future rulemaking, under section 1848(q)(1)(G) of the Act.

(b) MACRA Requirements

Section 1848(q)(2)(A)(ii) of the Act establishes “resource use” as a performance category under the MIPS. Section 1848(q)(2)(B)(ii) of the Act describes the measures of the resource use performance category as the measurement of resource use for a MIPS performance period under section1848(p)(3) of the Act, using the methodology under section 1848(r) of the Act as appropriate, and, as feasible and applicable, accounting for the cost of drugs under Part D.

As discussed in section II.E.5.e.(1)(c) of this proposed rule, we previously established in rulemaking a value-based payment modifier, as required by section 1848(p) of the Act, that provides for differential payment to a physician or a group of physicians under the Physician Fee Schedule based on the quality of care furnished compared to cost. For the evaluation of costs of care, section 1848(p)(3) refers to appropriate measures of costs established by the Secretary that eliminate the effect of geographic adjustments in payment rates and take into account risk factors (such as socioeconomic and demographic characteristics, ethnicity, and health status of individuals, such as to recognize that less healthy individuals may require more intensive interventions) and other factors determined appropriate by the Secretary.

Section 1848(r) of the Act specifies a series of steps and activities for the Secretary to undertake to involve the physician, practitioner, and other stakeholder communities in enhancing the infrastructure for resource use measurement, including for purposes of MIPS and APMs. Section 1848(r)(2) of the Act requires the development of care episode and patient condition groups, and classification codes for such groups. That section provides for care episode and patient condition groups to account for a target of an estimated one-half of expenditures under Parts A and B (with this target increasing over time as appropriate). We are required to take into account several factors when establishing these groups. For care episode groups, we must consider the patient's clinical problems at the time items and services are furnished during an episode of care, such as clinical conditions or diagnoses, whether or not inpatient hospitalization occurs, the principal procedures or services furnished, and other factors determined appropriate by the Secretary. For patient condition groups, we must consider the patient's clinical history at the time of a medical visit, such as the patient's combination of chronic conditions, current health status, and recent significant history (such as hospitalization and major surgery during a previous period), and other factors determined appropriate. We are required to post on the CMS Web site a draft list of care episode and patient condition groups and codes for solicitation of input from stakeholders, and subsequently post on the Web site an operational list of such groups and codes. As required by section 1848(r)(2)(H) of the Act, not later than November 1 of each year (beginning with 2018), the Secretary shall, through rulemaking, revise the operational list as the Secretary determines may be appropriate.

To facilitate the attribution of patients and episodes to one or more clinicians, section 1848(r)(3) of the Act requires the development of patient relationship categories and codes that define and distinguish the relationship and responsibility of a physician or applicable practitioner with a patient at the time of furnishing an item or service. These categories shall include different relationships of the clinician to the patient and reflect various types of responsibility for and frequency of furnishing care. We are required to post on the CMS Web site a draft list of patient relationship categories and codes for solicitation of input from stakeholders, and subsequently post on the Web site an operational list of such categories and codes. As required by section 1848(r)(3)(F) of the Act, not later than November 1 of each year (beginning with 2018), the Secretary shall, through rulemaking, revise the operational list as the Secretary determines may be appropriate.

Section 1848(r)(4) of the Act requires that claims submitted for items and services furnished by a physician or applicable practitioner on or after January 1, 2018, shall, as determined appropriate by the Secretary, include the applicable codes established for care episode groups, patient condition groups, and patient relationship categories under sections 1848(r)(2) and (3) of the Act, as well as the NPI of the ordering physician or applicable practitioner (if different from the billing physician or applicable practitioner).

Under section 1848(r)(5) of the Act, to evaluate the resources used to treat patients, the Secretary shall, as determined appropriate, use the codes reported on claims under section 1848(r)(4) of the Act to attribute patients to one or more physicians and applicable practitioners and as a basis to compare similar patients, and conduct an analysis of resource use. In measuring such resource use, the Secretary shall use per patient total allowed charges for all services under Parts A and B (and, if the Secretary determines appropriate, Part D) and may use other measures of allowed charges and measures of utilization of items and services. The Secretary shall seek comments through one or more mechanisms (other than notice and comment rulemaking) from stakeholders regarding the resource use methodology established under section 1848(r)(5) of the Act.

On October 15, 2015, as required by section 1848(r)(2)(B) of the Act, we posted on the CMS Web site for public comment a list of the episode groups developed under section 1848(n)(9)(A) of the Act with a summary of the background and context to solicit stakeholder input as required by section 1848(r)(2)(C) of the Act. That posting is available at https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/Value-Based-Programs/MACRA-MIPS-and-APMs/MACRA-MIPS-and-APMs.html . The public comment period closed on February 15, 2016.

(c) Relationship to the Value Modifier

Currently, the physician value-based payment modifier established under section 1848(p) of the Act utilizes six cost measures (see 42 CFR 414.1235): (1) A total per capita costs for all-attributed beneficiaries measure (which we will refer to as the total per capita cost measure); (2) a total per capita costs for all attributed beneficiaries with chronic obstructive pulmonary disease (COPD) measure; (3) a total per capita costs for all attributed beneficiaries with congestive heart failure (CHF) measure; (4) a total per capita costs for all attributed beneficiaries with coronary artery disease (CAD) measure; (5) a total per capita costs for all attributed beneficiaries with diabetes mellitus (DM) measure; and (6) a Medicare Spending Per Beneficiary (MSPB) measure.

Total per capita costs include payments under both Part A and Part B, but do not include Medicare payments under Part D for drug expenses. All cost measures for the VM are attributed at the physician group and solo practice level using the Medicare-enrolled billing TIN under a two-step attribution methodology. They are risk-adjusted and payment-standardized, and the expected cost is adjusted for the TIN's specialty composition. We refer readers to our discussions of these total per capita cost measures (76 FR 73433 through 73434, 77 FR 69315 through 69316), MSPB measure (78 FR 74774 through 74780, 80 FR 71295 through 71296), payment standardization methodology (77 FR 69316 through 69317 ), risk adjustment methodology (77 FR 69317 through 69318), and specialty adjustment methodology (78 FR 74781 through 74784) in earlier rulemaking for the VM. More information about these total per capita cost measures may be found in documents under the links titled “Measure Information Form: Overall Total Per Capita Cost Measure,” “Measure Information Form: Condition-Specific Total Per Capita Cost Measures,” and “Measure Information Form: Medicare Spending Per Beneficiary Measure” available at https://www.cms.gov/medicare/medicare-fee-for-service-payment/physicianfeedbackprogram/valuebasedpaymentmodifier.html .

The total per capita cost measures use a two-step attribution methodology that is similar, but not exactly the same, as the assignment methodology used for the Shared Savings Program. The attribution focuses on the delivery of primary care services (77 FR 69320) by both primary care clinicians and specialists. The MSPB measure has a different attribution methodology. It is attributed to the TIN that provides the plurality of Medicare Part B claims (as measured by allowable charges) during the index inpatient hospitalization. We refer readers to the discussion of our attribution methodologies (77 FR 69318 through 69320, 79 FR 67960 through 67964) in prior rulemaking for the VM.

These total per capita cost measures include payments for a calendar year and have been reported to TINs for several years through the Quality and Resource Use Reports (QRURs), which are issued as part of the Physician Feedback Program under section 1848(n) of the Act. The total per capita cost measures have been used in the calculation of the VM payment adjustments beginning with the 2015 payment adjustment period and the MSPB measure has been used in the calculation of the VM payment adjustments beginning with the 2016 payment adjustment period. More information about the current attribution methodology for these measures is available in the “Fact Sheet for Attribution in the Value-Based Payment Modifier Program” document available at https://www.cms.gov/medicare/medicare-fee-for-service-payment/physicianfeedbackprogram/valuebasedpaymentmodifier.html .

In the MIPS and APMs RFI (80 FR 59102 through 59113), we solicited feedback on the resource use performance category. Commenters directed our attention towards the “2015 Value-Based Payment Modifier Program Experience Report” (document available at https://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/PhysicianFeedbackProgram/Downloads/2015-VM-Program-Experience-Rpt.pdf ) for data demonstrating that physicians treating the largest shares of the Medicare's sickest patients are most likely to incur downward adjustments under the current program. Commenters suggested that CMS could risk adjust cost measures for differences in beneficiary characteristics impacting health and cost outcomes, and suggested that cost measure benchmarks could be stratified so that groups and solo practitioners are compared to other groups and individual practitioners treating beneficiaries with similar risk profiles. Commenters also expressed concern that current attribution methods are holding many clinicians accountable for costs they have no control over, while other clinicians have no patients attributed and no way of calculating accurate scores. Commenters generally believe episode-based costs could provide a more accurate measure in calculating resource use and comparing clinicians based on the cost of patient treatment episodes. Many commenters agreed that if properly selected and designed, measures tied to episodes of care could increase the relevance, reliability, and applicability of resource use measures and make feedback reports more actionable. However, in order for clinicians to be responsible for resource use, including episode-based costs, commenters strongly emphasized the need for access to timely and actionable information regarding these costs. Commenters have expressed concern that because certain VM measures were developed for hospitals they are not properly applied to clinician practices, which do not have Medicare patient populations large enough or heterogeneous enough to produce an accurate picture for resource use. Commenters requested that CMS make an effort to use resource measures which have been tested for use in clinician practices. Commenters supported development of new measures based on clinical guidelines and/or appropriate use criteria (AUC), and support the related “Choosing Wisely” campaign. In future years, individual specialties might decide to use AUC or “Choosing Wisely” guidelines in the creation of resource use measures applicable to their members. In these cases, CMS could consider adoption of evidence-based measures developed through a multi-specialty, clinician-led process.

(2) Weighting in the Composite Performance Score

As required by section 1848(q)(5)(E)(i)(II)(bb) of the Act, the resource use performance category shall make up no more than 10 percent of the CPS for the first MIPS payment year (CY 2019) and not more than 15 percent of the CPS the second MIPS payment year (CY 2020). Therefore, we propose at § 414.1350 that the resource use performance category would make up 10 percent of the CPS for the first MIPS payment year (CY 2019) and 15 percent of the CPS for the second MIPS payment year (CY 2020). As required by section 1848(q)(5)(E)(i)(II)(aa) of the Act and proposed at § 414.1350, starting with the third MIPS payment year and for each MIPS payment year thereafter, the resource use performance category would make up 30 percent of the CPS.

(3) Resource Use Criteria

As discussed above in section II.E.5.a. of this proposed rule, performance in the resource use performance category would be assessed using measures based on administrative Medicare claims data. At this time, we are not proposing any additional data submissions for the resource use performance category. As such, MIPS eligible clinicians and groups would be assessed based on resource use for Medicare patients only and only for patients that are attributed to them. MIPS eligible clinicians or groups that do not have enough attributed cases to meet or exceed the case minimums proposed in sections II.E.5.e.(3)(a)(ii) and II.E.5.e.(3)(b)(ii) would not be measured on resource use. For more discussion of MIPS eligible clinicians and groups without a resource use performance category score, please refer to II.E.6.a.(3)(d) and II.E.6.b.

(a) Value Modifier Cost Measures Proposed for the MIPS Resource Use Performance Category

For purposes of assessing performance of MIPS eligible clinicians on the resource use performance category, we propose at § 414.1350 to specify resource use measures for a performance period. For the CY 2017 MIPS performance period, we propose to utilize the total per capita cost measure, the MSPB measure, and several episode-based measures discussed in section II.E.5.e.3.b. of this proposed rule for the resource use performance category. The total per capita costs measure and the MSPB measure are described above in section II.E.5.e.(1)(c) of this proposed rule.

We propose including the total per capita cost measure as it is a global measure of all Part A and Part B resource use during the performance period and inclusive of the four condition specific measures under the VM (chronic obstructive pulmonary disease, congestive heart failure, coronary artery disease, and diabetes mellitus) for which performance tends to be correlated and its inclusion was supported by commenters on the MIPS and APMs RFI (80 FR 59102 through 59113). We also anticipate that MIPS eligible clinicians are familiar with the total per capita cost measure as the measure has been in the VM since 2015 and feedback has been reported through the annual QRUR to all groups starting in 2014.

We propose to adopt the MSPB measure because by the beginning of the initial MIPS performance period in 2017, we believe most MIPS eligible clinicians will be familiar with the measure in the VM or its variant under the Hospital Value Based Purchasing program. However, we propose two technical changes to the MSPB measure calculations for purposes of its adoption in MIPS which are discussed in the reliability section II.E.5.e.3.a.ii. of this proposed rule.

We propose to use the same methodologies for payment standardization, and risk adjustment for these measures for the resource use performance category as are defined for the VM. For more details on the previously adopted payment standardization methodology see 77 FR 69316 through 69317. For more details on the previously adopted risk adjustment methodology see 77 FR 69317 through 69318.

We are not proposing to include the VM total per capita cost measures for the four condition-specific groups (chronic obstructive pulmonary disease, congestive heart failure, coronary artery disease, and diabetes mellitus). Instead, we are generally proposing to assess performance as part of the episode-based measures proposed under section II.E.5.e.3.b. of this proposed rule. This shift is in response to feedback received as part of the MIPS and APMs RFI (80 FR 59102 through 59113). In the MIPS and APMs RFI, commenters stated that they do not believe the existing condition-based measures under the VM are relevant to their practice and expressed support for episode-based measures under MIPS.

(i) Attribution

In the VM, all cost measures are attributed to a TIN. In MIPS, however, we are proposing to evaluate performance at the individual and group levels. Please refer to section II.E.5.e.(3)(c) of this proposed rule, for our proposals to address attribution differences for individuals and groups. For purposes of this section, we will use the general term MIPS eligible clinicians to indicate attribution for individuals or groups.

For the MSPB measure, we propose to use attribution logic that is similar to what is used in the VM. MIPS eligible clinicians with the plurality of claims (as measured by allowable charges) for Medicare Part B services, rendered during an inpatient hospitalization that is an index admission for the MSPB measure during the applicable performance period would be assigned the episode. The only difference from the VM attribution methodology would be that the MSPB measure would be assigned differently for individuals than for groups. For the total per capita cost measure, we propose to use a two-step attribution methodology that is similar to the methodology used in the 2017 and 2018 VM. We also propose to have the same two-step attribution process for the claims-based population measures in the quality performance category (section II.E.5.b.6.), CMS Web Interface measures, and CAHPS for MIPS. However, we also propose to make some modifications to the primary care services definition that is used in the attribution methodology to align with policies adopted under the Shared Savings Program.

The VM currently defines primary care services as the set of services identified by the following HCPCS/CPT codes: 99201 through 99215, 99304 through 99340, 99341 through 99350, the welcome to Medicare visit (G0402), and the annual wellness visits (G0438 and G0439). We propose to update this set to include new care coordination codes that have been implemented in the Medicare Physician Fee Schedule: Transitional care management (TCM) codes (CPT codes 99495 and 99496) and the chronic care management (CCM) code (CPT code 99490). These services were added to the primary care service definition used by the Shared Saving Program in June 2015 (80 FR 32746 through 32748). We believe that these care coordination codes would also be appropriate for assigning services in the MIPS.

In the CY 2016 PFS final rule, the Shared Saving Program also finalized another modification to the primary care service definition: To exclude nursing visits that occur in a skilled nursing facility (SNF) (80 FR 71271 through 71272). Patients in SNFs (POS 31) are generally shorter stay patients who are receiving continued acute medical care and rehabilitative services. While their care may be coordinated during their time in the SNF, they are then transitioned back to the community. Patients in a SNF (POS 31) require more frequent practitioner visits—often from 1 to 3 times a week. In contrast, patients in nursing facilities (NFs) (POS 32) are almost always permanent residents and generally receive their primary care services in the facility for the duration of their life. Patients in the NF (POS 32) are usually seen every 30 to 60 days unless medical necessity dictates otherwise. We believe that it would be appropriate to follow a similar policy in MIPS; therefore, we propose to exclude services billed under CPT codes 99304 through 99318 when the claim includes the POS 31 modifier from the definition of primary care services.

We believe that making these two modifications would help align the primary care service definition between MIPS and Shared Savings Program and would improve the results from the 2-step attribution process.

We note, however, that while we are aligning the definition for primary care services, the 2-step attribution for MIPS would be different than the one used for the Shared Saving Program. We believe there are valid reasons to have differences between MIPS and the Shared Savings Program attribution. For example, as discussed in CY 2015 PFS final rule (79 FR 67960 through 67962), we eliminated the primary care service pre-step that is statutorily required for the Shared Savings Program from the VM. We noted that without the pre-step, the beneficiary attribution method would more appropriately reflect the multiple ways in which primary care services are provided, which are not limited to physician groups. As MIPS eligible clinicians include more than physicians, we continue to believe it is appropriate to exclude the pre-step.

In addition, in the 2015 Shared Saving Program final rule, we finalized a policy for the Shared Savings Program that we did not extend to the VM 2-step attribution: to exclude select specialties (such as several surgical specialties) from the second attribution step (80 FR 32749 through 32754). We do not believe it is appropriate to restrict specialties from the second attribution step for MIPS. If such a policy were adopted under MIPS, then all specialists on the exclusion list, unless they were part of a multispecialty group, would automatically be excluded from measurement on the total per capita cost measure, as well as on the claims-based population measures which rely on the same 2-step attribution. While we do not believe that many MIPS eligible clinicians or clinician groups with these specialties would be attributed enough cases to meet or exceed the case minimum, we believe that an automatic exclusion could remove some MIPS eligible clinicians and groups that should be measured for resource use.

We request comments on these proposed changes.

(ii) Reliability

Additionally, we seek to ensure that MIPS eligible clinicians and groups are measured reliably; therefore, we intend to use the 0.4 reliability threshold currently applied to measures under the VM to evaluate their reliability. A 0.4 reliability threshold standard means that the majority of MIPS eligible clinicians and groups who meet the case minimum required for scoring under a measure have measure reliability scores that exceed 0.4. We generally consider reliability levels between 0.4 and 0.7 to indicate “moderate” reliability and levels above 0.7 to indicate “high” reliability. In cases where we have considered high participation in the applicable program to be an important programmatic objective, such as the Hospital VBP Program, we have selected this 0.4 moderate reliability standard. We believe this standard ensures moderate reliability but does not substantially limit participation.

To ensure sufficient measure reliability for the resource use performance category in MIPS, we also propose at § 414.1380(b)(2)(ii) to use the minimum of 20 cases for the total per capita cost measure, the same case minimum that is being used for the VM. An analysis in the CY 2016 PFS final rule (80 FR 71282) confirms that this measure has high average reliability for solo practitioners (0.74) as well as for groups with more than 10 professionals (0.80).

In the CY 2016 PFS final rule, we finalized a policy that increases the minimum cases for the MSPB measure from 20 to 125 cases (80 FR 71295 through 71296) due to reliability concerns with the measure including the specialty adjustment. That said, we recognize that a case size increase of this nature also may limit the ability of MIPS eligible clinicians to be scored on MSPB, and have been evaluating alternative measure calculation strategies for potential inclusion under MIPS that better balance participation, accuracy, and reliability. As a result of this, we are proposing two modifications to the MSPB measure.

The first technical change we are proposing is to remove the specialty-adjustment from the MSPB measure's calculation. As currently reported on the QRURs, the MSPB measure is risk adjusted to ensure that these comparisons account for case-mix differences between practitioners' patient populations and the national average. It is unclear that the current additional adjustment for physician specialty improves the accounting for case-mix differences for acute care patients, and thus, may not be needed.

The second technical change we propose is to modify the cost ratio used within the MSPB equation to evaluate the difference between observed and expected episode cost at the episode level before comparing the two at the individual or group level. In other words, rather than summing all of the observed costs and dividing by the sum of all the expected costs, we would take the observed to expected cost ratio for each MSPB episode assigned to the MIPS eligible clinician or group and take the average of the assigned ratios. As we did previously, we would take the average for the MIPS eligible clinician or group and multiply it by the average of observed costs across all episodes nationally.

Our analysis, which is based on all Medicare Part A and B claims data for beneficiaries discharged from an acute inpatient hospital between January 1, 2013 and December 1, 2013, indicates that these two changes would improve the MSPB measure's ability to calculate costs and the accuracy with which it can be used to make clinician-level performance comparisons. We also believe that these changes would help ensure the MSPB measure can be applied to a greater number of MIPS eligible clinicians while still maintaining its status as a reliable measure. More specifically, our analysis indicates that after making these changes to the MSPB measure's calculations, the MSPB measure meets the desired 0.4 reliability threshold used in the VM for over 88 percent of all TINs with a 20 case minimum, including solo practitioners. While this percentage is lower than our current policy for the VM (where virtually all TINs with 125 or more episodes have moderate reliability), setting the case minimum at 20 allows for an increase in participation in the MSPB measure. Therefore, we propose at § 414.1380(b)(2)(ii) to use a minimum of 20 cases for the MSPB measure. As noted previously, we consider expanded participation of MIPS eligible clinicians, particularly individual reporters, to be of great import for the purposes of transitioning to MIPS and believe that this justifies a slight decrease of the percentage of TINs meeting the reliability threshold.

We welcome public comment on these proposals.

(b) Episode-Based Measures Proposed for the MIPS Resource Use Performance Category

As noted in the previous section, we are proposing to calculate several episode-based measures for inclusion in the resource use performance category. Groups have received feedback on their performance on episode-based measures through the Supplemental Quality and Resource Use Report (sQRUR), which are issued as part of the Physician Feedback Program under section 1848(n) of the Act; however, these measures have not been used for payment adjustments through the VM. Several stakeholders expressed in the MIPS and APMs RFI the desire to transition to episode-based measures and away from the general total per capita measures used in the VM. Therefore, in lieu of using the total per capita cost measures for populations with specific conditions that are used for the VM, we are proposing episode-based measures for a variety of conditions and procedures that are high cost, have high variability in resource use, or are for high impact conditions. In addition, as these measures are payment standardized and risk adjusted, we believe they meet the statutory requirements for appropriate measures of cost as defined in section 1848(p)(3) of the Act because the methodology eliminates the effects of geographic adjustments in payment rates and takes into account risk factors.

We also reiterate that while we transition to using episode-based measures for payment adjustments, we will continue to engage stakeholders through the process specified in section 1848(r)(2) of the Act to refine and improve the episodes moving forward.

As noted earlier, we have provided performance information on episode-based measures to MIPS eligible clinicians through the Supplemental Quality and Resource Use Reports (sQRURs), which are released in the Fall. The sQRURs provide groups and solo practitioners with information to evaluate their resource utilization on conditions and procedures that are costly and prevalent in the Medicare FFS population. To accomplish this goal, various episodes are defined and attributed to one or more groups or solo practitioners most responsible for the patient's care. The episode-based measures include Medicare Part A and Part B payments for services determined to be related to the triggering condition or procedure. The payments included are standardized to remove the effect of differences in geographic adjustments in payment rates and incentive payment programs and they are risk adjusted for the clinical condition of beneficiaries. Although the sQRURs provide detailed information on these care episodes, the calculations are not used to determine a TIN's VM payment adjustment and are only used to provide feedback.

We propose to include in the resource use performance category several clinical condition and treatment episode-based measures that have been reported in the sQRUR or were included in the list of the episode groups developed under section 1848(n)(9)(A) of the Act published on the CMS Web site: https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/Value-Based-Programs/MACRA-MIPS-and-APMs/MACRA-MIPS-and-APMs.html . The identified episode-based measures have been tested and previously published. Tables 4 and 5 list the 41 clinical condition and treatment episode-based measures proposed for the CY 2017 MIPS performance period, as well as whether the episodes have previously been reported in a sQRUR.

The measures listed in Table 4 were developed under section 1848(n)(9)(A) of the Act, which required the Secretary to develop an episode grouper that combines separate but clinically related items and services into an episode of care for an individual, as appropriate, and provide reports on utilization to physicians (episode grouping Method A). The proposed measures accommodate both chronic and acute procedure episodes. The measures are also specifically designed to accommodate episodes that are initiated by physician claims, and section 1848(r)(4) of the Act requires claims submitted for items and services furnished by a physician or applicable practitioner on or after January 1, 2018, to include (as determined appropriate by the Secretary) the applicable codes established for care episode groups, patient condition groups, and patient relationship categories. The episodes and logic have undergone detailed and rigorous evaluation by an independent evaluation contractor and CMS also reviewed for clinical validity.

Attribution and reliability for the measures are discussed later in this section. Information about how the measures are constructed can be found at https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/Value-Based-Programs/MACRA-MIPS-and-APMs/MACRA-Feedback.html under the link for “Method A—Technical.” Detailed episode logic can be found under the “Method A” link on the same page.

Table 5 shows a second set of proposed measures that were developed to complement previous CMS efforts and to provide additional episode types to report in the supplemental QRURs. These measures represent acute conditions and procedures that are costly and prevalent in the Medicare FFS population. These measures examine services independently, regardless of other episodes a patient may be experiencing, and episodes do not interact with each other (episode grouping Method B).

Some of the episode types listed in Table 5 have subtypes that provide additional clinical detail and improve the actionability of data reported on these episode types, as well as comparability to expected costs. All episode types were developed with clinical input and complement the existing MSPB measure currently used in the VM. In addition, all episode types were reported in 2014 sQRURs.

Information about how the measures are constructed can be found at https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/Value-Based-Programs/MACRA-MIPS-and-APMs/MACRA-Feedback.html under the link for “Method B—Technical.” Detailed episode logic can be found under the “Method B” link on the same page.

While we are proposing the measures listed in Tables 4 and 5 for the resource use performance category, we are uncertain as to how many of these measures we will ultimately include in the final rule. As these measures have never been used for payment purposes, we may choose to specify a subset of these measures in the final rule. We request public comment on which of the measures listed in Tables 4 and 5 to include in the final rule. In addition to considering public comments, we intend to consider the number of MIPS eligible clinicians able to be measured, the episode's impact on Medicare Part A and Part B spending, and whether the measure has been reported through sQRUR. In addition, while we do not believe specialty adjustment is necessary for the episode-based measures, we will continue to explore this further given the diversity of episodes. We seek comment on whether we should specialty adjust the episode-based measures.

(i) Attribution

For the episode-based measures listed in Tables 4 and 5, we propose to use the attribution logic used in the 2014 sQRUR (full description available at https://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/PhysicianFeedbackProgram/Downloads/Detailed-Methods-2014SupplementalQRURs.pdf ), with modifications to adjust for whether performance is being assessed at an individual level or group level. Please refer to section II.E.5.e.(3)(c) of this proposed rule for our proposals to address attribution differences for individuals and groups. For purposes of this section, we will use the general term MIPS eligible clinicians to indicate attribution for individuals or groups.

Acute condition episodes would be attributed to all MIPS eligible clinicians that bill at least 30 percent of inpatient evaluation and management (IP E&M) visits during the initial treatment, or “trigger event,” that opened the episode. E&M visits during the episode's trigger event represent services directly related to the management of the beneficiary's acute condition episode. MIPS eligible clinicians that bill at least 30 percent of IP E&M visits are therefore likely to have been responsible for the oversight of care for the beneficiary during the episode. It is possible for more than one MIPS eligible clinician to be attributed a single episode using this rule. If an acute condition episode has no IP E&M claims during the episode, then that episode is not attributed to any MIPS eligible clinician.

Procedural episodes would be attributed to all MIPS eligible clinicians that bill a Medicare Part B claim with a trigger code during the trigger event of the episode. For inpatient procedural episodes, the trigger event is defined as the IP stay that triggered the episode plus the day before the admission to the IP hospital. For outpatient procedural episodes constructed using Method A, the trigger event is defined as the day of the triggering claim plus the day before and two days after the trigger date. For outpatient procedural episodes constructed using Method B, the trigger event is defined as only the day of the triggering claim. Any Medicare Part B claim or line during the trigger event with the episode's triggering procedure code is used for attribution. If more than one MIPS eligible clinician bills a triggering claim during the trigger event, the episode is attributed to each of the MIPS eligible clinicians. If co-surgeons bill the triggering claim, the episode is attributed to each MIPS eligible clinician. If only an assistant surgeon bills the triggering claim, the episode is attributed to the assistant surgeon or group. If an episode does not have a concurrent Part B claim with a trigger code for the episode, then that episode is not attributed to any MIPS eligible clinician.

(ii) Reliability

To ensure moderate reliability, we propose at § 414.1380(b)(2)(ii) to use the minimum of 20 cases for all episode-based measures listed in Tables 4 and 5. We propose to not include any measures that do not have average moderate reliability (at least 0.4) at 20 episodes.

(c) Attribution for Individual and Groups

In the VM and sQRUR, all resource use measurement was attributed at the solo practitioner and group level, as identified by TIN. In MIPS, however, we are proposing to evaluate performance at the individual and group levels. For MIPS eligible clinicians whose performance is being assessed individually across the other MIPS performance categories, we propose to attribute resource use measures using TIN/NPI rather than TIN. Attribution at the TIN/NPI level allows individual MIPS eligible clinicians, as identified by their TIN/NPI, to be measured based on cases that are specific to their practice, rather than being measured on all the cases attributed to the group TIN. For MIPS eligible clinicians that choose to have their performance assessed as a group across the other MIPS performance categories, we propose to attribute resource use measures at the TIN level (the group TIN under which they report). The logic for attribution would be similar whether attributing to the TIN/NPI level or the TIN level. As an alternative proposal, we seek comment on whether MIPS eligible clinicians that choose to have their performance assessed as a group should first be attributed at the individual TIN/NPI level and then have all cases assigned to the individual TIN/NPIs attributed to the group under which they bill. This alternative would apply one consistent methodology to both groups and individuals, compared to having a methodology that assigns cases using TIN/NPI for assessment at the individual level and another that assigns cases using only TIN for assessment at the group level. For example, the general attribution logic for MSPB is to assign the MSPB measure based on the plurality of claims (as measured by allowable charges) for Medicare Part B services rendered during an inpatient hospitalization that is an index admission for the MSPB measure. Our proposed approach would determine “plurality of claims” separately for individuals and groups. For individuals, we would assign the MSPB measure using the “plurality of claims” by TIN/NPI, but for groups we would determine the “plurality of claims” by TIN. The alternative proposal, in contrast, would determine the “plurality of claims” by TIN/NPI for both groups and individuals. However, for individuals, only the MSPB measure attributed to the TIN/NPI would be evaluated, while for groups the MSPB measure attributed to any TIN/NPI billing under the TIN would be evaluated.

We request comment on this proposal and alternative considered.

(d) Application of Measures to Non-Patient Facing MIPS Eligible Clinicians

Section 101(c) of the MACRA added section 1848(q)(2)(C)(iv) to the Act, which requires the Secretary to give consideration to the circumstances of professional types who typically furnish services without patient facing interaction (non-patient-facing) when determining the application of measures and activities. In addition, this section allows the Secretary to apply alternative measures or activities to non-patient facing MIPS eligible clinicians that fulfill the goals of a performance category. Section 101(c) of the MACRA also added section 1848(q)(5)(F) to the Act, which allows the Secretary to re-weight MIPS performance categories if there are not sufficient measures and activities applicable and available to each type of eligible clinician involved.

For the 2017 MIPS performance period, we are not proposing any alternative measures for non-patient facing MIPS eligible clinicians or groups. This means that non-patient facing MIPS eligible clinicians or groups may not be attributed any resource use measures that are generally attributed to clinicians who have patient facing encounters with patients. We therefore anticipate that, similar to MIPS eligible clinicians or groups that do not meet the required case minimum for any resource use measures, many non-patient facing MIPS eligible clinicians may not have sufficient measures and activities available to report and would not be scored on the resource use performance category under MIPS. We refer readers to section II.E.6.b.2. of this proposed rule where we discuss how we would address performance category weighting for MIPS eligible clinicians or groups who do not receive a performance category score for a given performance category. We also intend to work with non-patient facing MIPS eligible clinicians and specialty societies to propose alternative resource use measures for non-patient facing MIPS eligible clinicians and groups under MIPS in future years. Lastly, we seek comment on how best to incorporate appropriate alternative resource use measures for all MIPS eligible clinician types, including non-patient facing MIPS eligible clinicians.

(e) Additional System Measures

Section 1848(q)(2)(C)(ii) of the Act, as added by section 101(c) of MACRA provides that the Secretary may use measures used for a payment system other than for physicians, such as measures for inpatient hospitals, for purposes of the quality and resource use performance categories of MIPS. The Secretary, however, may not use measures for hospital outpatient departments, except in the case of items and services furnished by emergency physicians, radiologists, and anesthesiologists.

We intend to align any facility-based MIPS measure decision across the quality and resource use performance categories to ensure consistent policies for MIPS in future years. We refer readers back to section II.E.5.b.5. of this proposed rule, which discusses our strategy and solicits comments related to this provision.

(4) Future Modifications to Resource Use Performance Category

In the future, we intend to consider how best to incorporate Part D costs into the resource use performance category, as described in section 1848(q)(2)(B)(ii) of the Act. We seek public comments on how we should incorporate those costs under MIPS for future years. We also intend to continue developing and refining episode groups for purposes of resource use performance category measure calculations.

f. Clinical Practice Improvement Activity (CPIA) Category

(1) Background

(a) General Overview and Strategy

The CPIA performance category focuses on one of our MIPS strategic goals, to use a patient-centered approach to program development that leads to better, smarter, and healthier care. We believe improving the health of all Americans can be accomplished by developing incentives and policies that drive improved patient health outcomes. CPIAs emphasize activities that have a proven association with improved health outcomes. The CPIA performance category also focuses on another MIPS strategic goal which is to use design incentives that drive movement toward delivery system reform principles and APMs. Another MIPS strategic goal we are striving to achieve is to establish policies that can be scaled in future years as the bar for improvement rises. Under the CPIA performance category we are proposing baseline requirements that will continue to have more stringent requirements in future years, and lay the groundwork for expansion towards continuous improvement over time.

(b) The MACRA Requirements

Section 1848(q)(2)(C)(v)(III) of the Act defines a CPIA as an activity that relevant eligible clinician organizations and other relevant stakeholders identify as improving clinical practice or care delivery, and that the Secretary determines, when effectively executed, is likely to result in improved outcomes. Section 1848(q)(2)(B)(iii) of the Act requires the Secretary to specify CPIAs under subcategories for the performance period, which must include at least the subcategories specified in section 1848(q)(2)(B)(iii)(I) through (VI) of the Act, and in doing so to give consideration to the circumstances of small practices (consisting of 15 or fewer clinicians), and practices located in rural areas and geographic health professional shortage areas (HPSAs).

Section 1848(q)(2)(C)(iv) of the Act generally requires the Secretary to give consideration to the circumstances of non-patient-facing MIPS eligible clinicians or groups and allows the Secretary, to the extent feasible and appropriate, to apply alternative measures and activities to such MIPS eligible clinicians and groups.

Section 1848(q)(2)(C)(v) of the Act required the Secretary to use a request for information (RFI) to solicit recommendations from stakeholders to identify CPIAs and specify criteria for such CPIAs, and provides that the Secretary may contract with entities to assist in identifying activities, specifying criteria for the activities, and determining whether MIPS eligible clinicians or groups meet the criteria set. In the MIPS and APMs RFI, we requested recommendations to identify activities and specify criteria for activities. In addition, we requested details on how data should be submitted, the number of activities, how performance should be measured, and what considerations should be made for small and/or rural practices. There were two overarching themes from the comments that we received. First, the majority of the comments indicated that all subcategories should be weighted equally and that MIPS eligible clinicians or groups should be allowed to select from whichever subcategories are most applicable to them during the performance period. Second, commenters supported inclusion of a diverse set of activities that are meaningful for individual MIPS eligible clinicians or groups. We have reviewed all of the comments that we received and have taken these recommendations into consideration while developing the proposed CPIA policies.

(2) Contribution to Composite Performance Score (CPS)

Section 1848(q)(5)(E)(i)(III) of the Act specifies that the CPIA performance category will account for 15 percent of the CPS, subject to the Secretary's authority to assign different scoring weights under section 1848(q)(5)(F) of the Act. Therefore, we propose at § 414.1355, that the CPIA performance category will account for 15 percent of the CPS.

Section 1848(q)(5)(C)(i) of the Act specifies that a MIPS eligible clinician or group that is certified as a patient-centered medical home or comparable specialty practice, as determined by the Secretary, with respect to a performance period must be given the highest potential score for the CPIA performance category for the performance period. For a further description of APMs that have a certified patient centered-medical home designation, we refer readers to section II.E.5.h.

A patient-centered medical home will be recognized if it is a nationally recognized accredited patient-centered medical home, a Medicaid Medical Home Model, or a Medical Home Model. The NCQA Patient-Centered Specialty Recognition will also be recognized, which qualifies as a comparable specialty practice. Nationally recognized accredited patient-centered medical homes are recognized if they are accredited by: (1) The Accreditation Association for Ambulatory Health Care; (2) the National Committee for Quality Assurance (NCQA) PCMH recognition; (3) The Joint Commission Designation; or (4) the Utilization Review Accreditation Commission (URAC). We refer readers to section II.F. of this proposed rule for further description of the Medicaid Medical Home Model or Medical Home Model.⁹ The criteria for being a nationally recognized accredited patient-centered medical home is that it must be national in scope and must have evidence of being used by a large number of medical organizations as the model for their patient-centered medical home. We seek comment on our proposal for determining which practices would qualify as patient-centered medical homes. We also note that practices may receive a patient-centered medical home designation at a practice level, and that individual TINs may be composed of both undesignated practices and practices that have received a designation as a patient-centered medical home (for example, only one practice site has received patient-centered medical home designation in a TIN that includes five practice sites). For MIPS eligible clinicians who choose to report at the group level, reporting is required at the TIN level. We solicit comment on how to provide credit for patient-centered medical home designations in the calculation of the CPIA performance category score for groups when the designation only applies to a portion of the TIN (for example, to only one practice site in a TIN that is comprised of five practice sites).

Section 1848(q)(5)(C)(ii) of the Act provides that MIPS eligible clinicians or groups who are participating in an APM (as defined in section 1833(z)(3)(C) of the Act) for a performance period must earn at least one half of the highest potential score for the CPIA performance category for the performance period. For further description of CPIA and the APM scoring standard for MIPS, we refer readers to section II.E.5.h. For all other MIPS eligible clinicians or groups, this section applies and we also refer readers to the scoring requirements for MIPS eligible clinicians and groups in section II.E.6. of this proposed rule.

Section 1848(q)(5)(C)(iii) of the Act provides that a MIPS eligible clinician or group must not be a MIPS eligible clinician or group required to perform activities in each CPIA subcategory or participate in an APM to achieve the highest potential score for the CPIA performance category.

Section 1848(q)(5)(B)(i) of the Act requires the Secretary to treat a MIPS eligible clinician or group that fails to report on an applicable measure or activity that is required to be reported, they will receive the lowest potential score applicable to the measure or activity.

(3) CPIA Data Submission Criteria

(a) Submission Mechanisms

For the purpose of submitting under the CPIA performance category, we proposed in section II.E.5.a. of this proposed rule to allow for submission of data for the CPIA performance category using the qualified registry, EHR, QCDR, CMS Web Interface and attestation data submission mechanisms. If technically feasible, we will use administrative claims data to supplement the CPIA submission. Regardless of the data submission method, all MIPS eligible clinicians or groups must select activities from the CPIA Inventory provided in Table H of the Appendices. We believe the proposed data submission methods will allow for greater access and ease in submitting data, as well as consistency throughout the MIPS program.

In addition, we propose at § 414.1360, that for the first year only, all MIPS eligible clinicians or groups, or third party entities such as health IT vendors, QCDRs and qualified registries that submit on behalf of a MIPS eligible clinician or group, must designate a yes/no response for activities on the CPIA Inventory. In the case where a MIPS eligible clinician or group is using a health IT vendor, QCDR, or qualified registry for their data submission, the MIPS eligible clinician or group will certify all CPIAs have been performed and the health IT vendor, QCDR, or qualified registry will submit on their behalf. An agreement between a MIPS eligible clinician or group and a health IT vendor, QCDR, or qualified registry for data submission for CPIA as well as other performance data submitted outside of the CPIA performance category could be contained in a single agreement, minimizing the burden on the MIPS eligible clinician or group. See section II.E.9 for additional details.

We propose to use the administrative claims method, if technically feasible, only to supplement CPIA submissions. For example, if technically feasible, MIPS eligible clinicians or groups, using the telehealth modifier GT, could get automatic credit for this activity. We request comments on these proposals.

(b) Weighted Scoring

While we considered both equal and differentially weighted scoring in this performance category, the statute requires a differentially weighted scoring model by requiring 100 percent of the potential score in the CPIA performance category for patient-centered medical home participants, and a minimum 50 percent score for APM participants. For additional activities in this category, we propose at § 414.1380 a differentially weighted model for the CPIA performance category with two categories: Medium and high. The justification for these two weights is to provide flexible scoring due to the undefined nature of activities (that is, CPIA standards are not nationally recognized and there is no entity for CPIA that serves the same function as the National Quality Forum does for quality measures). CPIAs are weighted as high based on alignment with CMS national priorities and programs such as the Quality Innovation Network-Quality Improvement Organization (QIN/QIO) or the Comprehensive Primary Care Initiative which recognizes specific activities related to expanded access and integrated behavioral health as important. Programs that require performance of multiple activities such as participation in the Transforming Clinical Practice Initiative, seeing new and follow-up Medicaid patients in a timely manner in the provider's State Medicaid Program, or an activity identified as a public health priority (such as emphasis on anticoagulation management or utilization of prescription drug monitoring programs) were weighted as high.

The statute references patient-centered medical homes as achieving the highest score for the MIPS program. MIPS eligible clinicians or groups may use that to guide them in the criteria or factors that should be taken into consideration to determine whether to weight an activity medium or high on comments for this proposal. We request comments on this proposal, including criteria or factors we should take into consideration to determine whether to weight an activity medium or high.

(c) Submission Criteria

We propose at § 414.1380 to set the CPIA submission criteria under MIPS, in order to achieve the highest potential score of 100 percent, at three high-weighted CPIAs (20 points each) or six medium-weighted CPIAs (10 points each), or some combination of high and medium-weighted CPIAs to achieve a total of 60 points for MIPS eligible clinicians participating as individuals or as groups (refer to Table H of the Appendices for CPIAs and weights). MIPS eligible clinicians or groups that select less than the designated number of CPIAs will receive partial credit based on the weighting of the CPIA selected. To achieve a 50 percent score, one high-weighted and one medium-weighted CPIA or three medium-weighted CPIAs are required for these MIPS eligible clinicians or groups.

Exceptions to the above apply for: MIPS small groups (consisting of 15 or fewer clinicians), MIPS eligible clinicians and groups located in rural areas, MIPS eligible clinicians and groups that are located in geographic HPSAs, non-patient-facing MIPS eligible clinicians or groups or MIPS eligible clinicians, or groups that participate in an APM and/or a patient-centered medical home submitting in MIPS.

For MIPS eligible clinicians and groups that are small, located in rural areas or geographic HPSAs, or non-patient-facing MIPS eligible clinicians or groups, in order to achieve the highest score of 100 percent, two CPIAs are required (either medium or high). For MIPS eligible clinicians or groups that are small, located in rural areas, located in HPSAs, or non-patient-facing MIPS eligible clinicians or groups, in order to achieve a 50 percent score, one CPIA is required (either medium or high).

MIPS eligible clinicians or groups that participate in APMs are considered eligible to participate under the CPIA performance category unless they are participating in an Advanced APM and they have met the Qualifying APM Participant (QP) thresholds or are Partial QPs that elect not to report information. A MIPS eligible clinician or group that is participating in an APM and participating under the CPIA performance category will receive 50 percent of the total CPIA score (30 points) just through their APM participation. These are MIPS eligible clinicians or groups that CMS identifies as participating in APMs for MIPS and may participate under the CPIA performance category. To achieve 100 percent of the total CPIA score, MIPS eligible clinicians or groups will need to identify that they participate in an alternative payment model (30 points) and also select additional CPIAs for an additional 30 points to reach the 60 point CPIA highest score.

For further description of MIPS eligible clinicians or groups that are required to report to MIPS under the APM scoring standard and their CPIA scoring requirements, we refer readers to section II.E.5.h. For all other MIPS eligible clinicians or groups participating in APMs that would report to MIPS, this section applies and we also refer readers to the scoring requirements for these MIPS eligible clinicians or groups in section II.E.6.

Since we cannot measure variable performance within a single CPIA, we propose at § 414.1380 to compare the CPIA points associated with the reported activities against the highest number of points that are achievable under the CPIA performance category which is 60 points. We propose that the highest potential score of 100 percent can be achieved by selecting a number of activities that will add up to 60 points. MIPS eligible clinicians and groups, including those that are participating as an APM, and all those that select activities under the CPIA performance category can achieve the highest potential score of 60 points by selecting activities that are equal to the 60-point maximum. We refer readers to scoring section II.E.6 for additional rationale for using 60 points for the first year.

If a MIPS eligible clinician or group reports only one CPIA, we will score that activity accordingly, as 10 points for a medium-level activity or 20 points for a high-level activity. If a MIPS eligible clinician or group reports no CPIAs, then the MIPS eligible clinician or group would receive a zero score for the CPIA performance category. We believe this proposal allows us to capture variation in the total CPIAs reported.

In addition, we believe these are reasonable criteria for MIPS eligible clinicians or groups to accomplish within the first year for three reasons: (1) In response to several stakeholder MIPS and APMs RFI comments, we are not recommending a minimum number of hours for performance of an activity; (2) we are offering a broad list of activities from which MIPS eligible clinicians or groups may select; and (3) also in response to MIPS and APMs RFI comments, we are proposing that an activity must be performed for at least 90 days during the performance period for CPIA credit. We intend to reassess this requirement threshold in future years. We do not believe it is appropriate to require a determined number of activities within a specific subcategory at this time. This proposal aligns with the requirements in section 1848(q)(2)(C)(iii) of the Act that states MIPS eligible clinicians or groups are not required to perform activities in each subcategory.

Lastly, we recognize that working with a QCDR could allow a MIPS eligible clinician or group to meet the measure and activity criteria for multiple CPIAs. For the first year of MIPS, there are several CPIAs in the inventory that incorporate QCDR participation. Each activity must be selected and achieved separately for the first year of MIPS. A MIPS eligible clinician or group cannot receive credit for multiple activities just by selecting one activity that includes participation in a QCDR. As the CPIA inventory expands over time we are interested in receiving comments on what restrictions, if any, should be placed around CPIA measures and activities that incorporate QCDR participation.

(d) Required Period of Time for Performing an Activity

We propose § 414.1360 that MIPS eligible clinicians or groups must perform CPIAs for at least 90 days during the performance period for CPIA credit. We understand there are some activities that are ongoing whereas others may be episodic. We considered setting the threshold for the minimum time required for performing an activity to longer periods up to a full calendar year. However, after researching several organizations we believe a minimum of 90 days is a reasonable amount of time. Two illustrative examples of organizations that used 90 days as a window for reviewing clinical practice improvements include practice improvement activities undertaken by anesthesiologists, as detailed in a study describing anesthesiologists' practice improvements as part of the Maintenance of Certification in Anesthesiology Program requiring a 90-day report back period, and a large Veteran's Administration health care program that set a 90-day window for reviewing improvements in the management of opioid dispensing.

Additional clarification for how some activities meet the 90-day rule or if additional time is required are reflected in the description of that activity in Table H of the Appendices. In addition we propose that activities, where applicable, may be continuing (that is, could have started prior to the performance period and are continuing) or be adopted in the performance period as long as an activity is being performed for at least 90 days during the performance period.

We anticipate in future years that extended CPIA time periods will be needed for certain activities. We will monitor the time period requirement to asses if allowing for extended time requirements may enhance the value associated with generating more effective outcomes, or conversely, the extended time may reveal that more time has little or no value added for certain activities when associated with desired outcomes. We request comments on this proposal.

(4) Application of CPIA to Non-Patient-Facing MIPS Eligible Clinicians and Groups

We understand that non-patient-facing MIPS eligible clinicians and groups may have a limited number of measures and activities to report. Therefore, we propose at § 414.1360 allowing non-patient-facing MIPS eligible clinicians and groups to report on a minimum of one activity to achieve partial credit or two activities to achieve full credit to meet the CPIA submission criteria. These non-patient-facing MIPS eligible clinicians and groups receive partial or full credit for submitting one or two activities irrespective of any type of weighting, medium or high (for example, two medium activities will qualify for full credit). For scoring purposes, non-patient-facing MIPS eligible clinicians or groups receive 30 points per activity, regardless of whether the activity is medium or high. For example, one high activity and one medium activity could be selected to receive 60 points. Similarly, two medium activities could also be selected to receive 60 points.

We anticipate the number of activities for non-patient-facing MIPS eligible clinicians or groups will increase in future years as we gather more data on the feasibility of performing CPIAs. As part of the process for identifying activities, we consulted with several organizations that represent a cross-section of non-patient-facing MIPS eligible clinicians and groups. An illustrative example of those consulted with include organizations that represent cardiologists involved in nuclear medicine, nephrologists who serve only in a consulting role to other providers, or pathologists who, while they typically function as a team, have different members that perform different roles within their specialty that are primarily non-patient-facing.

In the course of those discussions these organizations identified CPIAs they believed would be applicable. Comments on activities appropriate for non-patient-facing MIPS eligible clinicians or groups are reflected in the proposed CPIA Inventory across multiple subcategories. For example, several of these organizations suggested consideration for Appropriate Use Criteria (AUC). As a result, we have incorporated AUC into some of the activities. We encourage MIPS eligible clinicians or groups who are already required to use AUC (for example, for advanced imaging) to report a CPIA other than one related to appropriate use. Another example, under Patient Safety and Practice Assessment, is the implementation of an antibiotic stewardship program that measures the appropriate use of antibiotics for several different conditions (Upper Respiratory Infection (URI) treatment in children, diagnosis of pharyngitis, bronchitis treatment in adults) according to clinical guidelines for diagnostics and therapeutics. In addition, we request comments on what activities would be appropriate for non-patient-facing MIPS eligible clinicians or groups to add to the CPIA Inventory in the future. We request comments on this proposal.

(5) Special Consideration for Small, Rural, or Health Professional Shortage Areas Practices

As noted previously in this proposed rule, section 1848(q)(2)(B)(iii) of the Act requires the Secretary, in establishing CPIAs, to give consideration to small practices (15 or fewer clinicians) and practices located in rural areas (proposed definition at § 414.1305) and in geographic based HPSAs as designated under section 332(a)(1)(A) of the Public Health Service Act. In the MIPS and APMs RFI, we requested comments on how CPIAs should be applied to MIPS eligible clinicians or groups in small practices, in rural areas, and geographic HPSAs: If a lower performance requirement threshold or different measures should be established that will better allow those MIPS eligible clinicians or groups to perform well in this performance category, what methods should be leveraged to appropriately identify these practices, and what best practices should be considered to develop flexible and adaptable CPIAs based on the needs of the community and its population.

We engaged high performing organizations, including several rural health clinics with 15 or fewer clinicians that are designated as geographic HPSAs, to provide feedback on relevant QIN/QIO activities based on their specific circumstances. Some examples provided include participation in implementation of self-management programs such as for diabetes, and early use of telemedicine, as in the one case for a top performing multi-specialty rural practice that covers 20,000 people over a 25,000-mile radius in a rural area of North Dakota. Comments on activities appropriate for MIPS eligible clinicians or groups located in rural areas or practices that are designated as geographic HPSAs are reflected in the proposed CPIA Inventory across multiple subcategories.

Based on the review of comments and listening sessions, we propose at § 414.1360 to accommodate small practices and practices located in rural areas, or geographic HPSAs for the CPIA performance category by allowing MIPS eligible clinicians or groups to submit a minimum of one activity to achieve partial credit or two activities to achieve full credit. These MIPS eligible clinicians or groups receive partial or full credit for submitting two activities of any type of weighting (for example, two medium activities will qualify for full credit). We anticipate the requirement on the number of activities for small practices and practices located in rural areas, or practices in geographic HPSAs will increase in future years as we gather more data on the feasibility of small practices and practices located in rural areas and practices located in geographic HPSAs to perform CPIAs. Therefore, we request comments on what activities would be appropriate for these practices for the CPIA Inventory in future years. We request comments on this proposal.

(6) CPIA Subcategories

Section 1848(q)(2)(B)(iii) of the Act provides that the CPIA performance category must include at least the subcategories listed below. The statute also provides the Secretary discretion to specify additional subcategories for the CPIA performance category, which have also been included below.

• Expanded practice access, such as same day appointments for urgent needs and after-hours access to clinician advice.
• Population management, such as monitoring health conditions of individuals to provide timely health care interventions or participation in a QCDR.
• Care coordination, such as timely communication of test results, timely exchange of clinical information to patients and other MIPS eligible clinicians or groups, and use of remote monitoring or telehealth.
• Beneficiary engagement, such as the establishment of care plans for individuals with complex care needs, beneficiary self-management assessment and training, and using shared decision-making mechanisms.
• Patient safety and practice assessment, such as through the use of clinical or surgical checklists and practice assessments related to maintaining certification.
• Participation in an APM, as defined in section 1833(z)(3)(C) of the Act.

In the MIPS and APMs RFI, we requested recommendations on the inclusion of the following five potential new subcategories:

• Promoting Health Equity and Continuity, including (a) serving Medicaid beneficiaries, including individuals dually eligible for Medicaid and Medicare, (b) accepting new Medicaid beneficiaries, (c) participating in the network of plans in the Federally Facilitated Marketplace or state exchanges, and (d) maintaining adequate equipment and other accommodations (for example, wheelchair access, accessible exam tables, lifts, scales, etc.) to provide comprehensive care for patients with disabilities.
• Social and Community Involvement, such as measuring completed referrals to community and social services or evidence of partnerships and collaboration with the community and social services.
• Achieving Health Equity, as its own performance category or as a multiplier where the achievement of high quality in traditional areas is rewarded at a more favorable rate for MIPS eligible clinicians or groups that achieve high quality for underserved populations, including persons with behavioral health conditions, racial and ethnic minorities, sexual and gender minorities, people with disabilities, people living in rural areas, and people in geographic HPSAs.
• Emergency preparedness and response, such as measuring MIPS eligible clinician or group participation in the Medical Reserve Corps, measuring registration in the Emergency System for Advance Registration of Volunteer Health Professionals, measuring relevant reserve and active duty military MIPS eligible clinician or group activities, and measuring MIPS eligible clinician or group volunteer participation in domestic or international humanitarian medical relief work.
• Integration of primary care and behavioral health, such as measuring or evaluating such practices as: Co-location of behavioral health and primary care services; shared/integrated behavioral health and primary care records; or cross-training of MIPS eligible clinicians or groups participating in integrated care. This subcategory also includes integrating behavioral health with primary care to address substance use disorders or other behavioral health conditions, as well as integrating mental health with primary care.

We recognize that quality improvement is a critical aspect of improving the health of individuals and the health care delivery system overall. We also recognize that this will be the first time MIPS eligible clinicians or groups will be measured on the quality improvement work on a national scale. We have approached the CPIA performance category with these principles in mind along with the overarching principle for the MIPS program that we are building a process that will have increasingly more stringent requirements over time.

Therefore, for the first year of MIPS, we propose at § 414.1365 that the CPIA performance category include the subcategories of activities provided at section 1848(q)(2)(B)(iii) of the Act. In addition, we propose at § 414.1365 adding the following subcategories: “Achieving Health Equity”, “Integrated Behavioral and Mental Health”, and “Emergency Preparedness and Response.” In response to multiple MIPS and APMs RFI comments requesting the inclusion of “Achieving Health Equity,” we are proposing to include this subcategory because: (1) It is important and may require targeted effort to achieve and so should be recognized when accomplished; (2) supports our national priorities and programs, such as Reducing Health Disparities; and (3) encourages “use of plans, strategies, and practices that consider the social determinants that may contribute to poor health outcomes.” (CMS, Quality Innovation Network Quality Improvement Organization Scope of Work: Excellence in Operations and Quality Improvement, 2014).

Similarly, MIPS and APMs RFI comments strongly supported the inclusion of the subcategory of “Integrated Behavioral and Mental Health”, citing that “statistics show 50 percent of all behavioral health disorders are being treated by primary care and behavioral health integration.” Additionally, according to MIPS and APMs RFI comments, behavioral health integration with primary care is already being implemented in numerous locations throughout the country. The third additional subcategory we propose to include is “Emergency Preparedness and Response,” based on MIPS and APMs RFI comments that encouraged us to consider this subcategory to help ensure that practices remain open during disaster and emergency situations and support emergency response teams as needed. Additionally, commenters were able to provide a sufficient number of recommended activities (that is, more than one) that could be included in the CPIA Inventory in all of these proposed subcategories and the subcategories included under section 1848(q)(2)(B)(iii) of the Act.

We also seek public comments on two additional subcategories for future consideration:

• Promoting Health Equity and Continuity, including (a) serving Medicaid beneficiaries, including individuals dually eligible for Medicaid and Medicare, (b) accepting new Medicaid beneficiaries, (c) participating in the network of plans in the Federally Facilitated Marketplace or state exchanges, and (d) maintaining adequate equipment and other accommodations (for example, wheelchair access, accessible exam tables, lifts, scales, etc.) to provide comprehensive care for patients with disabilities; and
• Social and Community Involvement, such as measuring completed referrals to community and social services or evidence of partnerships and collaboration with community and social services.

For these two subcategories, we are requesting activities that can demonstrate some improvement over time and go beyond current practice expectations. For example, maintaining existing medical equipment would not qualify for a CPIA, but implementing some improved clinical workflow processes that reduce wait times for patients with disabilities or improve coordination of care including activities that regularly provide additional assistance to find other care needed for patients with disabilities, would be some examples of activities that could show improvement in clinical practice over time.

We request comments on these proposals.

(7) CPIA Inventory

To implement the MIPS program, we are required to create an inventory of CPIAs. Consistent with our MIPS strategic goals, we believe it is important to create a broad list of activities that can be used by multiple practice types to demonstrate CPIAs and activities that may lend themselves to being measured for improvement in future years.

We took several steps to ensure the initial CPIA Inventory is inclusive of activities in line with the statutory intent. We had numerous interviews with highly performing organizations of all sizes, conducted an environmental scan to identify existing models, activities, or measures that met all or part of the CPIA category, including the patient centered medical homes, the Transforming Clinical Practice Initiative (TCPI), Consumer Assessment of Healthcare Providers and Systems (CAHPS) surveys, and AHRQ's Patient Safety Organizations. In addition, we reviewed the CY 2016 PFS final rule with comment period (80 FR 70886) and the comments received in response to the MIPS and APMs RFI regarding the CPIA performance category. The CPIA Inventory was compiled as a result of the stakeholder input, an environmental scan, MIPS and MIPS and APMs RFI comments, and subsequent working sessions with AHRQ and ONC and additional communications with CDC, SAMHSA and HRSA.

Based on the above discussions we established guidelines for CPIA inclusion based on one or more of the following criteria (in any order):

• Relevance to an existing CPIA subcategory (or a proposed new subcategory);
• Importance of an activity toward achieving improved beneficiary health outcome;
• Importance of an activity that could lead to improvement in practice to reduce health care disparities;
• Aligned with patient-centered medical homes;
• Representative of activities that multiple MIPS eligible clinicians or groups could perform (for example, primary care, specialty care);
• Feasible to implement, recognizing importance in minimizing burden, especially for small (15 or fewer clinicians) practices, practices in rural areas, or in areas designated as geographic HPSAs by HRSA;
• CMS is able to validate the activity; or
• Evidence supports that an activity has a high probability of contributing to improved beneficiary health outcomes.

Activities that overlap with other performance categories were excluded unless there was a strong policy rationale to include it in the CPIA Inventory. We propose to use the CPIA Inventory for the first year of MIPS, as provided in Table H of the Appendices. For further description of how MIPS eligible clinicians or groups will be designated to submit to MIPS for CPIA, we refer readers to section II.E.6.h. For all other MIPS eligible clinicians or groups participating in APMs that would report to MIPS, this section applies and we also refer readers to the scoring requirements for these MIPS eligible clinicians or groups in section II.E.5. of this proposed rule.

We request comments on the inventory and welcome suggestions for CPIAs for future years as well.

(a) CMS Study on CPIA and Measurement

(1) Study Purpose

From our experience under the PQRS, VM, and Medicare EHR Incentive programs we have discovered that many providers have errors within their data sets, as well as issues understanding the data that corresponds to their selected quality measures. To help better understand the current processes and limitations, we propose to conduct a study on CPIAs and measurement to examine clinical quality workflows and data capture using a simpler approach to quality measures. The study will allow a limited number of selected MIPS eligible clinicians and groups to receive full credit (60 points) for the CPIA category.

The lessons learned in this study on practice improvement and measurement may or may not influence changes to future MIPS data submission requirements. The goals of the study are to see whether there will be improved outcomes, reduced burden in reporting, and enhancements in clinical care by selected MIPS eligible clinicians desiring:

• A more data driven approach to quality measurement.
• Measure selection unconstrained with a CEHRT program or system.
• Improving data quality submitted to CMS.
• Enabling CMS get data more frequently and provide feedback more often.

(2) Study Participation Credit and Requirements

Eligible clinicians and groups in the CMS study on practice improvement and measurement will receive full credit for the CPIA category of MIPS after successfully electing, participating and submitting data to CMS. Based on feedback and surveys from MIPS eligible clinicians, study measurement data will be made available to CMS throughout the study on at least a quarterly basis unless the MIPS eligible clinician or group agrees to submit data on a more frequent basis. Participants will be required to attend a monthly focus group to share lessons learned along with providing survey feedback to monitor effectiveness. The focus group will also include providing visual displays of data, workflows, and best practices to be shared amongst the participants to obtain feedback and make further improvements. The monthly focus groups will be used to learn from the practices on how to be more agile as we test new ways of measure recording and workflow.

For the 2017 performance period, the participating MIPS eligible clinicians or groups would submit their data and workflows for a minimum of three MIPS clinical quality measures that are relevant and prioritized by their practice. One of the measures must be an outcome measure, and one must be a patient experience measure. The participating MIPS eligible clinicians could elect to report on more measures as this would provide more options from which to select in subsequent years for purposes of measuring improvement.

If MIPS eligible clinicians or groups calculate the measures working with a QCDR, qualified registry, or CMS-approved third party intermediary, CMS will use the same data validation process described in section II.E.8.e. CMS will only collect the numerator and denominator for the measures selected for the overall population, all patients/all payers. This will enable the practices to build the measures based on what is important for their area of practice while increasing the quality of care.

In future years, participating MIPS eligible clinicians or groups would select three of the measures for which they have baseline data from the 2017 performance period to compare against later performance years. Participants electing to continue in future years will be afforded the opportunity opt-in or opt-out following the successful submission of data to CMS. The first opportunity to continue in the study will be at the end of the 2017 performance period. Eligible clinicians who elect to join the study but fail to participate and/or fail to successfully submit the data required will be removed from the study. Unsuccessful study participants will then be subject to the full requirements for the CPIA category.

(3) Study Participation Eligibility

Participation will be open to a limited number of MIPS eligible clinicians in rural settings and non-rural settings. A rural area is defined at § 414.1305 and a non-rural area would be any MIPS eligible clinicians or groups not included as part of the rural definition. This test will be open to include up to 10 non-rural individual MIPS eligible clinicians or groups of less than three non-rural MIPS eligible clinician's, 10 rural individual MIPS eligible clinicians or groups of less than three rural MIPS eligible clinician's, 10 groups of three to eight MIPS eligible clinicians, five groups of nine to twenty MIPS eligible clinicians, three groups of twenty-one to one hundred MIPS eligible clinicians, two groups of greater than 100 MIPS eligible clinicians, and two specialist groups of MIPS eligible clinicians. Eligible clinicians and groups will need to sign up from January 1, 2017, to January 31, 2017. The sign up process will utilize this web-based interface— http://oncprojectracking.org/ . Participants will be approved on a first come first served basis and must meet all the required criteria.

We request comment on the study and welcome suggestions on future study topics.

(8) CPIA Policies for Future Years of the MIPS Program

(a) Proposed Approach for Identifying New Subcategories and New Activities

We propose, for future years of, MIPS, to consider the addition of a new subcategory or activity to the CPIA Inventory only when the following criteria are met:

• The new subcategory represents an area that could highlight improved beneficiary health outcomes, patient engagement and safety based on evidence.
• The new subcategory has a designated number of activities that meet the criteria for a CPIA activity and cannot be classified under the existing subcategories.
• Newly identified subcategories would contribute to improvement in patient care practices or improvement in performance on quality measures and resource use performance categories.

In future years, MIPS eligible clinicians or groups will have an opportunity to nominate additional subcategories, along with activities associated with each of those subcategories that are based on criteria specified for these activities, as discussed above.

We request comments on this proposal.

(b) Request for Comments on Call for Measures and Activities Process for Adding New Activities and New Subcategories

We plan to develop a call for measures and activities process for future years of MIPS, where MIPS eligible clinicians or groups and other relevant stakeholders may recommend activities for potential inclusion in the CPIA Inventory. As part of the process, MIPS eligible clinicians or groups would be able to nominate additional activities that we could consider adding to the CPIA Inventory. The MIPS eligible clinician or group or relevant stakeholder would be able to provide an explanation of how the activity meets all the criteria we have identified. This nomination and acceptance process would, to the best extent possible, parallel the annual call for measures process already conducted by CMS for quality measures. The final CPIA Inventory for the performance year would be published in accordance with the overall MIPS rulemaking timeline and program. In addition, in future years we anticipate developing a process and establishing criteria to remove or add new activities to CPIA.

Additionally, prospective activities that are submitted through a QCDR could also be included as part of a beta-test process that may be instrumental for future years to determine whether that activity should be included in the CPIA Inventory based on specific criteria noted above. MIPS eligible clinicians or groups and groups that use QCDRs to capture data associated with an activity, for example the frequency in administering depression screening and a follow-up plan, may be asked to voluntarily submit that same data in year 2 to begin identifying a baseline for improvement for subsequent year analysis. This is not intended to require any MIPS eligible clinician or group to submit CPIAs only via QCDR from one year to the next or to require the same activity from one year to the next. Participation in doing so, however, can help to identify how activities can contribute to improve outcomes. This data submission process will be considered part of a beta-test to: (1) Determine if the activity is being regularly conducted and effectively executed and (2) if the activity warrants continued inclusion on the CPIA Inventory. The data will help capture baseline information to begin measuring improvement and inform the Secretary of the likelihood that the activity would result in improved outcomes. If an activity is submitted and reported by a QCDR, it would be reviewed by CMS for final inclusion in the CPIA Inventory the following year, even if these activities are not submitted through the future call for measures and activities process. We intend, in future performance years, to begin measuring CPIA data points for all eligible clinicians and to award scores based on performance and improvement. We solicit comment on how best to collect such CPIA data and factor it into future scoring under MIPS.

We request comments on this approach and on any other considerations we should take into account when developing this type of approach for future rulemaking.

(c) Request for Comments on Use of QCDRs for Identification and Tracking of Future Activities

In future years, we expect to learn more about CPIAs and how the inclusion of additional measures and activities captured by QCDRs could enhance the ability of MIPS eligible clinicians or groups to capture and report on more meaningful activities. This is especially true for specialty groups. In the future, we may propose use of QCDRs for identification and acceptance of additional measures and activities which is in alignment with section 1848(q)(1)(E) of the Act which encourages the use of QCDRs, as well as under section 1848(q)(2)(B)(iii)(II) of the Act related to the population management subcategory. We recognize, through the MIPS and APMs RFI comments and interviews with organizations that represent non-patient-facing MIPS eligible clinicians or groups and specialty groups that QCDRs may provide for a more diverse set of measures and activities under CPIA than are possible to list under the current CPIA Inventory. This diverse set of measures and activities, which we can validate, affords specialty practices additional opportunity to report on more meaningful activities in future years. QCDRs may also provide the opportunity for longer-term data collection processes which will be needed for future year submission on improvement, in addition to achievement. Use of QCDRs also supports ongoing performance feedback and allows for implementation of continuous process improvements. We believe that for future years, QCDRs will be allowed to define specific CPIAs for specialty and non-patient-facing MIPS eligible clinicians or groups through the already-established QCDR approval process for measures and activities. We request comments on this approach.

g. Advancing Care Information Performance Category

(1) Background and Relationship to Prior Programs

(a) Background

The American Recovery and Reinvestment Act of 2009 (ARRA), which included the Health Information Technology for Economic and Clinical Health Act (HITECH Act), amended Titles XVIII and XIX of the Act to authorize incentive payments and Medicare payment adjustments for EPs to promote the adoption and meaningful use of certified EHR technology (CEHRT). Section 1848(o) of the Act provides the statutory basis for the Medicare incentive payments made to meaningful EHR users. Section 1848(a)(7) of the Act also establishes downward payment adjustments, beginning with calendar year (CY) 2015, for EPs who are not meaningful users of certified EHR technology for certain associated EHR reporting periods. (For a more detailed explanation of the statutory basis for the Medicare and Medicaid EHR Incentive Programs, see the July 28, 2010 Stage 1 final rule titled, “Medicare and Medicaid Programs; Electronic Health Record Incentive Program; Final Rule” (75 FR 44316 through 44317).)

A primary policy goal of the EHR Incentive Program is to encourage and promote the adoption and use of certified EHR technology among Medicare and Medicaid health care providers to help drive the industry as a whole toward the use of certified EHR technology. As described in the final rule titled “Medicare and Medicaid Programs; Electronic Health Record Incentive Program—Stage 3 and Modifications to Meaningful Use in 2015 Through 2017” (Hereinafter referred to as the “2015 EHR Incentive Programs Final Rule”) (80 FR 62769), the HITECH Act outlined several foundational requirements for meaningful use and for EHR technology. CMS and ONC have subsequently outlined a number of key policy goals which are reflected in the current objectives and measures of the program and the related certification requirements (80 FR 62790). Current Medicare EP performance on these key goals is varied, with EPs demonstrating high performance on some objectives while others represent a greater challenge.

(b) MACRA Changes

Section 1848(q)(2)(A) of the Act, as added by section 101(c) of the MACRA, includes the meaningful use of certified EHR technology as a performance category under the MIPS, referred to in this proposed rule as the advancing care information performance category, which will be reported by MIPS eligible clinicians as part of the overall MIPS program. As required by sections 1848(q)(2) and (5) of the Act, the four performance categories shall be used in determining the MIPS CPS for each MIPS eligible clinician. In general, MIPS eligible clinicians will be evaluated under all four of the MIPS performance categories, including the advancing care information performance category. This includes MIPS eligible clinicians who were not previously eligible for the EHR Incentive Program incentive payments under section 1848(o) of the Act or subject to the EHR Incentive Program payment adjustments under section 1848(a)(7) of the Act, such as physician assistants, nurse practitioners, clinical nurse specialists, certified registered nurse anesthetists, and hospital-based EPs (as defined in section 1848(o)(1)(C)(ii) of the Act). Understanding that these MIPS eligible clinicians may not have prior experience with certified EHR technology and the objectives and measures under the EHR Incentive Program, we have proposed a scoring methodology within the advancing care information performance category that provides flexibility for MIPS eligible clinicians from early adoption of certified EHR technology through advanced use of health IT. We note that in section II.e.5.g.8.a of this proposed rule, we have also proposed to reweight the advancing care information performance category to zero in the MIPS composite performance score for certain hospital-based and other MIPS eligible clinicians where the measures proposed for this performance category may not be available or applicable to these types of MIPS eligible clinicians.

(c) Considerations in Defining Advancing Care Information Performance Category

In implementing MIPS, we intend to develop the requirements for the advancing care Information performance category to continue supporting the foundational objectives of the HITECH Act, and to encourage continued progress on key uses such as health information exchange and patient engagement. These more challenging objectives are essential to leveraging certified EHR technology to improve care coordination and they represent the greatest potential for improvement and for significant impact on delivery system reform in the context of MIPS quality reporting.

In developing the requirements and structure for the advancing care information performance category, we considered several approaches for establishing a framework that would naturally integrate with the other MIPS performance categories. We considered historical performance on the EHR Incentive Program objectives and measures, feedback received through public comment, and the long term goals for delivery system reform and quality improvement strategies.

One approach we considered would be to maintain the current structure of the Medicare EHR Incentive Program and award full points for the advancing care information performance category for meeting all of the objectives and measures finalized in the 2015 EHR Incentive Programs final rule, and award zero points for failing to meet all of these requirements. This method would be consistent with the current EHR Incentive Program and is based on objectives and measures already established in rulemaking. However, we considered and dismissed this approach as it would not allow flexibility for MIPS eligible clinicians and would not allow CMS to effectively measure performance for MIPS eligible clinicians in the advancing care information performance category who have taken incremental steps toward the use of certified EHR technology, or to recognize exceptional performance for MIPS eligible clinicians who have excelled in any one area. This is particularly important as many MIPS eligible clinicians may not have had past experience relevant to the advancing care information performance category and use of EHR technology because they were not previously eligible to participate in the Medicare EHR Incentive Program. This approach also does not allow for differentiation among the objectives and measures that have high adoption and those where there is potential for continued advancement and growth.

We subsequently considered several methods which would allow for more flexibility and provide CMS the opportunity to recognize partial or exceptional performance among MIPS eligible clinicians for the measures under the advancing care information performance category. We decided to design a framework that would allow for flexibility and multiple paths to achievement under this category while recognizing MIPS eligible clinicians' efforts at all levels. Part of this framework requires moving away from the concept of requiring a single threshold for a measure, and instead incentivizes continuous improvement, and recognizes onboarding efforts among late adopters and MIPS eligible clinicians facing continued challenges in full implementation of certified EHR technology in their practice.

(2) Advancing Care Information Performance Category Within MIPS

In defining the advancing care information performance category for the MIPS, we considered stakeholder feedback and lessons learned from our experience with the Medicare EHR Incentive Program. Specifically, we considered feedback from the Stage 1 (75 FR 44313) and Stage 2 (77 FR 53967) EHR Incentive Program rules, and the 2015 EHR Incentive Programs final rule (80 FR 62769), as well as comments received from the MIPS and APMs RFI (80 FR 59102). We have learned from this feedback that clinicians desire flexibility to focus on health IT implementation that is right for their practice. We have also learned that updating software, training staff and changing practice workflows to accommodate new technology can take time, and that clinicians need time and flexibility to focus on the health IT activities that are most relevant to their patient population. Clinicians also desire consistent timelines and reporting requirements in order to simplify and streamline the reporting process. Recognizing this, we have worked to align the advancing care information performance category with the other MIPS performance categories, which would streamline reporting requirements, timelines and measures in an effort to reduce burden on MIPS eligible clinicians.

The implementation of the advancing care information performance category is an important opportunity to increase clinician and patient engagement, improve the use of health IT to achieve better patient outcomes, and continue to meet the vision of enhancing the use of certified EHR technology as defined under the HITECH Act. As discussed later in this section, we are proposing in section II.E.5.g.6.a. new flexibility in how we would assess MIPS eligible clinician performance for the advancing care information performance category. We propose to emphasize performance in the objectives and measures that are the most critical and would lead to the most improvement in the use of health IT and health care quality. We intend to promote innovation so that technology can be interconnected easily and securely, and data can be accessed and directed where and when it is needed to support patient care. These objectives include Patient Electronic Access, Coordination of Care Through Patient Engagement and Health Information Exchange, which are essential to leveraging certified EHR technology to improve care. At the same time, we propose to eliminate reporting on objectives and measures in which the vast majority of clinicians already achieve high performance—which would reduce burden, encourage greater participation and direct MIPS eligible clinicians' attention to higher-impact measures. Our proposal balances program participation with rewarding performance on high-impact objectives and measures, which we believe would make the overall program stronger and further the goals of the HITECH Act.

(a) Advancing the Goals of the HITECH Act in MIPS

Section 1848(o)(2)(A) of the Act requires that the Secretary seek to improve the use of electronic health records and health care quality over time by requiring more stringent measures of meaningful use. In implementing MIPS and the advancing care information performance category, we seek to improve and encourage the use of certified EHR technology over time by adopting a new, more flexible scoring methodology, as discussed in section II.E.5.g.6. of this proposed rule, that would more effectively allow MIPS eligible clinicians to reach the goals of the HITECH Act, and would allow MIPS eligible clinicians to use EHR technology in a manner more relevant to their practice. This new, more flexible scoring methodology puts a greater focus on Patient Electronic Access, Coordination of Care Through Patient Engagement, and Health Information Exchange—objectives we believe are essential to leveraging certified EHR technology to improve care by engaging patients and furthering interoperability. This methodology would also de-emphasize objectives in which clinicians have historically achieved high performance with median performance rates of over 90 percent for the last 2 years. We believe shifting focus away from these objectives would reduce burden, encourage greater participation, and direct attention to other objectives and measures which require more attention. Through this flexibility, MIPS eligible clinicians would be incentivized to focus on those aspects of certified EHR technology that are most relevant to their practice, which we believe would lead to improvements in health care quality.

We also seek to increase the adoption and use of certified EHR technology by incorporating such technology into the other MIPS performance categories. For example, in section II.6.a.2.f. of this proposed rule, we are proposing to incentivize electronic reporting by awarding a bonus point for submitting quality measure data using certified EHR technology. Additionally, in section II.E.5.f. of this proposed rule, we have aligned some of the activities under the CPIA performance category such as Care Coordination, Beneficiary Engagement and Achieving Health Equity with a focus on enhancing the use of certified EHR technology. We believe this approach would strengthen the adoption and use of EHR systems and program participation consistent with the provisions of section 1848(o)(2)(A) of the Act.

(b) Future Considerations

We note that the increased flexibility and removal of previously established thresholds for reporting, as proposed in this section of this proposed rule, may appear to be a lower standard than what previously existed in the Medicare EHR Incentive Program. In reality, this restructuring of program requirements is geared toward increasing participation and EHR adoption. We believe this is the most effective way to encourage the adoption of certified EHR technology, and introduce new MIPS eligible clinicians to the use of EHR technology and health IT overall.

We will continue to review and evaluate MIPS eligible clinician performance in the advancing care information performance category, and will consider evolutions in health IT over time as it relates to this performance category. Based on our ongoing evaluation, we expect to adopt changes to the scoring methodology for the advancing care information performance category to ensure the efficacy of the program and to ensure increased value for MIPS eligible clinicians, as well as to adopt more stringent measures of meaningful use as required by section 1848(o)(2)(A) of the Act.

Potential changes may include establishing benchmarks for MIPS eligible clinician performance on the advancing care information performance category measures, and using these benchmarks as a baseline or threshold for future reporting. This may include scoring for performance improvement over time and the potential to reevaluate the efficacy of measures based on these analyses. For example, in future years we may use a MIPS eligible clinician's prior performance on the advancing care information performance category measures as comparison for the subsequent year's performance category score, or compare a MIPS eligible clinician's performance category score to peer groups to measure their improvement and determine a performance category score based on improvement over those benchmarks or peer group comparisons. This type of approach would drive continuous improvement over time through the adoption of more stringent performance standards for the advancing care information performance category measures.

We are committed to continual review, improvement and increased stringency of the advancing care information performance category measures as directed under section 1848(o)(2)(A) of the Act both for the purposes of ensuring program efficacy as well as ensuring value for the MIPS eligible clinicians reporting the advancing care information performance category measures. We seek comment on further methods to increase the stringency of the advancing care information performance category measures in the future.

We additionally seek comment on the concept of a holistic approach to health IT—one that we believe is similar to the concept of outcome measures in the quality performance category in the sense that MIPS eligible clinicians could potentially be measured more directly on how the use of health IT contributes to the overall health of their patients. Under this concept, MIPS eligible clinicians would be able to track certain use cases or patient outcomes to tie patient health outcomes with the use of health IT.

We believe this approach would allow us to directly link health IT adoption and use to patient outcomes, moving MIPS beyond the measurement of EHR adoption and process measurement and into a more patient-focused health IT program. From comments and feedback we have received from the health care provider community, we understand that this type of approach would be a welcome enhancement to the measurement of health IT. At this time, we recognize that technology and measurement for this type of program is currently unavailable. We seek comment on what this type of measurement would look like under MIPS, including the type of measures that would be needed within the advancing care information performance category and the other performance categories to measure this type of outcome, what functionalities with certified EHR technology would be needed, and how such an approach could be implemented.

(3) Clinical Quality Measurement

Section 1848(o)(2)(A)(iii) of the Act requires the reporting of clinical quality measures (CQMs) using certified EHR technology. Section 1848(q)(5)(B)(ii)(II) provides that under the methodology for assessing the total performance of each MIPS eligible clinician, the Secretary shall, with respect to a performance period for a year, for which a MIPS eligible clinician reports applicable measures under the quality performance category through the use of certified EHR technology, treat the MIPS eligible clinician as satisfying the CQMs reporting requirement under section 1848(o)(2)(A)(iii) of the Act for such year. We note that in the context and overall structure of MIPS, the quality performance category allows for a greater focus on patient-centered measurement, and multiple pathways for MIPS eligible clinicians to report their quality measure data. Therefore, we are not proposing separate requirements for clinical quality measure reporting within the advancing care information performance category and instead would require submission of quality data for measures specified for the quality performance category, in which we encourage reporting of CQMs with data captured in certified EHR technology. We refer readers to section II.E.5.a of this proposed rule for discussion of reporting of CQMs with data captured in certified EHR technology under the quality performance category.

(4) Performance Period Definition for Advancing Care Information Performance Category

In the Medicare and Medicaid Programs; Electronic Health Record Incentive Program—Stage 3 proposed rule, we proposed to eliminate the 90-day EHR reporting period beginning in 2017 for EPs who had not previously demonstrated meaningful use, with a limited exception for the Medicaid EHR Incentive Program (80 FR 16739-16740, 16774-16775). We received many comments from respondents stating their preference for maintaining the 90-day EHR reporting period to allow first time participants to avoid payment adjustments. In addition, commenters indicated that the 90-day time period reduced administrative burden and allowed for needed time to adapt their EHRs to ensure they could achieve program objectives. As a result, we did not finalize our proposal and established a 90-day EHR reporting period for all EPs in 2015 and for new participants in 2016, as well as a 90-day EHR reporting period for new participants in 2015, 2016, and 2017 with regard to the payment adjustments (80 FR 62777-62779; 62904-62906).

Moving forward, the implementation of MIPS creates a critical opportunity to align performance periods to ensure that quality, CPIA, resource use, and the advancing care information performance categories are all measured and scored based on the same period of time. We believe this would lower reporting burden, focus clinician quality improvement efforts and align administrative actions so that clinicians can use common systems and reporting pathways.

Under MIPS, we propose to align the performance period for the advancing care information performance category to the proposed MIPS performance period of one full calendar year. Thus, the performance period for the advancing care information performance category would be the same as the performance periods for the other performance categories as indicated in section II.E.4. We note that there would not be a separate 90-day performance period for the advancing care information performance category. Under this proposal, MIPS eligible clinicians would need to submit data based on performance period starting January 1, 2017, and ending December 31, 2017 for the first year of MIPS. We recognize that stakeholders may still have concerns related to a full year performance period. We note that, as discussed in section II.E.4. of this proposed rule, MIPS eligible clinicians that only have data for a portion of the year can still submit data, be assessed and be scored for the advancing care information performance category. Under the proposal, MIPS eligible clinicians would need to possess certified EHR technology and report on the objectives and measures (without meeting any thresholds) during the calendar year performance period to achieve the advancing care information category base score. We note that MIPS eligible clinicians would be required to submit all of the data they have available for the performance period, even if the time period they have data for is less than one full calendar year.

We believe this proposal would reduce reporting burden and streamline requirements so that MIPS eligible clinicians and third party intermediaries, such as registries and QCDRs, would have a common timeline for data submission to all performance categories. We refer readers to section II.E.4. of this proposed rule for discussion of the performance period for MIPS and solicit feedback on our proposal.

(5) Advancing Care Information Performance Category Data Submission and Collection

(a) Definition of Meaningful EHR User and Certification Requirements

The use of certified health IT continues to be an important component of care delivery for clinicians. Certified health IT that advances patient engagement, interoperability, and privacy and security are key to care coordination, and a critical component in improving health outcomes.

We anticipate that as certified health IT and related standards continue to evolve to support health information exchange, care coordination (for example, referral management), and other capabilities, we will consider updates to the certified health IT requirements for MIPS. We continue to work with the Office of the National Coordinator for Health IT to identify certified health IT that would aid clinicians in MIPS.

Throughout this proposed rule, we use the terms “certified health IT” and “certified EHR technology”. These terms refer to health information technologies and systems that are certified to various standards and functions under the ONC Health IT Certification Program. In general, the full range of potential technologies, functions, standards, and systems for which ONC has established certification criteria are referred to as “certified health IT” (See the 2015 Edition Health IT Certification Criteria final rule (80 FR 62604)). In contrast, the term “certified EHR technology” is a statutory and regulatory term that defines the technology that MIPS eligible clinicians and participants in Advanced APMs must use.

It is important to note that certified EHR technology is a part of the larger category of certified health IT. Therefore when discussing certified health IT in a broad and general manner; such a discussion includes both the functions included in certified EHR technology and other additional potential functions and criteria. In other words, certified EHR technology is a subset of the broader definition of certified health IT.

“Certified health IT” is used in two different ways within this proposed rule. The first is stated as “certified health IT” to identify where the text is referencing a broad range of technology that is included in the ONC Health IT Certification Program. The second use is where the term “a certified Health IT Module” identifies a technology or function used independently from the clinicians' EHR. An example of this second use of the term includes the certified functions leveraged by Health Information Exchange organizations, QCDRs, and public health agencies to support actions like information exchange, quality measurement, and data submission. These individual functions may also be a part of the certified EHR technology definition and may connect with the EHR, but are in these cases used independently from the clinicians' EHR systems.

ONC and CMS worked closely to identify the set of certified health IT that are part of the certified EHR technology definitions proposed in this rule. For example, ONC's 2015 Edition Health Information Technology (Health IT) Certification Criteria, 2015 Edition Base Electronic Health Record (EHR) Definition, and ONC Health IT Certification Program Modifications (80 FR 62602 through 62759) hereinafter referred to as “2015 Edition final rule”, defines the technological requirements for health IT systems used by EHR Incentive Program participants. In this proposed rule, we are proposing to adopt a definition of certified EHR technology at § 414.1305 for MIPS eligible clinicians that is based on the definition that applies in the EHR Incentive Programs under 42 CFR 495.4.

In the 2015 EHR Incentive Programs final rule (80 FR 62873) we outlined the requirements for EPs using certified EHR technology in 2017 as it relates to the objectives and measures they select to report. We propose at § 414.1375 similar requirements for the use of certified EHR technology in relation to the selection of objectives and measures under the MIPS advancing care information performance category.

For 2017, the first MIPS performance period, MIPS eligible clinicians would be able to use EHR technology certified to either the 2014 or 2015 Edition certification criteria as follows:

• A MIPS eligible clinician who only has technology certified to the 2015 Edition may choose to report: (1) On the objectives and measures specified for the advancing care information performance category in section II.E.5.g.7 of this proposed rule, which correlate to Stage 3 requirements; or (2) on the alternate objectives and measures specified for the advancing care information performance category in section II.E.5.g.7 of this proposed rule, which correlate to modified Stage 2 requirements.
• A MIPS eligible clinician who has technology certified to a combination of 2015 Edition and 2014 Edition may choose to report: (1) On the objectives and measures specified for the advancing care information performance category in section II.E.5.g.7 of this proposed rule, which correlate to Stage 3; or (2) on the alternate objectives and measures specified for the advancing care information performance category as described in section II.E.5.g.7 of this proposed rule, which correlate to modified Stage 2, if they have the appropriate mix of technologies to support each measure selected.
• A MIPS eligible clinician who only has technology certified to the 2014 Edition would not be able to report on any of the measures specified for the advancing care information performance category described in section II.E.5.g.7 of this proposed rule that correlate to a Stage 3 measure that requires the support of technology certified to the 2015 Edition. These MIPS eligible clinicians would be required to report on the alternate objectives and measures specified for the advancing care information performance category as described in section II.E.5.g.7. of this proposed rule, which correlate to modified Stage 2 objectives and measures.

Beginning with the performance period in 2018, MIPS eligible clinicians:

• Must only use technology certified to the 2015 Edition to meet the objectives and measures specified for the advancing care information performance category in section II.E.5.g.7. of this proposed rule, which correlate to Stage 3.

We welcome comments on this proposal, which is intended to maintain consistency across MIPS, the Medicare EHR Incentive Program and the Medicaid EHR Incentive Program.

Finally, we propose to define at § 414.1305 a meaningful EHR user under MIPS as a MIPS eligible clinician who possesses certified EHR technology, uses the functionality of certified EHR technology, and reports on applicable objectives and measures specified for the advancing care information performance category for a performance period in the form and manner specified by CMS.

We invite comments on our proposals.

(b) Method of Data Submission

Under the Medicare EHR Incentive Program, EPs attest to the numerators and denominators for certain objectives and measures, through a CMS web portal. For the purpose of reporting advancing care information performance category objectives and measures under the MIPS, we propose at § 414.1325 to allow for MIPS eligible clinicians to submit advancing care information performance category data through qualified registry, EHR, QCDR, attestation and CMS Web Interface submission methods. Regardless of data submission method, all MIPS eligible clinicians must follow the reporting requirements for the objectives and measures to meet the requirements of the advancing care information performance category.

We note that under this proposal, 2017 would be the first year that EHRs (through the QRDA submission method), QCDRs and qualified registries would be able to submit EHR Incentive Program objectives and measures (as adopted for the advancing care information performance category) to CMS, and the first time this data would be reported through the CMS Web Interface. We recognize that some Health IT vendors, QCDRs and qualified registries may not be able to conduct this type of data submission for the 2017 performance period given that the development efforts associated with this data submission capability. However, we are including these data submission mechanisms in 2017 to support early adopters and to signal our longer-term commitment to working with organizations that are agile, effective and can create less burdensome data submission mechanisms for MIPS eligible clinicians. We believe the proposed data submission methods could reduce reporting burden by synchronizing reporting requirements and data submission, and systems, allow for greater access and ease in submitting data throughout the MIPS program. We note that specific details about the form and manner for data submission will be addressed by CMS in the future.

(c) Group Reporting

Under the Medicare EHR Incentive Program, CMS adopted a reporting mechanism for EPs that are part of a group to attest using one common form, or batch reporting process. Under that batch reporting process CMS assessed the individual performance of the EPs that made up the group, not the group as a whole, to determine whether those EPs meaningfully used certified EHR technology.

The structure of the MIPS and our desire to achieve alignment across the MIPS performance categories appropriately necessitates the ability to assess the performance of MIPS eligible clinicians at the group level for all MIPS performance categories. We believe MIPS eligible clinicians should be able to submit data as a group, and be assessed at the group level, for all of the MIPS performance categories, including the advancing care information performance category. For this reason, we are proposing a group reporting mechanism for individual MIPS eligible clinicians to have their performance assessed as a group for all performance categories in section II.E.1.e. of this proposed rule, consistent with section 1848(q)(1)(D)(i)(I) & (II) of the Act.

Under this option, we are proposing that performance on advancing care information performance category objectives and measures would be assessed and reported at the group level, as opposed to the individual MIPS eligible clinician level. We note that the data submission criteria would be the same when submitted at the group-level as if submitted at the individual-level, but the data submitted would be aggregated for all MIPS eligible clinicians within the group practice. We believe this approach to data submission better reflects the team dynamics of groups, and would reduce the overall reporting burden for MIPS eligible clinicians that practice in groups, incentivize practice-wide approaches to data submission, and provide enterprise-level continuous improvements strategies for submitting data to the advancing care information performance category. Please see section II.E.1.e. of this proposed rule for more discussion of how to participate as a group under MIPS.

(6) Reporting Requirements & Scoring Methodology

(a) Scoring Method

Section 1848(q)(5)(E)(i)(IV) of the Act, as added by section 101(c) of the MACRA, states that 25 percent of the MIPS CPS shall be based on performance for the advancing care information performance category. Therefore, we propose at § 414.1375 that performance in the advancing care information performance category will comprise 25 percent of a MIPS eligible clinician's CPS for payment year 2019 and each year thereafter. We received many comments in the MIPS and APMs RFI from stakeholders regarding the importance of flexible scoring for the advancing care information performance category and provisions for multiple performance pathways. We agree that this is the best approach moving forward with the adoption and use of certified EHR technology as it becomes part of a single coordinated program under the MIPS. For the reasons described here and previously in this preamble, we are proposing a methodology which balances the goals of incentivizing participation and reporting while recognizing exceptional performance by awarding points through a performance score. In this methodology, we are proposing at § 414.1380(b)(4) that the score for the advancing care information performance category would be comprised of a score for participation and reporting, hereinafter referred to as the “base score,” and a score for performance at varying levels above the base score requirements, hereinafter referred to as the “performance score”.

(b) Base Score

To earn points toward the base score, a MIPS eligible clinician must report the numerator and denominator of certain measures specified for the advancing care information performance category (see measure specifications in section II.E.5.g.7 of this proposed rule), which are based on the measures adopted by the EHR Incentive Programs for Stage 3 in the 2015 EHR Incentive Programs Final Rule, to account for 50 percent (out of a total 100 percent) of the advancing care information performance category score. For measures that include a percentage-based threshold for Stage 3 of the EHR Incentive Program, we would not require those thresholds to be met for purposes of the advancing care information performance category under MIPS, but would instead require MIPS eligible clinicians to report the numerator (of at least one) and denominator (or a yes/no statement for applicable measures, which would be submitted together with data for the other measures) for each measure being reported. We note that for any measure requiring a yes/no statement, only a yes statement would qualify for credit under the base score. Under the proposal, the base score of the advancing care information performance category would incorporate the objective and measures adopted by the EHR Incentive Programs with an emphasis on privacy and security. We are proposing two variations of a scoring methodology for the base score, a primary and an alternate proposal, which are outlined below. Both proposals would require the MIPS eligible clinician to meet the requirement to protect patient health information created or maintained by certified EHR technology to earn any score within the advancing care information performance category; failure to do so would result in a base score of zero, a performance score of zero (discussed in section II.E.5.g of this proposed rule), and an advancing care information performance category score of zero.

The primary proposal at section II.E.5.g.6.b.ii. of this proposed rule would require a MIPS eligible clinician to report the numerator (of at least one) and denominator or yes/no statement (only a yes statement would qualify for credit under the base score) for a subset of measures adopted by the EHR Incentive Program for EPs in the 2015 EHR Incentive Programs Final Rule. In an effort to streamline and simplify the reporting requirements under the MIPS, and reduce reporting burden on MIPS eligible clinicians, two objectives (Clinical Decision Support and Computerized Provider Order Entry) and their associated measures would not be required for reporting the advancing care information performance category. Given the consistently high performance on these two objectives in the EHR Incentive Program with EPs accomplishing a median score of over 90 percent for the last 3 years, we believe these objectives and measures are no longer an effective measure of EHR performance and use. In addition, we do not believe these objectives and associated measures contribute to the goals of patient engagement and interoperability, and thus believe these objectives can be removed in an effort to reduce reporting burden without negatively impacting the goals of the advancing care information performance category. We note that the removed objectives and associated measures would still be required as part of ONC's functionality standards for certified EHR technology, however, MIPS eligible clinicians would not be required to report the numerator and denominator or yes/no statement for those measures. In the 2015 EHR Incentive Programs Final Rule we also established that, for measures that were removed, the technology requirements would still be a part of the definition of certified EHR technology. For example, in that final rule, the Stage 1 Objective to Record Demographics was removed, but the technology and standard for this function in the EHR were still required (80 FR 62784). This means that the MIPS eligible clinician would still be required to have these functions as a part of their certified EHR technology.

The alternate proposal at section II.E.5.g.6.b.iii. of this proposed rule would require a MIPS eligible clinician to report the numerator (of at least one) and denominator or yes/no statement (only a yes statement would qualify for credit under the base score) for all objectives and measures adopted for Stage 3 in the 2015 EHR Incentive Programs Final Rule to earn the base score portion of the advancing care information performance category, which would include reporting a yes/no statement for Clinical Decision Support and a numerator and denominator for Computerized Provider Order Entry objectives. We include these objectives in the alternate proposal as MIPS eligible clinicians may feel the continued measurement of these objectives is valuable to the continued use of EHR technology as this would maintain the previously established objectives under the EHR Incentive Program.

We believe both proposed approaches to the base score are consistent with the statutory requirements and previously established certified EHR technology requirements as we transition to MIPS. We also believe both approaches, in conjunction with the advancing care information performance score, recognize the need for greater flexibility in scoring CEHRT use across different clinician types and practice settings by allowing MIPS eligible clinicians to focus on the objectives and measures most applicable to their practice.

(i) Privacy and Security; Protect Patient Health Information

In the 2015 EHR Incentive Programs Final Rule (80 FR 62832), we finalized the Protect Patient Health Information objective and its associated measure for Stage 3, which requires EPs to protect electronic protected health information (ePHI) created or maintained by the certified EHR technology through the implementation of appropriate technical, administrative, and physical safeguards. As privacy and security is of paramount importance and applicable across all objectives, the Protect Patient Health Information objective and measure would be an overarching requirement for the base score under both the primary proposal and alternate proposal, and therefore would be an overarching requirement for the advancing care information performance category. We propose that a MIPS eligible clinician must meet this objective and measure in order to earn any score within the advancing care information performance category. Failure to do so would result in a base score of zero under either the primary proposal or alternate proposal outlined below, as well as a performance score of zero (discussed in section II.E.5.g. of this proposed rule) and an advancing care information performance category score of zero.

(ii) Advancing Care Information Performance Category Base Score Primary Proposal

In the 2015 EHR Incentive Programs Final Rule (80 FR 62829-62871), we finalized certain objectives and measures EPs would report to demonstrate meaningful use of certified EHR technology for Stage 3. Under our proposal for the base score of the advancing care information performance category, MIPS eligible clinicians would be required to submit the numerator (of at least one) and denominator, or yes/no statement as appropriate (only a yes statement would qualify for credit under the base score), for each measure within a subset of objectives (Electronic Prescribing, Patient Electronic Access to Health Information, Care of Coordination Through Patient Engagement, Health Information Exchange, and Public Health and Clinical Data Registry Reporting) adopted in the 2015 EHR Incentive Programs Final Rule for Stage 3 as outlined in Table 6 to account for the base score of 50 percent of the advancing care information performance category score. Successfully submitting a numerator and denominator or yes/no statement for each measure of each objective would earn a base score of 50 percent for the advancing care information performance category. Failure to meet the submission criteria (numerator/denominator or yes/no statement as applicable) and measure specifications (as defined in section II.E.5.g.7. of this proposed rule) for any measure in any of the objectives would result in a score of zero for the advancing care information performance category base score, a performance score of zero (discussed in section II.E.5.g. of this proposed rule) and an advancing care information performance category score of zero.

For the Public Health and Clinical Data Registry Reporting objective there is no numerator and denominator to measure; rather, the measure is a “yes/no” statement of whether the MIPS eligible clinician has completed the measure, noting that only a yes statement would qualify for credit under the base score. Therefore we are proposing that MIPS eligible clinicians would include a yes/no statement in lieu of the numerator/denominator statement within their submission for the advancing care information performance category for the Public Health and Clinical Data Registry Reporting objective. We further propose that, to earn points in the base score, a MIPS eligible clinician would only need to complete submission on the Immunization Registry Reporting measure of this objective. Completing any additional measures under this objective would earn one additional bonus point in the advancing care information performance category score. For further information on this proposed objective, we direct readers to section II.E.5.g.7. of this proposed rule.

(iii) Advancing Care Information Performance Category Base Score Alternate Proposal

Under our alternate proposal for the base score of the advancing care information performance category, a MIPS eligible clinician would be required to submit the numerator (of at least one) and denominator, or yes/no statement as appropriate, for each measure, for all objectives and measures for Stage 3 in the 2015 EHR Incentives Program Final Rule (80 FR 62829-62871) as outlined in Table 7. Successfully submitting a numerator and denominator for each measure of each objective would earn a base score of 50 percent for the advancing care information performance category. Failure to meet the submission requirements, or measure specifications for any measure in any of the objectives would result in a score of zero for the advancing care information performance category base score, a performance score of zero (discussed in Section II.E.5.g.), and an advancing care information performance category score of zero.

We propose the same approach in the alternate proposal for the Public Health and Clinical Data Registry Reporting objective as for the primary proposal outlined above. We direct readers to section II.E.5.g.7. for further details on the individual objectives and measures.

(iv) Modified Stage 2 in 2017

In the 2015 EHR Incentive Programs final rule (80 FR 62772), we streamlined reporting for EPs by adopting a single set of objectives and measures for EPs regardless of their prior stage of participation. This was the first step in synchronizing the objectives and eliminating the separate stages of meaningful use in the EHR Incentive Program. In doing so, we also sought to provide some flexibility and to allow adequate time for EPs to move toward the more advanced use of EHR technology. This flexibility included alternate exclusions and specifications for EPs scheduled to demonstrate Stage 1 in 2015 and 2016 (80 FR 62788) and allowed clinicians to select either the Modified Stage 2 Objectives or the Stage 3 Objectives in 2017 (80 FR 62772) with all EPs moving to the Stage 3 Objectives in 2018. We note that in section II.E.5.g. of this proposed rule, we proposed the requirements for MIPS eligible clinicians using various editions of certified EHR technology in 2017 as it relates to the objectives and measures they select to report.

In connection with that proposal, and in an effort not to unfairly burden MIPS eligible clinicians who are still utilizing EHR technology certified to the 2014 Edition certification criteria in 2017, we propose at § 414.1380(b)(4) modified primary and alternate proposals for the base score for those MIPS eligible clinicians utilizing EHR technology certified to the 2014 Edition. We note that these modified proposals are the same as the primary and alternate proposals outlined above in regard to scoring and data submission, but vary in the measures required under the Coordination of Care Through Patient Engagement and Health Information Exchange objectives as demonstrated in Table 8.

This approach allows MIPS eligible clinicians to continue moving toward advanced use of certified EHR technology in 2018, but allows for flexibility in the implementation of upgraded technology and in the selection of measures for reporting in 2017.

We invite comments on our proposal.

(c) Performance Score

In addition to the base score, which includes submitting each of the objectives and measures in order to achieve 50 percent of the possible points within the advancing care information performance category, we propose to allow multiple paths to achieve a score greater than the 50 percentage base score. The performance score is based on the priority goals established by CMS to focus on leveraging certified EHR technology to support the coordination of care. A MIPS eligible clinician would earn additional points above the base score for performance in the objectives and measures for Patient Electronic Access, Coordination of Care through Patient Engagement, and Health Information Exchange. These measures have a focus on patient engagement, electronic access and information exchange, which promote healthy behaviors by patients and lay the ground work for interoperability. These measures also have significant opportunity for improvement among eligible clinicians and the industry as a whole based on adoption and performance data. We believe this approach for achievement above a base score in the advancing care information performance category would provide MIPS eligible clinicians a flexible and realistic incentive towards the adoption and use of certified EHR technology.

We are proposing at § 414.1380(b)(4) that, for the performance score, the eight associated measures under these three objectives would each be assigned a total of 10 possible points. For each measure, a MIPS eligible clinician may earn up to 10 percent of their performance score based on their performance rate for the given measure. For example, a performance rate of 95 percent on a given measure would earn 9.5 percentage points of the performance score for the advancing care information performance category. This scoring approach is consistent with the performance score approach outlined for other MIPS categories in this proposed rule. Table 9 provides an example of the proposed performance score methodology.

We note that in this methodology, a MIPS eligible clinician has the potential to earn a performance score of up to 80 percent, which, in combination with the base score would be greater than the total possible 100 percent for the advancing care information performance category. This methodology allows flexibility for MIPS eligible clinicians to focus on measures which are most relevant to their practice to achieve the maximum performance category score, while deemphasizing concentration in other measures which are not relevant to their practice.

This proposed methodology recognizes the importance of promoting health IT adoption and standards and the use of certified EHR technology to support quality improvement, interoperability, and patient engagement. We invite comments on our proposal.

(d) Overall Advancing Care Information Performance Category Score

To determine the MIPS eligible clinician's overall advancing care information performance category score, we propose to use the sum of the base score, performance score, and the potential Public Health and Clinical Data Registry Reporting bonus point. We note that if the sum of the MIPS eligible profession's base score (50 percent) and performance score (out of a possible 80 percent) with the Public Health and Clinical Data Registry Reporting bonus point are greater than 100 percent, we would apply an advancing care information performance category score of 100 percent. For example, if the MIPS eligible clinician earned the base score of 50 percent, a performance score of 60 percent and the bonus point for Public Health and Clinical Data Registry Reporting for a total of 111 percent, the MIPS eligible clinician's overall advancing care information performance category score would be 100 percent. The total percentage score (out of 100) for the advancing care information performance category would then be applied to the 25 points allocated for the advancing care information performance category and incorporated into the MIPS CPS, as described in section II.E.6. of this proposed rule. Table 10 provides an example of the calculation of the advancing care information performance category score based on these proposals.

(e) Scoring Considerations

Section 1848(q)(5)(E)(ii) of the Act, as added by section 101(c) of the MACRA, provides that in any year in which the Secretary estimates that the proportion of EPs (as defined in section 1848(o)(5) of the Act) who are meaningful EHR users (as determined under section 1848(o)(2) of the Act) is 75 percent or greater, the Secretary may reduce the applicable percentage weight of the advancing care information performance category in the MIPS CPS, but not below 15 percent, and increase the weightings of the other performance categories such that the total percentage points of the increase equals the total percentage points of the reduction. We note section 1848(o)(5) of the Act defines an EP as a physician, as defined in section 1861(r) of the Act. For purposes of applying section 1848(q)(5)(E)(ii) of the Act, we propose to estimate the proportion of physicians as defined in section 1861(r) who are meaningful EHR users as those physician MIPS eligible clinicians who earn an advancing care information performance category score of at least 75 percent under our proposed scoring methodology for the advancing care information performance category for a performance period. This would require the MIPS eligible clinician to earn the advancing care information base score of 50 percent, and an advancing care information performance score of at least 25 percent (or 24 percent plus the Public Health and Clinical Data Registry Reporting bonus point) for an overall performance category score of 75 percent for the advancing care information performance category. We are alternatively proposing to estimate the proportion of physicians as defined in section 1861(r) who are meaningful EHR users as those physician MIPS eligible clinicians who earn an advancing care information performance category score of 50 percent (which would only require the MIPS eligible clinician to earn the advancing care information base score) under our proposed scoring methodology for the advancing care information performance category for a performance period, and we seek comments on both of these proposed thresholds.

We propose to base this estimation on data from the relevant performance period, if we have sufficient data available from that period. For example, if feasible, we would consider whether to reduce the applicable percentage weight of the advancing care information performance category in the MIPS CPS for the 2019 MIPS payment year based on an estimation using the data from the 2017 performance period. We note that in section II.E.5.g.8. of this proposed rule, we have proposed to reweight the advancing care information performance category to zero for certain hospital-based physicians and other physicians. These physicians meet the definition of MIPS eligible clinicians, but would not be included in the estimation because the advancing care information performance category would be weighted at zero for them. We note that any adjustments of the performance category weights specified in section 1848(q)(5)(E) of the Act based on this policy would be established in future notice and comment rulemaking.

We invite comments on our proposals.

(7) Advancing Care Information Performance Category Objectives and Measures Specifications

(a) MIPS Objectives and Measures Specifications

We propose the objectives and measures for the advancing care information performance category of MIPS as outlined in this section of the proposed rule. We note that these objectives and measures have been adapted from the Stage 3 objectives and measures as finalized in the 2015 EHR Incentive Programs Final Rule (80 FR 62829-62871), however, we have not proposed to maintain the previously established thresholds for MIPS. Any additional changes to the objectives and measures are outlined in this section of the proposed rule. For a more detailed discussion of the Stage 3 objectives and measures, including explanatory material and defined terms, we refer readers to the 2015 EHR Incentive Programs Final Rule (80 FR 62829-62871).

Objective: Protect Patient Health Information

Objective: Protect electronic protected health information (ePHI) created or maintained by the certified EHR technology through the implementation of appropriate technical, administrative, and physical safeguards

Security Risk Analysis Measure: Conduct or review a security risk analysis in accordance with the requirements in 45 CFR 164.308(a)(1), including addressing the security (to include encryption) of ePHI data created or maintained by certified EHR technology in accordance with requirements in 45 CFR164.312(a)(2)(iv) and 45 CFR 164.306(d)(3), and implement security updates as necessary and correct identified security deficiencies as part of the MIPS eligible clinician's risk management process.

Objective: Electronic Prescribing

Objective: MIPS eligible clinicians must generate and transmit permissible prescriptions electronically.

ePrescribing Measure: At least one permissible prescription written by the MIPS eligible clinician is queried for a drug formulary and transmitted electronically using certified EHR technology.

• Denominator: Number of prescriptions written for drugs requiring a prescription in order to be dispensed other than controlled substances during the performance period; or number of prescriptions written for drugs requiring a prescription in order to be dispensed during the performance period.
• Numerator: The number of prescriptions in the denominator generated, queried for a drug formulary, and transmitted electronically using certified EHR technology.

For this objective, we note that the 2015 EHR Incentive Program final rule included a discussion of controlled substances in the context of the Stage 3 objective and measure (80 FR 62834), which we understand from stakeholders has caused confusion. We are therefore proposing for both MIPS and for the EHR Incentive Programs that health care providers would continue to have the option to include or not include controlled substances that can be electronically prescribed in the denominator. This means that health care providers may choose to include controlled substances in the definition of “permissible prescriptions” at their discretion where feasible and allowable by law in the jurisdiction where they provide care. The health care provider may also choose not to include controlled substances in the definition of “permissible prescriptions” even if such electronic prescriptions are feasible and allowable by law in the jurisdiction where they provide care.

Objective: Clinical Decision Support (Alternate Proposal Only)

Objective: Implement clinical decision support (CDS) interventions focused on improving performance on high-priority health conditions

Clinical Decision Support (CDS) Interventions Measure: Implement three clinical decision support interventions related to three CQMs at a relevant point in patient care for the entire performance period. Absent three CQMs related to a MIPS eligible clinician's scope of practice or patient population, the clinical decision support interventions must be related to high-priority health conditions.

Drug Interaction and Drug-Allergy Checks Measure: The MIPS eligible clinician has enabled and implemented the functionality for drug-drug and drug-allergy interaction checks for the entire performance period.

Objective: Computerized Provider Order Entry (Alternate Proposal Only)

Objective: Use computerized provider order entry (CPOE) for medication, laboratory, and diagnostic imaging orders directly entered by any licensed healthcare professional, credentialed medical assistant, or a medical staff member credentialed to and performing the equivalent duties of a credentialed medical assistant, who can enter orders into the medical record per state, local, and professional guidelines.

Medication Orders Measure: At least one medication order created by the MIPS eligible clinician during the performance period is recorded using CPOE.

• Denominator: Number of medication orders created by the MIPS eligible clinician during the performance period.
• Numerator: The number of orders in the denominator recorded using CPOE.

Laboratory Orders Measure: At least one laboratory order created by the MIPS eligible clinician during the performance period is recorded using CPOE.

• Denominator: Number of laboratory orders created by the MIPS eligible clinician during the performance period.
• Numerator: The number of orders in the denominator recorded using CPOE.

Diagnostic Imaging Orders Measure: At least one diagnostic imaging order created by the MIPS eligible clinician during the performance period is recorded using CPOE.

• Denominator: Number of diagnostic imaging orders created by the MIPS eligible clinician during the performance period.
• Numerator: The number of orders in the denominator recorded using CPOE.

Objective: Patient Electronic Access.

Objective: The MIPS eligible clinician provides patients (or patient authorized representative) with timely electronic access to their health information and patient-specific education.

Patient Access Measure: For at least one unique patient seen by the MIPS eligible clinician: (1) The patient (or the patient authorized representative) is provided timely access to view online, download, and transmit his or her health information; and (2) The MIPS eligible clinician ensures the patient's health information is available for the patient (or patient—authorized representative) to access using any application of their choice that is configured to meet the technical specifications of the Application Programing Interface (API) in the MIPS eligible clinician's certified EHR technology.

• Denominator: The number of unique patients seen by the MIPS eligible clinician during the performance period.
• Numerator: The number of patients in the denominator (or patient authorized representative) who are provided timely access to health information to view online, download, and transmit to a third party and to access using an application of their choice that is configured meet the technical specifications of the API in the MIPS eligible clinician's certified EHR technology.

Patient-Specific Education Measure: The MIPS eligible clinician must use clinically relevant information from certified EHR technology to identify patient-specific educational resources and provide electronic access to those materials to at least one unique patient seen by the MIPS eligible clinician.

• Denominator: The number of unique patients seen by the MIPS eligible clinician during the performance period.
• Numerator: The number of patients in the denominator who were provided electronic access to patient-specific educational resources using clinically relevant information identified from certified EHR technology during the performance period.

Objective: Coordination of Care Through Patient Engagement.

Objective: Use certified EHR technology to engage with patients or their authorized representatives about the patient's care.

View, Download, Transmit (VDT) Measure: During the performance period, at least one unique patient (or patient-authorized representatives) seen by the MIPS eligible clinician actively engages with the EHR made accessible by the MIPS eligible clinician. An MIPS eligible clinician may meet the measure by either—(1) view, download or transmit to a third party their health information; or (2) access their health information through the use of an API that can be used by applications chosen by the patient and configured to the API in the MIPS eligible clinician's certified EHR technology; or (3) a combination of (1) and (2).

• Denominator: Number of unique patients seen by the MIPS eligible clinician during the performance period.
• Numerator: The number of unique patients (or their authorized representatives) in the denominator who have viewed online, downloaded, or transmitted to a third party the patient's health information during the performance period and the number of unique patients (or their authorized representatives) in the denominator who have accessed their health information through the use of an API during the performance period.

Secure Messaging Measure: For at least one unique patient seen by the MIPS eligible clinician during the performance period, a secure message was sent using the electronic messaging function of certified EHR technology to the patient (or the patient-authorized representative), or in response to a secure message sent by the patient (or the patient-authorized representative).

• Denominator: Number of unique patients seen by the MIPS eligible clinician during the performance period.
• Numerator: The number of patients in the denominator for whom a secure electronic message is sent to the patient (or patient-authorized representative) or in response to a secure message sent by the patient (or patient-authorized representative), during the performance period.

Patient-Generated Health Data Measure: Patient-generated health data or data from a non-clinical setting is incorporated into the certified EHR technology for at least one unique patient seen by the MIPS eligible clinician during the performance period.

• Denominator: Number of unique patients seen by the MIPS eligible clinician during the performance period.
• Numerator: The number of patients in the denominator for whom data from non-clinical settings, which may include patient-generated health data, is captured through the certified EHR technology into the patient record during the performance period.

Objective: Health Information Exchange.

Objective: The MIPS eligible clinician provides a summary of care record when transitioning or referring their patient to another setting of care, receives or retrieves a summary of care record upon the receipt of a transition or referral or upon the first patient encounter with a new patient, and incorporates summary of care information from other health care providers into their EHR using the functions of certified EHR technology.

Patient Care Record Exchange Measure: For at least one transition of care or referral, the MIPS eligible clinician that transitions or refers their patient to another setting of care or health care provider—(1) creates a summary of care record using certified EHR technology; and (2) electronically exchanges the summary of care record.

• Denominator: Number of transitions of care and referrals during the performance period for which the MIPS eligible clinician was the transferring or referring clinician.
• Numerator: The number of transitions of care and referrals in the denominator where a summary of care record was created using certified EHR technology and exchanged electronically.

Request/Accept Patient Care Record Measure: For at least one transition of care or referral received or patient encounter in which the MIPS eligible clinician has never before encountered the patient, the MIPS eligible clinician receives or retrieves and incorporates into the patient's record an electronic summary of care document.

• Denominator: Number of patient encounters during the performance period for which a MIPS eligible clinician was the receiving party of a transition or referral or has never before encountered the patient and for which an electronic summary of care record is available.
• Numerator: Number of patient encounters in the denominator where an electronic summary of care record received is incorporated by the clinician into the certified EHR technology.

Clinical Information Reconciliation Measure: For at least one transition of care or referral received or patient encounter in which the MIPS eligible clinician has never before encountered the patient, the MIPS eligible clinician performs clinical information reconciliation. The clinician must implement clinical information reconciliation for the following three clinical information sets: (1) Medication. Review of the patient's medication, including the name, dosage, frequency, and route of each medication. (2) Medication allergy. Review of the patient's known medication allergies. (3) Current Problem list. Review of the patient's current and active diagnoses.

• Denominator: Number of transitions of care or referrals during the performance period for which the MIPS eligible clinician was the recipient of the transition or referral or has never before encountered the patient.
• Numerator: The number of transitions of care or referrals in the denominator where the following three clinical information reconciliations were performed: Medication list, medication allergy list, and current problem list.

Objective: Public Health and Clinical Data Registry Reporting

Objective: The MIPS eligible clinician is in active engagement with a public health agency or clinical data registry to submit electronic public health data in a meaningful way using certified EHR technology, except where prohibited, and in accordance with applicable law and practice.

Immunization Registry Reporting Measure: The MIPS eligible clinician is in active engagement with a public health agency to submit immunization data and receive immunization forecasts and histories from the public health immunization registry/immunization information system (IIS).

(Optional) Syndromic Surveillance Reporting Measure: The MIPS eligible clinician is in active engagement with a public health agency to submit syndromic surveillance data from a non-urgent care ambulatory setting where the jurisdiction accepts syndromic data from such settings and the standards are clearly defined.

(Optional) Electronic Case Reporting Measure: The MIPS eligible clinician is in active engagement with a public health agency to electronically submit case reporting of reportable conditions.

(Optional) Public Health Registry Reporting Measure: The MIPS eligible clinician is in active engagement with a public health agency to submit data to public health registries.

(Optional) Clinical Data Registry Reporting Measure: The MIPS eligible clinician is in active engagement to submit data to a clinical data registry.

(b) Modified Stage 2 Advancing Care Information Objectives and Measures Specifications for MIPS

We propose the Modified Stage 2 objectives and measures for the advancing care information performance category of MIPS as outlined in this section of the proposed rule. We note that these objectives and measures have been adapted from the Modified Stage 2 objectives and measures as finalized in the 2015 EHR Incentive Programs Final Rule (80 FR 62793—62825), however, we have not proposed to maintain the previously established thresholds for MIPS. Any additional changes to the objectives and measures are outlined in this section of the proposed rule. For a more detailed discussion of the Modified Stage 2 objectives and measures, including explanatory material and defined terms, we refer readers to the 2015 EHR Incentive Programs Final Rule (80 FR 62793—62825).

Objective: Protect Patient Health Information

Objective: Protect electronic protected health information (ePHI) created or maintained by the certified EHR technology through the implementation of appropriate technical, administrative, and physical safeguards.

Security Risk Analysis Measure: Conduct or review a security risk analysis in accordance with the requirements in 45 CFR 164.308(a)(1), including addressing the security (to include encryption) of ePHI data created or maintained by certified EHR technology in accordance with requirements in 45 CFR164.312(a)(2)(iv) and 45 CFR 164.306(d)(3), and implement security updates as necessary and correct identified security deficiencies as part of the MIPS eligible clinician's risk management process.

Objective: Electronic Prescribing

Objective: MIPS eligible clinicians must generate and transmit permissible prescriptions electronically.

ePrescribing Measure: At least one permissible prescription written by the MIPS eligible clinician is queried for a drug formulary and transmitted electronically using certified EHR technology.

• Denominator: Number of prescriptions written for drugs requiring a prescription in order to be dispensed other than controlled substances during the performance period; or number of prescriptions written for drugs requiring a prescription in order to be dispensed during the performance period.
• Numerator: The number of prescriptions in the denominator generated, queried for a drug formulary, and transmitted electronically using certified EHR technology.

Objective: Clinical Decision Support (alternate proposal only)

Objective: Implement clinical decision support (CDS) interventions focused on improving performance on high-priority health conditions.

Clinical Decision Support (CDS) Interventions Measure: Implement three clinical decision support interventions related to three CQMs at a relevant point in patient care for the entire performance period. Absent three CQMs related to a MIPS eligible clinician's scope of practice or patient population, the clinical decision support interventions must be related to high-priority health conditions.

Drug Interaction and Drug-Allergy Checks Measure: The MIPS eligible clinician has enabled and implemented the functionality for drug-drug and drug-allergy interaction checks for the entire performance period.

Objective: Computerized Provider Order Entry

Objective: Use computerized provider order entry (CPOE) for medication, laboratory, and diagnostic imaging orders directly entered by any licensed healthcare professional, credentialed medical assistant, or a medical staff member credentialed to and performing the equivalent duties of a credentialed medical assistant, who can enter orders into the medical record per state, local, and professional guidelines.

Medication Orders Measure: At least one medication order created by the MIPS eligible clinician during the performance period is recorded using CPOE.

• Denominator: Number of medication orders created by the MIPS eligible clinician during the performance period.
• Numerator: The number of orders in the denominator recorded using CPOE.

Laboratory Orders Measure: At least one laboratory order created by the MIPS eligible clinician during the performance period is recorded using CPOE.

• Denominator: Number of laboratory orders created by the MIPS eligible clinician during the performance period.
• Numerator: The number of orders in the denominator recorded using CPOE.

Diagnostic Imaging Orders Measure: At least one diagnostic imaging order created by the MIPS eligible clinician during the performance period is recorded using CPOE.

• Denominator: Number of diagnostic imaging orders created by the MIPS eligible clinician during the performance period.
• Numerator: The number of orders in the denominator recorded using CPOE.

Objective: Patient Electronic Access

Objective: The MIPS eligible clinician provides patients (or patient authorized representative) with timely electronic access to their health information and patient-specific education.

Patient Access Measure: At least one patient seen by the MIPS eligible clinician during the performance period is provided timely access to view online, download, and transmit to a third party their health information subject to the MIPS eligible clinician's discretion to withhold certain information.

• Denominator: The number of unique patients seen by the MIPS eligible clinician during the performance period.
• Numerator: The number of patients in the denominator (or patient authorized representative) who are provided timely access to health information to view online, download, and transmit to a third party.

View, Download, Transmit (VDT) Measure: At least one patient seen by the MIPS eligible clinician during the performance period (or patient-authorized representative) views, downloads or transmits their health information to a third party during the performance period.

• Denominator: Number of unique patients seen by the MIPS eligible clinician during the performance period.
• Numerator: The number of unique patients (or their authorized representatives) in the denominator who have viewed online, downloaded, or transmitted to a third party the patient's health information during the performance period.

Objective: Patient-Specific Education

Objective: The MIPS eligible clinician provides patients (or patient authorized representative) with timely electronic access to their health information and patient-specific education.

Patient-Specific Education Measure: The MIPS eligible clinician must use clinically relevant information from certified EHR technology to identify patient-specific educational resources and provide access to those materials to at least one unique patient seen by the MIPS eligible clinician.

• Denominator: The number of unique patients seen by the MIPS eligible clinician during the performance period.
• Numerator: The number of patients in the denominator who were provided access to patient-specific educational resources using clinically relevant information identified from certified EHR technology during the performance period.

Objective: Secure Messaging

Objective: Use certified EHR technology to engage with patients or their authorized representatives about the patient's care.

Secure Messaging Measure: For at least one patient seen by the MIPS eligible clinician during the performance period, a secure message was sent using the electronic messaging function of certified EHR technology to the patient (or the patient-authorized representative), or in response to a secure message sent by the patient (or the patient authorized representative) during the performance period.

• Denominator: Number of unique patients seen by the MIPS eligible clinician during the performance period.
• Numerator: The number of patients in the denominator for whom a secure electronic message is sent to the patient (or patient-authorized representative) or in response to a secure message sent by the patient (or patient-authorized representative), during the performance period.

Objective: Health Information Exchange

Objective: The MIPS eligible clinician provides a summary of care record when transitioning or referring their patient to another setting of care, receives or retrieves a summary of care record upon the receipt of a transition or referral or upon the first patient encounter with a new patient, and incorporates summary of care information from other health care providers into their EHR using the functions of certified EHR technology.

Health Information Exchange Measure: The MIPS eligible clinician that transitions or refers their patient to another setting of care or health care provider (1) uses certified EHR technology to create a summary of care record; and (2) electronically transmits such summary to a receiving health care provider for at least one transition of care or referral.

• Denominator: Number of transitions of care and referrals during the performance period for which the EP was the transferring or referring health care provider.
• Numerator: The number of transitions of care and referrals in the denominator where a summary of care record was created using certified EHR technology and exchanged electronically.

Objective: Medication Reconciliation

Medication Reconciliation Measure: The MIPS eligible clinician performs medication reconciliation for at least one transition of care in which the patient is transitioned into the care of the MIPS eligible clinician.

• Denominator: Number of transitions of care or referrals during the performance period for which the MIPS eligible clinician was the recipient of the transition or referral or has never before encountered the patient.
• Numerator: The number of transitions of care or referrals in the denominator where the following three clinical information reconciliations were performed: Medication list, medication allergy list, and current problem list.

Objective: Public Health Reporting

Objective: The MIPS eligible clinician is in active engagement with a public health agency or clinical data registry to submit electronic public health data in a meaningful way using certified EHR technology, except where prohibited, and in accordance with applicable law and practice.

Immunization Registry Reporting Measure: The MIPS eligible clinician is in active engagement with a public health agency to submit immunization data.

Syndromic Surveillance Registry Reporting Measure: The MIPS eligible clinician is in active engagement with a public health agency to submit syndromic surveillance data.

Specialized Registry Reporting Measure: The MIPS eligible clinician is in active engagement to submit data to a specialized registry.

We invite comments on our proposal.

(c) Exclusions

In the 2015 EHR Incentive Programs Final Rule (80 FR 62829-62871) we outlined certain exclusions from the objectives and measures of meaningful use for EPs who perform low numbers of a particular action or activity for a given measure (for example, an EP who writes fewer than 100 permissible prescriptions during the EHR reporting period would be granted an exclusion for the Electronic Prescribing measure) or for EPs who had no office visits during the EHR reporting period. Moving forward, we believe that the proposed MIPS exclusion criteria as outlined in section II.E.3. of this proposed rule, and advancing care information performance category scoring methodology together accomplish the same end as the previously established exclusions for the majority of the advancing care information measures. By excluding from MIPS those clinicians who do not exceed the low-volume threshold (proposed in section II.E.3.c. as MIPS eligible clinicians who, during the performance period, have Medicare billing charges less than or equal to $10,000 and provide care for 100 or fewer Part B-enrolled Medicare beneficiaries), we believe exclusions for most of the individual advancing care information measures are no longer necessary. The additional flexibility afforded by the proposed advancing care information performance category scoring methodology eliminates required thresholds for measures and allows MIPS eligible clinicians to focus on, and therefore report higher numbers for, measures that are more relevant to their practice.

We note that EPs who write less than 100 permissible prescriptions during the EHR reporting period are allowed an exclusion for the Electronic Prescribing measure under the EHR Incentive Program (80 FR 62834), which we do not propose for MIPS. We note that the Electronic Prescribing objective would not be part of the performance score under our proposals, and thus MIPS eligible clinicians who write very low numbers of permissible prescriptions would not be at a disadvantage in relation to other MIPS eligible clinicians when seeking to achieve a maximum advancing care information performance category score. For the purposes of the base score, we are proposing that those MIPS eligible clinicians who write fewer than 100 permissible prescriptions in a performance period may elect to report their numerator and denominator (if they have at least one permissible prescription for the numerator), or they may report a null value. This is consistent with prior policy which allowed flexibility for clinicians in similar circumstances to choose an alternate exclusion (80 FR 62789).

In addition, in the 2015 EHR Incentive Programs final rule, we adopted a set of exclusions for the Immunization Registry Reporting measure under the Public Health and Clinical Data Registry Reporting objective (80 FR 62870). We recognize that some types of clinicians do not administer immunizations, and are therefore proposing to maintain the previously established exclusions for the Immunization Registry Reporting measure. We are therefore proposing that these MIPS eligible clinicians may elect to report their yes/no statement if applicable, or they may report a null value (if the previously established exclusions apply) for purposes of reporting the base score.

We note that we are not proposing to maintain any of the other exclusions established under the EHR Incentive Program, however, we are seeking comment on whether other exclusions should be considered under the advancing care information performance category under the MIPS.

(8) Additional Considerations

(a) Reweighting of the Advancing Care Information Performance Category for MIPS Eligible Clinicians Without Sufficient Measures Applicable and Available

As discussed previously in this proposed rule, section 101(b)(1)(A) of the MACRA amended section 1848(a)(7)(A) of the Act to sunset the meaningful use payment adjustment at the end of CY 2018. Section 1848(a)(7) of the Act includes certain statutory exceptions to the meaningful use payment adjustment under section 1848(a)(7)(A) of the Act. Specifically, section 1848(a)(7)(D) of the Act exempts hospital-based EPs from the application of the payment adjustment under section 1848(a)(7)(A) of the Act. In addition, section 1848(a)(7)(B) of the Act provides that the Secretary may exempt an EP who is not a meaningful EHR user for the EHR reporting period for the year from the application of the payment adjustment under section 1848(a)(7)(A) of the Act if the Secretary determines that compliance with the requirements for being a meaningful EHR user would result in a significant hardship, such as in the case of an EP who practices in a rural area without sufficient internet access. The MACRA did not maintain these statutory exceptions for the advancing care information performance category of the MIPS. Thus, the exceptions under sections 1848(a)(7)(B) and (D) of the Act are limited to the meaningful use payment adjustment under section 1848(a)(7)(A) of the Act and do not apply in the context of the MIPS.

Section 1848(q)(5)(F) of the Act provides, if there are not sufficient measures and activities applicable and available to each type of MIPS eligible clinician, the Secretary shall assign different scoring weights (including a weight of zero) for each performance category based on the extent to which the category is applicable to each type of MIPS eligible clinician, and for each measure and activity specified for each such category based on the extent to which the measure or activity is applicable and available to the type of MIPS eligible clinician.

We believe that under our proposals for the advancing care information performance category of the MIPS, there may not be sufficient measures that are applicable and available to certain types of MIPS eligible clinicians as outlined in this section of this proposed rule, some of whom may have qualified for a statutory exception to the meaningful use payment adjustment under section 1848(a)(7)(A) of the Act. For the reasons stated below, we propose to assign a weight of zero to the advancing care information performance category for purposes of calculating a MIPS CPS for these MIPS eligible clinicians. We refer readers to section II.E.6. of this proposed rule for more information regarding how the quality, resource use and CPIA performance categories would be reweighted.

(i) Hospital-Based MIPS Eligible Clinicians

Section 1848(a)(7)(D) of the Act exempts hospital-based EPs from the application of the meaningful use payment adjustment under section 1848(a)(7)(A) of the Act. We defined a hospital-based EP for the EHR Incentive Program under § 495.4 as an EP who furnishes 90 percent or more of his or her covered professional services in sites of service identified by the codes used in the HIPAA standard transaction as an inpatient hospital or emergency room setting in the year preceding the payment year, or in the case of a payment adjustment year, in either of the 2 years before the year preceding such payment adjustment year. Under this definition, EPs that have 90 percent or more of payments for covered professional services associated with claims with Place of Service Codes 21 (inpatient hospital) or 23 (emergency department) are considered hospital-based (75 FR 44442).

We believe there may not be sufficient measures applicable and available to hospital-based MIPS eligible clinicians under our proposals for the advancing care information performance category of MIPS.

Hospital-based MIPS eligible clinicians may not have control over the decisions that the hospital makes regarding the use of health IT and certified EHR technology. These MIPS eligible clinicians therefore may have no control over the type of certified EHR technology available, the way that the technology is implemented and used, or whether the hospital continually invests in the technology to ensure it is compliant with ONC certification criteria. In addition, some of the specific advancing care information performance category measures, such as the Patient Access measure under the Patient Electronic Access objective requires that patients have access to view, download and transmit their health information from the EHR which is made available by the health care provider, in this case the hospital. Thus the measure is more attributable and applicable to the hospital and not to the MIPS eligible clinician, as the hospital controls the availability of the EHR technology. Further, the requirement under the Protect Patient Health Information objective to conduct a security risk analysis, would rely on the actions of the hospital, rather than the actions of the MIPS eligible clinician, as the hospital controls the access and availability and secure implementation of the EHR technology. In this case, the measure is again more attributable and applicable to the hospital than to the MIPS eligible clinician. Further, certain specialists (such as pathologists, radiologists and anesthesiologists) who often practice in a hospital setting and may be hospital-based MIPS eligible clinicians often lack face-to-face interaction with patients, and thus may not have sufficient measures applicable and available to them under our proposals. For example, hospital-based MIPS eligible clinicians who lack face-to-face patient interaction may not have patients for which they could transfer or create an electronic summary of care record.

In addition, we note that eligible hospitals and CAHs are subject to meaningful use requirements under sections 1886(b)(3)(B) and (n) and 1814(l) of the Act, respectively, which were not affected by the enactment of the MACRA. Eligible hospitals and CAHs are required to report on objectives and measures of meaningful use under the EHR Incentive Program, as outlined in the 2015 EHR Incentive Programs Final Rule. We note the objectives and measures of the EHR Incentive Programs for eligible hospitals and CAHs are specific to these facilities, and are more applicable and better represent the EHR technology available in these settings.

For these reasons, we propose to rely on section 1848(q)(5)(F) of the Act to assign a weight of zero to the advancing care information performance category for hospital-based MIPS eligible clinicians. We propose to define a “hospital-based MIPS eligible clinician” at § 414.1305 as a MIPS eligible clinician who furnishes 90 percent or more of his or her covered professional services in sites of service identified by the codes used in the HIPAA standard transaction as an inpatient hospital or emergency room setting in the year preceding the performance period, otherwise stated as the year three years preceding the MIPS payment year. For example, under this proposal, hospital-based determinations would be made for the 2019 MIPS payment year based on covered professional services furnished in 2016. We also propose, consistent with the EHR Incentive Program, that CMS would determine which MIPS eligible clinicians qualify as “hospital-based” for a MIPS payment year. We invite comments on these proposals.

In addition, we are seeking comment on how the advancing care information performance category could be applied to hospital-based MIPS eligible clinicians in future years of MIPS, and the types of measures that would be applicable and available to these types of MIPS eligible clinicians.

We are also seeking comment on whether the previously established 90 percent threshold of payments for covered professional services associated with claims with Place of Service (POS) Codes 21 (inpatient hospital) or 23 (emergency department) is appropriate, or whether we should consider lowering this threshold to account for hospital-based MIPS eligible clinicians who bill more than 10 percent of claims with a POS other than 21 or 23. Although we have proposed a threshold of 90 percent, we are considering whether a lower threshold would be more appropriate for hospital-based MIPS eligible clinicians. In particular, we are interested in what factors should be applied to determine the threshold for hospital-based MIPS eligible clinicians. We will continue to evaluate the data to determine whether there are certain thresholds which naturally define a hospital-based MIPS eligible clinician.

(ii) MIPS Eligible Clinicians Facing a Significant Hardship

Section 1848(a)(7)(B) of the Act provides that the Secretary may exempt an EP who is not a meaningful EHR user for the EHR reporting period for the year from the application of the payment adjustment under section 1848(a)(7)(A) of the Act if the Secretary determines that compliance with the requirements for being a meaningful EHR user would result in a significant hardship. In the Stage 2 Final Rule (77 FR 54097-54100), we defined certain categories of significant hardships that may prevent an EP from meeting the requirements of being a meaningful EHR user. These categories include:

• Insufficient Internet Connectivity (as specified in 42 CFR 495.102(d)(4)(i)).
• Extreme and Uncontrollable Circumstances (as specified in 42 CFR 495.102(d)(4)(iii)).
• Lack of Control over the Availability of certified EHR technology (as specified in 42 CFR 495.102(d)(4)(iv)(A)).
• Lack of Face-to-Face Patient Interaction (as specified in 42 CFR 495.102(d)(4)(iv)(B)).

We believe that under our proposals for the advancing care information performance category, there may not be sufficient measures applicable and available to MIPS eligible clinicians within the categories above. For these MIPS eligible clinicians, we propose to rely on section 1848(q)(5)(F) of the Act to re-weight the advancing care information performance category to zero.

Sufficient internet access is fundamental to many of the measures proposed for the advancing care information performance category. For example, the ePrescribing measure requires sufficient access to the Internet to transmit prescriptions electronically, and the Secure Messaging measure requires sufficient Internet access to receive and respond to patient messages. These measures may not be applicable to MIPS eligible clinicians who practice in areas with insufficient internet access. We propose to require MIPS eligible clinicians to demonstrate insufficient internet access through an application process in order to be considered for a reweighting of the advancing care information performance category. The application would have to demonstrate that the MIPS eligible clinicians lacked sufficient internet access, during the performance period, and that there were insurmountable barriers to obtaining such infrastructure, such as a high cost of extending the internet infrastructure to their facility.

Extreme and uncontrollable circumstances, such as a natural disaster in which an EHR or practice building are destroyed, can happen at any time and are outside a MIPS eligible clinician's control. If a MIPS eligible clinician's certified EHR technology is unavailable as a result of such circumstances, the measures specified for the advancing care information performance category may not be available for the MIPS eligible clinician to report. We propose that these MIPS eligible clinicians submit an application to include the circumstances by which the EHR technology was unavailable, and for what period of time it was unavailable, to be considered for reweighting of their advancing care information performance category.

In the Stage 2 Final Rule (77 FR 54100) we discussed EPs who practice at multiple locations, and may not have the ability to impact their practices' health IT decisions. We noted the case of surgeons using ambulatory surgery centers or a physician treating patients in a nursing home who does not have any other vested interest in the facility, and may have no influence or control over the health IT decisions of that facility. If MIPS eligible clinicians lack control over the EHR technology in their practice locations, then the measures specified for the advancing care information performance category may not be available to them for reporting. To be considered for a reweighting of the advancing care information performance category, we propose that these MIPS eligible clinicians would need to submit an application demonstrating that a majority (50 percent or more) of their outpatient encounters occur in locations where they have no control over the health IT decisions of the facility, and request their advancing care information performance category score be reweighted to zero. We note that in such cases, the MIPS eligible clinician must have no control over the availability of certified EHR technology. Control does not imply final decision-making authority. For example, we would generally view MIPS eligible clinicians practicing in a large group as having control over the availability of certified EHR technology, because they can influence the group's purchase of certified EHR technology, they may reassign their claims to the group, they may have a partnership/ownership stake in the group, or any payment adjustment would affect the group's earnings and the entire impact of the adjustment would not be borne by the individual MIPS eligible clinician. These MIPS eligible clinicians can influence the availability of certified EHR technology and the group's earnings are directly affected by the payment adjustment. Thus, such MIPS eligible clinicians would not, as a general rule, be viewed as lacking control over the availability of certified EHR technology and would not be eligible for their advancing care information performance category to be reweighted based on their membership in a group practice that has not adopted certified EHR technology.

In the Stage 2 Final Rule (77 FR 54099), we noted the challenges faced by EPs who lack face-to-face interaction with patients (EPs that are non-patient facing), or lack the need to provide follow-up care with patients. Many of the measures proposed under the advancing care information performance category require face-to-face interaction with patients, including all eight of the measures that make up the three performance score objectives (Patient Electronic Access, Coordination of Care Through Patient Engagement and Health Information Exchange). Because these proposed measures rely so heavily on face-to-face patient interactions, we do not believe there would be sufficient measures applicable to non-patient-facing MIPS eligible clinicians under the advancing care information performance category. We propose to automatically reweight the advancing care information performance category to zero for a MIPS eligible clinician who is classified as a non-patient facing MIPS eligible clinician (based on the number of patient-facing encounters billed during a performance period) without requiring an application to be submitted by the MIPS eligible clinician. We refer readers to section II.E.1.b. of this proposed rule for further discussion of non-patient facing MIPS eligible clinicians. We are seeking comment on how the advancing care information performance category could be applied to non-patient facing MIPS eligible clinicians in future years of MIPS, and the types of measures that would be applicable and available to these types of MIPS eligible clinicians.

We propose that all applications for reweighting the advancing care information performance category be submitted by the MIPS eligible clinician or designated group representative in the form and manner specified by CMS. We propose that all applications may be submitted on a rolling basis, but must be received by CMS no later than the close of the submission period for the relevant performance period, or a later date specified by CMS. For example, for the 2017 performance period, applications must be submitted no later than March 31, 2018 (or later date as specified by CMS) to be considered for reweighting the advancing care information performance category for the 2019 MIPS payment year. An application would need to be submitted annually to be considered for reweighting each year.

We invite comments on our proposals.

(iii) Nurse Practitioners, Physician Assistants, Clinical Nurse Specialists, and Certified Registered Nurse Anesthetists

The definition of a MIPS EP under section 1848(q)(1)(C) of the Act includes certain non-physician practitioners, including Nurse Practitioners (NPs), Physicians Assistants (PAs), Certified Registered Nurse Anesthetists (CRNAs) and Clinical Nurse Specialists (CNSs)). CRNAs and CNSs are not eligible for the incentive payments under Medicare or Medicaid for the adoption and meaningful use of certified EHR technology (sections 1848(o) and 1903(t) of the Act, respectively) or subject to the meaningful use payment adjustment under Medicare (section 1848(a)(7)(A) of the Act), and thus they may have little to no experience with the adoption or use of certified EHR technology. Similarly, NPs and PAs may also lack experience with the adoption or use of certified EHR technology, as they are not subject to the payment adjustment under section 1848(a)(7)(A) of the Act. We further note that only 19,281 NPs and only 1,379 PAs have attested to the Medicaid EHR Incentive Program. Nurse practitioners are eligible for the Medicaid incentive payments under section 1903(t) of the Act, as are PAs practicing in a Federally Qualified Health Center (FQHC) or a rural health clinic (RHC) that is led by a PA, if they meet patient volume requirements and other eligibility criteria.

Because many of these non-physician clinicians are not eligible to participate in the Medicare and/or Medicaid EHR Incentive Program, we have little evidence as to whether there are sufficient measures applicable and available to these types of MIPS eligible clinicians under our proposals for the advancing care information performance category. The low numbers of NPs and PAs who have attested for the Medicaid incentive payments may indicate that EHR Incentive Program measures required to earn the incentive are not applicable or available, and thus would not be applicable or available under the advancing care information performance category. For these reasons, we propose to rely on section 1848(q)(5)(F) of the Act to assign a weight of zero to the advancing care information performance category if there are not sufficient measures applicable and available to NPs, PAs, CRNAs, and CNSs. We would assign a weight of zero only in the event that an NP, PA, CRNA, or CNS does not submit any data for any of the measures specified for the advancing care information performance category. We encourage all NPs, PAs, CRNAs, and CNSs to report on these measures to the extent they are applicable and available, however, we understand that some NPs, PAs, CRNAs, and CNSs may choose to accept a weight of zero for this performance category if they are unable to fully report the advancing care information measures. We believe this approach is appropriate for the first MIPS performance period based on the payment consequences associated with reporting, the fact that many of these types of MIPS eligible clinicians may lack experience with EHR use, and our current uncertainty as to whether we have proposed sufficient measures that are applicable and available to these types of MIPS eligible clinicians. We note that we would use the first MIPS performance period to further evaluate the participation of these MIPS eligible clinicians in the advancing care information performance category and would consider for subsequent years whether the measures specified for this category are applicable and available to these MIPS eligible clinicians.

We invite comments on our proposal. We are additionally seeking comment on how the advancing care information performance category could be applied to NPs, PAs, CRNAs, and CNSs in future years of MIPS, and the types of measures that would be applicable and available to these types of MIPS eligible clinicians.

(iv) Medicaid

In the 2015 EHR Incentive Programs Final Rule we adopted an alternate method for demonstrating meaningful use for certain Medicaid EPs that would be available beginning in 2016, for EPs attesting for an EHR reporting period in 2015 (80 FR 62900). Medicaid EPs who previously received an incentive payment under the Medicaid EHR Incentive Program, but failed to meet the eligibility requirements for the program in subsequent years, are permitted to attest using the CMS Registration and Attestation system for the purpose of avoiding the Medicare payment adjustment (80 FR 62900). However, as discussed previously in this proposed rule, section 101(b)(1)(A) of the MACRA amended section 1848(a)(7)(A) of the Act to sunset the meaningful use payment adjustment for Medicare EHR Incentive Program EPs at the end of CY 2018. This means that after the CY 2018 payment adjustment year, there will no longer be a separate Medicare EHR Incentive Program for EPs, and therefore Medicaid EPs who may have used this alternate method for demonstrating meaningful use cannot potentially be subject to a payment adjustment under the Medicare EHR Incentive Program at that time. Accordingly, there will no longer be a need for this alternate method of demonstrating meaningful use after the CY 2018 payment adjustment year.

Similarly, beginning in 2014, states were required to collect, upload and submit attestation data for Medicaid EPs for the purposes of demonstrating meaningful use to avoid the Medicare payment adjustment (80 FR 62915). This form of reporting will also no longer need to continue with the sunset of the meaningful use payment adjustment for Medicare EHR Incentive Program EPs at the end of CY 2018. Accordingly, we are proposing to amend the reporting requirement described at 42 CFR 495.316(g) by adding an ending date such that after the CY 2018 payment adjustment year states would no longer be required to report on meaningful EHR users.

We note that the Medicaid EHR Incentive Program for EPs was not impacted by the MACRA and the requirement under section 1848(q) of the Act to establish the MIPS program. In this rule, we do not propose any changes to the objectives and measures previously established in rulemaking for the Medicaid EHR Incentive Program, and thus EPs participating in that program must continue to report on the objectives and measures under the guidelines and regulations of that program.

Accordingly, reporting on the measures specified for the advancing care information performance category under MIPS cannot be used as a demonstration of meaningful use for the Medicaid EHR Incentive Programs. Similarly, a demonstration of meaningful use in the Medicaid EHR Incentive Programs cannot be used for purposes of reporting under MIPS.

Therefore, MIPS eligible clinicians who are also participating in the Medicaid EHR Incentive Programs must report their data for the advancing care information performance category through the submission methods established for MIPS in order to earn a score for the advancing care information performance category under MIPS and must separately demonstrate meaningful use in their state's Medicaid EHR Incentive Program in order to earn a Medicaid incentive payment. The Medicaid EHR Incentive Program continues through payment year 2021, with 2016 being the final year an EP can begin receiving incentive payments (§ 495.310(a)(1)(iii)). We solicit comments on alternative reporting or proxies for EPs who provide services to both Medicaid and Medicare patients and are eligible for both MIPS and the Medicaid EHR Incentive Payment.

h. APM Scoring Standard for MIPS Eligible Clinicians Participating in MIPS APMs

Under section 1848(q)(1)(C)(ii) of the Act, as added by section 101(c)(1) of the MACRA and discussed above in section II.E.3.b. of this proposed rule, Qualifying APM Participants (QPs) are not MIPS eligible clinicians and are thus excluded from MIPS payment adjustments. Partial Qualifying APM Participants (Partial QPs) are also not MIPS eligible clinicians unless they opt to report and be scored under MIPS. All other eligible clinicians participating in APMs are MIPS eligible clinicians and subject to MIPS requirements, including reporting requirements and payment adjustments. However, most current APMs already assess their participants on cost and quality of care and require engagement in certain care improvement activities.

We propose at § 414.1370 to establish a scoring standard for MIPS eligible clinicians participating in certain types of APMs in order to reduce participant reporting burden by eliminating the need for such APM eligible clinicians to submit data for both MIPS and their respective APMs. For purposes of this APM scoring standard, we propose to consider a participant in an APM to be an entity participating in an APM under an agreement with CMS that may either include eligible clinicians or be an eligible clinician and that is directly tied to beneficiary attribution, quality measurement or cost/utilization measurement under the APM. In accordance with section 1848(q)(1)(D)(i) of the Act, we propose to assess the performance of a group of MIPS eligible clinicians in an APM Entity that participates in certain types of APMs based on their collective performance as an APM Entity group, as defined at § 414.1305.

In addition to reducing reporting burden, we seek to ensure that eligible clinicians in APM Entity groups are not assessed in multiple ways on the same performance activities. For instance, performance on the generally applicable resource use measures under MIPS could contribute to upward or downward adjustments to payments under MIPS in a way that is not aligned with the strategy in an ACO initiative for reducing total Medicare costs for a specified population of beneficiaries attributed through the unique ACO initiative's attribution methodology. Depending on the terms of the particular APM, we believe similar misalignments could be common between the MIPS quality and resource use performance categories and the evaluation of quality and resource use in APMs. We believe requiring eligible clinicians in APM Entity groups to submit data, be scored on measures, and be subject to payment adjustments that are not aligned between MIPS and an APM could potentially undermine the validity of testing or performance evaluation under the APM. We also believe imposition of these requirements would result in reporting activity that provides little or no added value to the assessment of eligible clinicians, and could confuse eligible clinicians as to which CMS incentives should take priority over others in designing and implementing care activities.

We are proposing to use the APM scoring standard for MIPS eligible clinicians in APM Entity groups participating in certain APMs that meet the criteria listed below (and are identified as “MIPS APMs” on the CMS Web site). In this section of the rule, we define the proposed criteria for MIPS APMs, the APM scoring standard, the performance period for APM Entity groups, the proposed MIPS scoring methodology for APM Entity groups, and other information related to the APM scoring standard.

(1) Criteria for MIPS APMs

We propose at § 414.1370 to specify that the APM scoring standard under MIPS would only be applicable to certain eligible clinicians participating in MIPS APMs, which we propose to define as APMs (as defined in section II.F.4. of this preamble) that meet the following criteria: (1) APM Entities participate in the APM under an agreement with CMS; (2) the APM Entities include one or more MIPS eligible clinicians on a Participation List; and (3) the APM bases payment incentives on performance (either at the APM Entity or eligible clinician level) on cost/utilization and quality measures. We understand that under some APMs the APM Entity may enter into agreements with clinicians or entities that have supporting or ancillary roles to the APM Entity's performance under the APM, but are not participating under the APM Entity and therefore are not on a Participation List. We would not consider eligible clinicians under such arrangements to be participants for purposes of the APM Entity group to which the APM scoring standard would apply. We understand that this policy would not accommodate certain APMs pursuant to statute or our regulations rather than under an agreement with CMS. We seek comments on how the APM scoring standard should apply to those APMs as well.

The criteria for the identification of MIPS APMs are independent of the criteria for Advanced APM determinations discussed in section II.F.3. of this proposed rule, so a MIPS APM may or may not also be an Advanced APM. As such, it would be possible that an APM meets all three proposed criteria to be a MIPS APM, but does not meet the Advanced APM criteria listed in section II.F.4. Conversely, it would be possible, that an Advanced APM does not meet the criteria listed above because it does not include MIPS eligible clinicians as participants.

The APM scoring standard would not apply to MIPS eligible clinicians involved in APMs that include only facilities as participants (such as the Comprehensive Care for Joint Replacement Model). APMs that do not base payment on cost/utilization and quality measures (such as the Accountable Health Communities Model) would also not meet the proposed criteria for the APM scoring standard. Instead, MIPS eligible clinicians participating in these APMs would need to meet the generally applicable MIPS data submission requirements for the MIPS performance period, and their performance would be assessed using the generally applicable MIPS standards, either as individual eligible clinicians or as a group under MIPS.

As discussed above, the APM scoring standard described in this proposed rule would require MIPS eligible clinicians to report certain data under MIPS regardless of whether they ultimately become QPs or Partial QPs through their participation in Advanced APMs. Although QPs (and Partial QPs who elect not to participate in MIPS) would be excluded from MIPS payment adjustments, we believe it is necessary, for the operational and administrative reasons discussed in section II.F.5.d., to treat these eligible clinicians as MIPS eligible clinicians unless and until the QP or Partial QP determination is made. We believe the proposed APM scoring standard would help to alleviate certain duplicative, unnecessary, or competing data submission requirements for MIPS eligible clinicians participating in MIPS APMs. However, we are interested in public comments on alternative methods that could reduce MIPS data submission requirements to enable MIPS eligible clinicians participating in Advanced APMs to maximize their focus on the care delivery redesign necessary to succeed within the Advanced APM while maintaining the statutory framework that excludes only certain eligible clinicians from MIPS, and reducing reporting burden on Advanced APM participants.

We invite public comment on alternative MIPS data submission and scoring methods. Specifically, if, during a future performance period, we are able to make QP determinations before MIPS reporting must occur, we seek to attain the least amount of required MIPS data submission while avoiding unnecessary operational complexity.

(2) APM Scoring Standard Performance Period

We propose that the performance period for MIPS eligible clinicians participating in MIPS APMs would match the generally applicable performance period for MIPS proposed in section II.E.4 of this preamble. We propose this policy would apply to all MIPS eligible clinicians participating in MIPS APMs (those that meet the criteria specified in section II.E.5.h.1. of this proposed rule) except for a new MIPS APM for which the first APM performance period begins after the start of the corresponding MIPS performance period. In this instance, the participating MIPS eligible clinicians in the new MIPS APM would submit data to MIPS in the first MIPS performance period for the APM either as individual MIPS eligible clinicians or as a group using one of the MIPS data submission mechanisms for all four performance categories, and report to CMS using the APM scoring standard for subsequent MIPS performance period(s). Additionally, we anticipate that there might be MIPS APMs that would not be able to use the APM scoring standard (even though they met the criteria for the APM scoring standard and were treated as a MIPS APMs in the prior MIPS performance period) in their last year of operation because of technical or resource issues. For example, a MIPS APM in its final year may end earlier than the end of the MIPS performance period (proposed to be December 31). CMS might not have continuing resources dedicated or available to continue to support the MIPS APM activities under the APM scoring standard if the MIPS APM ends during the MIPS performance period. Therefore, if we determine it is not feasible for the MIPS eligible clinicians participating in the APM Entity to report to MIPS using this APM scoring standard in an APM's last year of operation, the MIPS eligible clinicians in the MIPS APM would need to submit data to MIPS either as individual MIPS eligible clinicians or as a group using one of the MIPS data submission mechanisms for the applicable performance period. We propose the eligible clinicians in the MIPS APM would be made aware of this decision in advance of the relevant MIPS performance period.

(3) How the APM Scoring Standard Differs From the Assessment of Groups and Individual MIPS Eligible Clinicians Under MIPS

We believe that establishing an APM scoring standard under MIPS would allow APM Entities and their participating eligible clinicians to focus on the goals and objectives of the APM to improve quality and lower costs of care while avoiding duplicative reporting that would occur as a result of having to submit data to MIPS separately. The APM scoring standard we propose is similar to group assessment under MIPS as described in section II.E.3.d. of this proposed rule, but would differ in one or more of the following ways: (1) Depending on the terms and conditions of the MIPS APM, an APM Entity could be comprised of a sole MIPS eligible clinician (for example, a physician practice with only one eligible clinician could be considered an APM Entity); (2) the APM Entity could include more than one unique TIN, as long as the MIPS eligible clinicians are identified as participants in the APM by their unique APM participant identifiers; (3) the composition of the APM Entity group could include APM participant identifiers with TIN/NPI combinations such that some MIPS eligible clinicians in a TIN are APM participants and other MIPS eligible clinicians in that same TIN are not APM participants. In contrast, assessment as a group under MIPS requires a group to be comprised of at least two MIPS eligible clinicians who have assigned their billing rights to a TIN. It also requires that all MIPS eligible clinicians in the group to use the same TIN.

In addition to the APM Entity group composition being potentially different than that of a group as generally defined under MIPS, we propose for the APM scoring standard that we will generate a MIPS CPS by aggregating all scores for MIPS eligible clinicians in the APM Entity that is participating in the MIPS APM to the level of the APM Entity. We believe that aggregating the MIPS performance category scores at the level of the APM Entity is more meaningful to, and appropriate for, these MIPS eligible clinicians because they have elected to participate in an APM and collectively focus on care transformation activities to improve the quality of care.

Further, we propose below that, depending on the type of MIPS APM, the weights associated with performance categories may be different than the generally applicable weights for MIPS eligible clinicians. The weights assigned to the MIPS performance categories under the APM scoring standard for MIPS eligible clinicians who are participating in a MIPS APM may be different from the performance category weights for MIPs eligible clinicians not participating in a MIPS APM for the same performance period. For example, we propose below that under the APM scoring standard, the weight for the resource use performance category will be zero. We also propose that for certain MIPS APMs, the weight for the quality performance category will be zero for the 2019 payment year. Where the weight for the performance category is zero, neither the APM Entity nor the MIPS eligible clinicians in the MIPS APM would need to report data in these categories, and we would redistribute the weights for the quality and resource use performance categories to the CPIA and advancing care information performance categories to maintain a CPS of 100 percent.

In order to implement certain elements of the APM scoring standard, we would need to use the Shared Savings Program (section 1899 of the Act) and CMS Innovation Center (section 1115A of the Act) authorities to waive specific statutory provisions related to MIPS reporting and scoring. Section 1899(f) of the Act authorizes waivers of title XVIII requirements as may be necessary to carry out the Shared Savings Program, and section 1115A(d)(1) of Act authorizes waivers of title XVIII requirements as may be necessary solely for purposes of testing models under section 1115A of the Act. In each section below in which we propose scoring methodologies and waivers to enable the proposed approaches, we describe how the use of waivers is necessary under the respective waiver authority standards. The underlying purpose of APMs is for CMS to pay for care in ways that are unique from fee-for-service payment and to test new ways of measuring and assessing performance. If the data submission requirements and associated adjustments under MIPS are not aligned with APM-specific goals and incentives, the participants receive conflicting messages from CMS on priorities, which could create uncertainty and severely degrade our ability to evaluate the impact of any particular APM on the overall cost and quality of care. Therefore, we believe that, for reasons stated in this section, certain waivers are necessary for testing and operating APMs and for maintaining the integrity of our evaluation of those APMs.

We note that for at least the first performance year, we do not anticipate that any APMs not authorized under sections 1115A or 1899 of the Act would meet the criteria to be MIPS APMs. In the event that we do anticipate other Federal demonstrations will become MIPS APMs, we will address MIPS scoring for participating eligible clinicians in future rulemaking.

(4) APM Participant Identifier and Participant Database

To ensure we have accurately captured performance data for all of the MIPS eligible clinicians that are participating in an APM, we would establish and maintain an APM participant database that will include all of the MIPS eligible clinicians who are part of the APM Entity. We would establish this database to track participation in all APMs, in addition to specifically tracking participation in MIPS APMs and Advanced APMs. We propose that each APM Entity be identified in the MIPS program by a unique APM Entity identifier. We also propose in section II.E.2.b. that the unique APM participant identifier for a MIPS eligible clinician would be a combination of four identifiers including: (1) APM identifier (established for the APM by CMS; for example, XXXXXX); (2) APM Entity identifier (established for the APM by CMS; for example, AA00001111); (3) the eligible clinician's billing TIN (for example, XXXXXXXXX); and (4) NPI (for example, 1111111111). For example, this APM participant identifier for the MIPS eligible clinician in this case would be APM XXXXXX, APM Entity AA00001111, TIN-XXXXXXXXX, NPI-11111111111. The use of the APM participant identifier will allow CMS to identify all MIPS eligible clinicians participating in an APM Entity, including instances when the MIPS eligible clinicians use a billing TIN that is shared with MIPS eligible clinicians who are not participating in the APM Entity. We would plan to communicate to each APM Entity the MIPS eligible clinicians who are included in the APM Entity group in advance of the applicable MIPS data submission deadline for the MIPS performance period.

Under the Shared Savings Program, each ACO is formed by a collection of Medicare-enrolled TINs (ACO participants). Pursuant to our regulation at 42 CFR 425.118, all Medicare enrolled individuals and entities that have reassigned their rights to receive Medicare payment to the TIN of the ACO participant must agree to participate in the ACO and comply with the requirements of the Shared Savings Program. Because all providers and suppliers that bill through the TIN of an ACO participant are required to agree to participate in the ACO, all MIPS eligible clinicians that bill through the TIN of an ACO participant are considered to be participating in the ACO. For purposes of the APM scoring standard, the ACO would be the APM Entity. The Shared Savings Program has established criteria for determining the list of eligible clinicians participating under the ACO, and we would use the same criteria for determining the list of MIPS eligible clinicians included in the APM Entity group for purposes of the APM scoring standard.

We recognize that there may be scenarios in which MIPS eligible clinicians may change TINs, use more than one TIN for billing Medicare, change their APM participation status, and/or change other practice affiliations during a performance period. Therefore, we propose that only those MIPS eligible clinicians who are listed as participants in the APM Entity in a MIPS APM on December 31 (the last day of the proposed performance period) would be considered part of the APM Entity group for purposes of the APM scoring standard. Consequently, MIPS eligible clinicians who are not listed as participants of an APM Entity in a MIPS APM at the end of the performance period would need to submit data to MIPS through one of the MIPS data submission mechanisms and would have their performance assessed either as individual MIPS eligible clinicians or as a group for all four performance categories. For example, a MIPS eligible clinician who participates in the APM Entity on January 1, 2017 and leaves the APM Entity on June 15, 2017 would need to submit data to MIPS using one of the MIPS data submission mechanisms and would have their performance assessed either as individual MIPS eligible clinicians or as a group. This approach for defining the applicable group of MIPS eligible clinicians is consistent with our proposal for identifying eligible clinician groups for purposes of QP determinations outlined in section II.F.5.b. of this proposed rule; the group of eligible clinicians CMS uses for purposes of a QP determination would be the same as that used for the APM scoring standard. This would be an annual process for each MIPS performance period. We propose to calculate one MIPS CPS for each APM Entity group, and that MIPS CPS would be applied to all MIPS eligible clinicians in the group. As previously explained in section II.E.7. of this proposed rule, the MIPS payment adjustment would be applied at the TIN/NPI level for each of the MIPS eligible clinicians in the APM Entity group.

(5) MIPS Eligible Clinicians Not Participating in a MIPS APM

The APM Entity group used for purposes of the APM scoring standard would be the same APM Entity group used for QP determinations under section II.F.5 of this proposed rule, except in the instances of APMs that do not meet the criteria to be MIPS APMs, as discussed in section II.E.5.h.(1) of this proposed rule. Examples of APMs that would not meet criteria to be MIPS APMs are those that do not have MIPS eligible clinicians as participants under the APM, or do not tie payment to cost/utilization and quality measures. We propose that the APM scoring standard would not apply to MIPS eligible clinicians participating in APMs that are not MIPS APMs. MIPS eligible clinicians who participate in an APM that is not a MIPS APM, would submit data to MIPS and have their performance assessed either as an individual MIPS eligible clinician or group as described in section II.E.2. of this proposed rule. Some APMs may involve certain types of MIPS eligible clinicians that are affiliated with an APM Entity but not included in the APM Entity group because they are not participants of the APM Entity. We propose that even if the APM meets the criteria to be a MIPS APM, MIPS eligible clinicians who are not included in the list of participants would not be considered part of the APM Entity group for purposes of the APM scoring standard. For instance, MIPS eligible clinicians in the Comprehensive Care for Joint Replacement Model might be involved in the APM through a business arrangement with the APM Entity (the inpatient hospital) but are not directly tied to beneficiary attribution, quality measurement, or care improvement activities under the APM. Additionally, we propose that if a MIPS eligible clinician participates in an APM Entity during the MIPS performance period but is no longer a participant in the APM Entity group on the last day of the performance period, the MIPS eligible clinician must submit either individual or group level data to MIPS. CMS will publish the list of MIPS APMs prior to the beginning of the MIPS performance period on the CMS Web site.

(6) APM Entity Group Scoring for the MIPS Performance Categories

As mentioned previously, section 1848(q)(3)(A) of the Act requires the Secretary to establish performance standards for the measures and activities under the following performance categories: (1) Quality; (2) resource use; (3) clinical practice improvement activities; and (4) advancing care information. We propose at § 414.1370 to calculate one CPS that is applied to the billing TIN/NPI combination of each MIPS eligible clinician in the APM Entity group. Therefore, each APM Entity group (for example, the MIPS eligible clinicians in a Shared Savings Program ACO or an Oncology Care Model practice) would receive a score for each of the four performance categories according to the proposals described in this section of the proposed rule, and we would calculate one CPS for the group. The APM Entity group score would be applied to each MIPS eligible clinician in the group, and subsequently used to develop the MIPS payment adjustment that is applicable for each MIPS eligible clinician in the group. Thus the APM Entity group score and the participating MIPS eligible clinician score are the same. For example, in the Shared Savings Program, the MIPS eligible clinicians in each ACO would be an APM Entity group. That group would receive a single CPS that would be applied to each of its participating MIPS eligible clinicians. Similarly, in the Oncology Care Model, the MIPS eligible clinicians in each oncology practice would be an APM Entity group. That group would receive a single CPS that would be applied to each of the MIPS eligible clinicians in the group. We note that this APM Entity group CPS is not used to evaluate eligible clinicians or the APM Entity for purposes of incentives within the APM, shared savings payments, or other potential payments under the APM, and we currently do not foresee APMs that would use the CPS for purposes of evaluation within the APM. Rather the APM Entity group CPS would be used only for the purposes of the APM scoring standard under MIPS for the first MIPS performance period. As proposed in this rule, all MIPS eligible clinicians listed as participating in the APM Entity on the last day of the performance period would be part of the APM Entity group and thus receive the same CPS. It should be noted that although we propose that the APM scoring standard only applies to participants in MIPS APMs, MIPS eligible clinicians that participate in an APM (including but not limited to a MIPS APM) and submit either individual or group level data to MIPS may earn a minimum score of 50 percent of the highest potential CPIA performance category score as long as such MIPS eligible clinicians are on the list of participants for an APM and are identifiable by the APM participant identifier.

Several commenters on the MIPS and APMs RFI suggested, and we generally agree, that MIPS eligible clinicians who collaborate under an APM Entity to accomplish the APM's goals should be treated as a group under MIPS and receive the same CPS. Furthermore, we want to avoid situations in which different MIPS eligible clinicians in the same APM Entity group receive different MIPS scores. APM Entities have a goal of collective success under the terms of the APM, so having a variety of differing MIPS adjustments for eligible clinicians within that collective unit would undermine the intent behind the APM to test a departure from a purely fee-for-service system based on independent clinician activity. Lastly, we believe that measurement of the performance for MIPS at the APM Entity level for eligible clinicians participating in MIPS APMs will result in more statistically valid performance scores for these eligible clinicians because the scores are aggregated to represent a larger group of MIPS eligible clinicians.

We propose, for the first MIPS performance period, a specific scoring and reporting approach for the MIPS eligible clinicians participating in MIPS APMs, which would include the Shared Savings Program, the Next Generation ACO Model, and other APMs that meet the criteria proposed above for a MIPS APM. Specifically, we propose that APM quality measure data submitted through the CMS Web Interface by ACOs participating in the Shared Savings Program and the Next Generation ACO Model would be used to evaluate performance for the MIPS quality performance category. We believe this is appropriate because all MIPS eligible clinicians that use the CMS Web Interface as their quality measure submission mechanism, e.g., MIPS eligible clinicians that report as a group and MIPS APM eligible clinicians that report as an APM Entity group, submit data on the same quality measures. Both the Shared Savings Program and the Next Generation ACO Model use additional quality measures for the purpose of APM performance assessment, but only the measures submitted to the CMS Web Interface would be used to evaluate performance for the MIPS quality performance category. Therefore, other measures that are required by the APM to assess APM quality performance will continue to be used for APM performance assessment only and not included in the MIPS quality performance category scoring. We also propose that MIPS eligible clinicians participating in MIPS APMs that do not use the CMS Web Interface as the mechanism for submitting APM quality data would not submit quality measure data to MIPS for the MIPS quality performance category until the second MIPS performance period (2018). In this section of the rule, we describe the APM Entity data submission requirements and propose a scoring approach for each of the MIPS performance categories for specific MIPS APMs (the Shared Savings Program, Next Generation ACO Model, and all other MIPS APMs).

(7) Shared Savings Program—Quality Performance Category Scoring Under the APM Scoring Standard

Beginning with the first MIPS performance period all Shared Savings Program ACOs would submit their quality measures to MIPS using the CMS Web Interface through the same process that they use to report to the Shared Savings Program and be scored as they normally would under Shared Savings Program rules. Shared Savings Program ACOs have used the CMS Web Interface for submitting their quality measures since the program's inception, making this a familiar data submission process. We also propose that the Shared Savings Program ACO quality measure data that is submitted through the CMS Web Interface will be submitted only once but will be used for two purposes. The Shared Savings Program quality measure data reported to the CMS Web Interface would be used by CMS to calculate the MIPS quality performance category score at the APM Entity group (ACO) level. The Shared Savings Program quality performance data that is not submitted to the CMS Web Interface, for example the CAHPS survey and other claims measures would not be included in the MIPS APM quality performance category score. We believe this will reduce the reporting burden for Shared Savings Program MIPS eligible clinicians by requiring quality measure data to be submitted only once and used for both programs. The MIPS quality performance category requirements and performance benchmarks for quality measures submitted via the CMS Web Interface would be used to determine the MIPS quality performance category score at the ACO level for the APM Entity group.

We believe that no waivers are necessary here because the quality measures submitted via the CMS Web Interface under the Shared Savings Program are also MIPS quality measures and will be scored under MIPS performance standards. In the event that Shared Savings Program quality measures depart from MIPS measures in the future, we will address such changes including whether further waivers are necessary at such a time in future rulemaking.

(8) Shared Savings Program—Resource Use Performance Category Scoring Under the APM Scoring Standard

We propose that for the first MIPS performance period, we will not assess MIPS eligible clinicians participating in the Shared Savings Program (the MIPS APM) under the resource use performance category. We propose this approach because: (1) Eligible clinicians participating in the Shared Savings Program are already subject to cost and utilization performance assessments under the APM; (2) the Shared Savings Program measures resource use in terms of an objective, absolute total cost of care expenditure benchmark for a population of attributed beneficiaries, and participating ACOs may share savings and/or losses based on that standard, whereas the MIPS resource use measures are relative measures such that clinicians are graded relative to their peers, and therefore different than assessing total cost of care for a population of attributed beneficiaries; and (3) the beneficiary attribution methodologies for measuring resource use under the Shared Savings Program and MIPS differ, leading to an unpredictable degree of overlap (for eligible clinicians and for CMS) between the sets of beneficiaries for which eligible clinicians would be responsible that would vary based on unique APM Entity characteristics such as which and how many TINs comprise an ACO. We believe that with an APM Entity's finite resource for engaging in efforts to improve quality and lower costs for a specified beneficiary population, the population identified through an APM must take priority to ensure that the goals and program evaluation associated with the APM are as clear and free of confounding factors as possible. The potential for different, conflicting results across Shared Savings Program and MIPS assessments—due to the differences in attribution, the inclusion in MIPS of episode-based measures that do not reflect the total cost of care, and the objective versus relative assessment factors listed above—creates uncertainty for eligible clinicians who are attempting to strategically transform their respective practices and succeed under the terms of the Shared Savings Program.

For example, Shared Savings Program ACOs are held accountable for expenditure benchmarks that reflect the total Medicare Parts A and B spending for their assigned beneficiaries, whereas many of the proposed MIPS resource use measures focus on spending for particular episodes of care or clinical conditions. For the reasons stated above, we consider it a programmatic necessity that the Shared Savings Program has the ability to structure its own measurement and payment for performance on total cost of care independent from other incentive programs such as the resource use performance category under MIPS. Thus, we propose to reduce the MIPS resource use performance category weight to zero for all MIPS eligible clinicians in APM Entities participating in the Shared Savings Program. Accordingly, under section 1899(f) of the Act, we propose to waive—for MIPS eligible clinicians participating in the Shared Savings Program—the requirement under section 1848(q)(5)(E)(i)(II) of the Act that specifies the scoring weight for the resource use performance category. With the proposed reduction of the resource use performance category weight to zero, we believe it would be unnecessary specify and use resource use measures in determining the MIPS CPS for these MIPS eligible clinicians. Therefore, under section 1899(f) of the Act, we propose to waive—for MIPS eligible clinicians participating in the Shared Savings Program—the requirements under sections 1848(q)(2)(B)(ii) and 1848(q)(2)(A)(ii) of the Act to specify and use, respectively, resource use measures in calculating the MIPS CPS for such MIPS eligible clinicians.

Given the proposal to waive requirements under section 1848(q)(5)(E)(i)(II) of the Act in order to reduce the weight of the resource use performance category to zero, we must subsequently specify how that weight would be redistributed among the remaining performance categories in order to maintain a total weight of 100 percent. We propose to redistribute the resource use performance category weight to both the CPIA and advancing care information performance categories as specified in Table 12. The MIPS resource use performance category is proposed to have a weight of 10 percent for the first performance period. Because the MIPS quality performance category bears a relatively higher weight than the other three MIPS performance categories, and its weight is scheduled to be reduced from 50 to 30 percent over time, we propose to evenly redistribute the 10 percent resource use performance category weight to the CPIA and advancing care information performance categories so that the distribution does not change the relative weight of the quality performance category in the opposite direction of its future state. The redistributed resource use performance category weight of 10 percent would result in a 5 percentage point increase (from 15 to 20 percent) for the CPIA performance category and a 5 percentage point increase (from 25 to 30 percent) for the advancing care information performance category. We invite comments on the proposed weights and specifically, whether we should increase the MIPS quality performance category weight.

We understand that as the MIPS resource use performance category evolves over time, there might be greater potential for alignment and less potential duplication or conflict with MIPS resource use measurement for MIPS eligible clinicians participating in APMs such as the Shared Savings Program. We will continue to monitor and consider how we might incorporate an assessment in the MIPS resource use performance category into the APM scoring standard for MIPS eligible clinicians participating in the Shared Savings Program. We also understand that reducing the resource use performance category weight to zero and redistributing the weight to the CPIA and advancing care information performance categories could, to the extent that CPIA and advancing care information scores are higher than the scores these MIPS eligible clinicians would have received under resource use, result in higher average scores for MIPS eligible clinicians participating in the Shared Savings Program. We seek comment on the possibility of assigning a neutral score to the Shared Savings Program APM Entity groups for the resource use performance category to moderate MIPS composite performance scores for APM Entities participating in the Shared Savings Program. We also generally seek comment on our proposed policy, and on whether and how we should incorporate the resource use performance category into the APM scoring standard under MIPS for eligible clinicians participating in the Shared Savings Program for future years.

(9) Shared Savings Program—CPIA and Advancing Care Information Performance Category Scoring Under the APM Scoring Standard

We propose that MIPS eligible clinicians participating in the Shared Savings Program would submit data for the MIPS CPIA and advancing care information performance categories through their respective ACO participant billing TINs independent of the Shared Savings Program ACO. Pursuant to section 1848(q)(5)(C)(ii) of the Act, all ACO participant group billing TINs would receive a minimum of one half of the highest possible score for the CPIA performance category. Additionally, pursuant to section 1848(q)(5)(C)(i) of the Act, any ACO participant TIN that is determined to be a patient-centered medical home or comparable specialty practice will receive the highest potential score for the CPIA performance category. The scores from all of the ACO participant billing TINs would be averaged to a weighted mean MIPS APM Entity group level score. We propose to use a weighted mean in computing the overall CPIA and advancing care information quality performance category score in order to account for difference in the size of each TIN and to allow each TIN to contribute to the overall score based on its size. Then all MIPS eligible clinicians in the APM Entity group, as identified by their APM participant identifiers, would receive that APM Entity score. The weights used for each ACO participant billing TIN would be the number of MIPS eligible clinicians in that TIN. Because all providers and suppliers that bill through the TIN of an ACO participant are required to agree to participate in the ACO, all MIPS eligible clinicians that bill through the TIN of an ACO participant are considered to be participating in the ACO. Any Shared Savings Program ACO participant billing TIN that does not submit data for the MIPS CPIA and/or advancing care information performance categories would contribute a score of zero for each performance category for which it does not report; and that score would be incorporated into the resulting weighted average score for the Shared Savings Program ACO. All MIPS eligible clinicians in the ACO (the APM Entity group) would receive the same score that is calculated at the ACO level (the APM Entity).

In this example, each eligible clinician participating in the APM Entity (Shared Savings Program ACO) would receive a CPIA performance category score of 78.5 and an advancing care information performance category score of 85. We recognize that the Shared Savings Program eligible clinicians participate as a complete TIN because all of the eligible clinicians that have reassigned their Medicare billing rights to the TIN of an ACO participant must agree to participate in the Shared Savings Program. This is different from other APMs, which may include APM Entity groups with eligible clinicians who share a billing TIN with other eligible clinicians who do not participate in the APM Entity. We seek comment on a possible alternative approach in which CPIA and advancing care information performance category scores would be applied to all MIPS eligible clinicians at the individual billing TIN level, as opposed to aggregated to the ACO level, for Shared Savings Program participants. If MIPS APM scores were applied to each TIN in an ACO at the TIN level, we would also likely need to permit those TINs to make the Partial QP election, as discussed elsewhere in this proposed rule, at the TIN level. We propose that under the APM scoring standard, the ACO-level APM Entity group score would be applied to each participating MIPS eligible clinician to determine the MIPS payment adjustment. We believe calculating the score at the APM Entity level mirrors the way APM participants are assessed for their shared savings and other incentive payments in the APM, but we understand there may be reasons why a group TIN, particularly one that believes it would achieve a higher score than the weighted average APM Entity level score, would prefer to be scored in the CPIA and advancing care information performance categories at the level of the group billing TIN rather than the ACO (APM Entity level). Therefore, we seek comment as to whether Shared Savings Program ACO eligible clinicians should be scored at the ACO level or the group billing TIN level for the CPIA and advancing care information performance categories. In Table 12, we provide a summary of the proposed MIPS data submission requirements and scoring under the APM scoring standard for MIPS eligible clinicians participating in a Shared Savings Program ACO.

(10) Next Generation ACO Model—Quality Performance Category Scoring Under the APM Scoring Standard

Beginning with the first MIPS performance period, all Next Generation ACO Model ACOs would submit their ACO quality measures to MIPS using the CMS Web Interface through the same process that they use to report to the Next Generation ACO Model and be scored as they normally would under Next Generation ACO Model rules. Next Generation ACO Model ACOs will have used the CMS Web Interface for submitting their quality measures since the model's inception and would most likely continue to use the CMS Web Interface as the submission method in future years. We also propose that the Next Generation ACO Model quality measure data that is submitted through the CMS Web Interface will be submitted only once but will be used for two purposes. The Next Generation ACO Model quality measure data reported to the CMS Web Interface would be used by CMS to calculate the MIPS APM quality performance score. The MIPS quality performance category requirements and performance benchmarks for reporting quality measures via the CMS Web Interface would be used to determine the MIPS quality performance category score at the ACO level for the APM Entity group. The Next Generation ACO Model quality performance data that is not submitted to the CMS Web Interface, for example the CAHPS survey and other claims measures would not be included in the MIPS APM quality performance score. The MIPS APM quality performance category score would be calculated using only quality measure data submitted through the CMS Web Interface, while the quality reporting requirements and performance benchmarks calculated by the Next Generation ACO Model would continue to be used to assess the ACO under the APM specific requirements. We believe this approach would reduce the reporting burden to Next Generation ACO Model participants by requiring quality measure data to be submitted only once and used for both MIPS and the Next Generation ACO Model.

We believe that no waivers are necessary here because the quality measures submitted via the CMS Web Interface under the Next Generation ACO Model are MIPS quality measures and will be scored under MIPS performance standards. In the event that Next Generation ACO Model quality measures depart from MIPS measures in the future, we will address such changes, including whether further waivers are necessary, at such a time in future rulemaking.

(11) Next Generation ACO Model—Resource Use Performance Category Scoring Under the APM Scoring Standard

We propose that for the first MIPS performance period, we will not assess MIPS eligible clinicians in the Next Generation ACO Model participating in the MIPS APM under the resource use performance category. We propose this approach because: (1) MIPS eligible clinicians participating in the Next Generation ACO Model are already subject to cost and utilization performance assessments under the APM; (2) the Next Generation ACO Model measures resource use in terms of an objective, absolute total cost of care expenditure benchmark for a population of attributed beneficiaries, and participating ACOs may share savings and/or losses based on that standard, whereas the MIPS resource use measures are relative measures such that clinicians are graded relative to their peers and therefore different than assessing total cost of care for a population of attributed beneficiaries; and (3) the beneficiary attribution methodologies for measuring resource use under the Next Generation ACO Model and MIPS differ, leading to an unpredictable degree of overlap (for eligible clinicians and for CMS) between the sets of beneficiaries for which eligible clinicians would be responsible that would vary based on unique APM Entity characteristics such as which and how many eligible clinicians comprise an ACO. We believe that with an APM Entity's finite resources for engaging in efforts to improve quality and lower costs for a specified beneficiary population, the population identified through the Next Generation ACO Model must take priority to ensure that the goals and model evaluation associated with the APM are as clear and free of confounding factors as possible. The potential for different, conflicting results across the Next Generation ACO Model and MIPS assessments—due to the differences in attribution, the inclusion in MIPS of episode-based measures that do not reflect the total cost of care, and the objective versus relative assessment factors listed above—creates uncertainty for eligible clinicians who are attempting to strategically transform their respective practices and succeed under the terms of the Next Generation ACO Model. For example, Next Generation ACOs are held accountable for expenditure benchmarks that reflect the total Medicare Parts A and B spending for their attributed beneficiaries, whereas many of the proposed MIPS resource use measures focus on spending for particular episodes of care or clinical conditions. For all the reasons stated above, we propose to reduce the MIPS resource use performance category weight to zero for all MIPS eligible clinicians participating in the Next Generation ACO Model. Accordingly, under section 1115A(d)(1) of the Act, we propose to waive—for MIPS eligible clinicians participating in the Next Generation ACO Model—the requirement under section 1848(q)(5)(E)(i)(II) of the Act that specifies the scoring weight for the resource use performance category. With the proposed reduction of the resource use performance category weight to zero, we believe it would be unnecessary to specify and use resource use measures in determining the MIPS CPS for these MIPS eligible clinicians. Therefore, under section 1115A(d)(1) of the Act, we propose to waive—for MIPS eligible clinicians participating in the Next Generation ACO Model—the requirements under sections 1848(q)(2)(B)(ii) and 1848(q)(2)(A)(ii) of the Act to specify and use, respectively, resource use measures in calculating the MIPS CPS for such eligible clinicians.

Given the proposal to waive requirements under section 1848(q)(5)(E) of the Act in order to reduce the weight of the resource use performance category to zero, we must subsequently specify how that weight would be redistributed among the remaining performance categories in order to maintain a total weight of 100 percent. We propose to redistribute the resource use performance category weight to both the CPIA and advancing care information performance categories as specified in Table 13. The MIPS resource use performance category is proposed to have a weight of 10 percent. Because the MIPS quality performance category bears a relatively higher weight than the other three MIPS performance categories and its weight is scheduled to be reduced from 50 to 30 percent over time, we propose to evenly redistribute the 10 percent resource use weight to the CPIA and advancing care information performance categories so that the distribution does not change the relative weight of the quality performance category in the opposite direction of its future state. The redistributed resource use performance category weight of 10 percent would result in a 5 percentage point increase (from 15 to 20 percent) for the CPIA performance category and a 5 percentage point increase (from 25 to 30 percent) for the advancing care information performance category. We invite comments on the proposed redistributed weights and specifically on whether we should also increase the MIPS quality performance category weight.

We understand that as the MIPS resource use performance category evolves over time, there might be greater potential for alignment and less potential duplication or conflict with MIPS resource use measurement for MIPS eligible clinicians participating in MIPS APMs such as the Next Generation ACO Model. We will continue to monitor and consider how we might incorporate an assessment in the MIPS resource use performance category into the APM scoring standard for the Next Generation ACO Model. We also understand that reducing the resource use weight to zero and redistributing the weight to the CPIA and advancing care information performance categories could, to the extent that CPIA and advancing care information scores are higher than the scores MIPS eligible clinicians would have received under resource use, result in higher average scores for MIPS eligible clinicians in APM Entity groups participating in the Next Generation ACO Model. We seek comment on the possible alternative of assigning a neutral score to APM Entity groups (ACOs) participating in the Next Generation ACO model for the resource use performance category in order to moderate APM Entity scores. We also generally seek comment on our proposed policy, and on whether and how we should incorporate the resource use performance category into the APM scoring standard for MIPS eligible clinicians in APM Entity groups participating in the Next Generation ACO model for future years.

(12) Next Generation ACO Model—CPIA and Advancing Care Information Performance Category Scoring Under the APM Scoring Standard

We propose that all MIPS eligible clinicians participating in the Next Generation ACO Model would submit data for the CPIA and advancing care information performance categories. Eligible clinicians in the Next Generation ACO Model may belong to a billing TIN that includes non-participating APM eligible clinicians. Therefore for both CPIA and the advancing care information performance category, we propose that these MIPS eligible clinicians would submit individual level data to MIPS and not group level data.

For both the CPIA and advancing care information performance categories, the scores from all of the individual MIPS eligible clinicians in the APM Entity group would be aggregated to the APM Entity level and averaged for a mean score. Any individual MIPS eligible clinicians that do not report the CPIA or advancing care information performance category would contribute a score of zero for that performance category in the calculation of the APM Entity score. All MIPS eligible clinicians in the APM Entity group would receive the same APM Entity score.

As noted above, because the MIPS quality performance category bears a relatively higher weight than the other three MIPS performance categories, we propose to evenly redistribute the 10 percent resource use performance category weight to the CPIA and advancing care information performance categories. Section 1848(q)(5)(C)(i) of the Act requires that MIPS eligible clinicians who are in a practice that is certified as a patient-centered medical home or comparable specialty practice, as determined by the Secretary, with respect to a performance period shall be given the highest potential score for the CPIA performance category. Accordingly, a MIPS eligible clinician participating in an APM Entity that meets the definition of a patient-centered medical home or comparable specialty practice, as discussed in section II.E.5.f. of this proposed rule, will receive the highest potential score. Additionally, section 1848(q)(5)(C)(ii) of the Act requires that MIPS eligible clinicians participating in APMs that are not patient-centered medical homes for a performance period shall earn a minimum score of one-half of the highest potential score for CPIA.

For the APM scoring standard for the first MIPS performance period, we propose to weight the CPIA and advancing care information performance categories for the Next Generation ACO Model in the same way that we propose to weight those categories for the Shared Savings Program: 20 percent and 30 percent for CPIA and advancing care information, respectively. We seek comment on our proposals for reporting and scoring the CPIA and advancing care information performance categories under the APM scoring standard. In particular, we seek comment on the appropriate weight distributions in the first year.

In Table 13, we provide a summary of the proposed MIPS data submission and scoring under the APM scoring standard for MIPS eligible clinicians participating in a Next Generation ACO.

(13) MIPS APMs Other Than the Shared Savings Program and the Next Generation ACO Model—Quality Performance Category Scoring Under the APM Scoring Standard

For MIPS APMs other than the Shared Savings Program and the Next Generation ACO Model, we propose that eligible clinicians or APM Entities would submit APM quality measures under their respective MIPS APM as usual, and those eligible clinicians or APM Entities would not also be required to submit quality information under MIPS. Current MIPS APMs have requirements regarding the number of quality measures, measure specifications, as well as the measure reporting method(s) and frequency of reporting, and have an established mechanism for submission of these measures to CMS. We believe there are operational considerations and constraints that would prevent us from being able to use the quality measure data from some MIPS APMs for the purpose of satisfying the MIPS data submission requirements for the quality performance category in the first performance period. For example, some current APMs use a quality measure data collection system or vehicle that is separate and distinct from the MIPS systems. We do not believe there is sufficient time to adequately implement changes to the current APM quality measure data collection timelines and infrastructure to conduct a smooth hand-off to the MIPS system that would enable use of APM quality measure data to satisfy the MIPS quality performance category requirements in the first MIPS performance period. As we have noted, we are concerned about subjecting MIPS eligible clinicians who participate in MIPS APMs to multiple performance assessments—under MIPS and under the APMs—that are not necessarily aligned and that could potentially undermine the validity of testing or performance evaluation under the APM. As stated previously, our goal is to reduce MIPS eligible clinician reporting burden by not requiring APM participants to report quality data twice to CMS, and to avoid misaligned performance incentives. Therefore, we propose that, for the first MIPS performance period only, for MIPS eligible clinicians participating in APM Entity groups in MIPS APMs (other than the Shared Savings Program or the Next Generation ACO Model), we would reduce the weight for the quality performance category to zero. We believe it is necessary to do this because CMS requires additional time to make adjustments in systems and processes related to the submission and collection of APM quality measures in order to align APM quality measures with the MIPS, and ensure APM quality measure data can be submitted in a time and in a manner sufficient for use in assessing quality performance under MIPS and under the APM. Additionally, due to the implementation of a new program that does not account for non-MIPS measures sets, the operational complexity of connecting APM performance to valid MIPS quality performance category scores in the necessary timeframe, as well as the uncertainty of the validity and equity of scoring results could unintentionally undermine the quality performance assessments in MIPS APMs. Finally, for purposes of performing valid evaluations of MIPS APMs, we must reduce the number of confounding factors to the extent feasible, which, in this case, would include reporting and assessment on non-APM quality measures. Thus, we propose to waive certain requirements of section 1848(q) of the Act for the first MIPS performance year to avoid risking adverse operational or program evaluation consequences for MIPS APMs while we work toward incorporating MIPS APM quality measures into MIPS scoring for future MIPS performance periods without. Accordingly, under section 1115A(d)(1) of the Act, we propose to waive—for MIPS eligible clinicians participating in MIPS APMs other than the Shared Savings Program or the Next Generation ACO Model—the requirement under section 1848(q)(5)(E)(i)(I) of the Act that specifies the scoring weight for the quality performance category. With the proposed reduction of the quality performance category weight to zero, we believe it would be unnecessary to establish an annual final list of quality measures as required under section 1848(q)(2)(D) of the Act, or to specify and use quality measures in determining the MIPS CPS for these MIPS eligible clinicians. Therefore, under section 1115A(d)(1) of the Act, we propose to waive—for MIPS eligible clinicians participating in MIPS APMs other than the Shared Savings Program or the Next Generation ACO Model—the requirements under sections 1848(q)(2)(D), 1848(q)(2)(B)(i) and 1848(q)(2)(A)(i) of the Act to establish a final list of quality measures (using certain criteria and processes); and to specify and use, respectively, quality measures in calculating the MIPS CPS, for these MIPS eligible clinicians.

We anticipate that beginning in the second MIPS performance period, the APM quality measure data submitted during the MIPS performance period to us would be used to derive a MIPs quality performance score for APM Entities in all APMs that meet criteria for application of the APM scoring standard. We anticipate that it may be necessary to propose policies and waivers of different requirements of the statute—such as one for section 1848(q)(2)(D) of the Act, to enable the use of non-MIPS quality measures in the quality performance category score—through future rulemaking. We expect that by the second MIPS performance period we will have had sufficient time to resolve operational constraints related to use of separate quality measure systems and adjust quality measure data submission timelines. Therefore, beginning with the second MIPS performance period, we anticipate that through use of the waiver authority under section 1115A(d)(1) of the Act, the quality measure data for APM Entities for which the APM scoring standard applies would be used for calculation of a MIPS quality performance score in a manner specified in future rulemaking. We seek comment on this transitional approach to use APM quality measures for the MIPS quality performance category for purposes of the APM scoring standard under MIPS in future years.

(14) MIPS APMs Other Than the Shared Savings Program and Next Generation ACO—Resource Use Performance Category Scoring Under the APM Scoring Standard

For the first MIPS performance period, we propose that, for MIPS eligible clinicians participating in MIPS APMs other than the Shared Savings Program or the Next Generation ACO, to reduce the weight of the resource use performance category to zero. We propose this approach because: (1) APM Entity groups are already subject to cost and utilization performance assessments under MIPS APMs; (2) MIPS APMs usually measure resource use in terms of total cost of care, which is a broader accountability standard inherently encompasses the purpose of the claims-based measures that have relatively narrow clinical scopes, and MIPS APMs that do not measure resource use in terms of total cost of care may depart entirely from MIPS measures; and (3) the beneficiary attribution methodologies differ for measuring resource use under APMs and MIPS, leading to an unpredictable degree of overlap (for eligible clinicians and for CMS) between the sets of beneficiaries for which eligible clinicians would be responsible that would vary based on unique APM Entity characteristics such as which and how many eligible clinicians comprise an APM Entity. We believe that with an APM Entity's finite resources for engaging in efforts to improve quality and lower costs for a specified beneficiary population, the population identified through an APM must take priority to ensure that the goals and model evaluation associated with the APM are as clear and free of confounding factors as possible. The potential for different, conflicting results across APM and MIPS assessments creates uncertainty for MIPs eligible clinicians who are attempting to strategically transform their respective practices and succeed under the terms of an APM. Accordingly, under section 1115A(d)(1) of the Act, we propose to waive—for MIPS eligible clinicians participating in MIPS APMs other than the Shared Savings Program or the Next Generation ACO Model—the requirement under section 1848(q)(5)(E)(i)(II) of the Act that specifies the scoring weight for the resource use performance category.

With the proposed reduction of the resource use performance category weight to zero, we believe it would be unnecessary to specify and use resource use measures in determining the MIPS CPS for these MIPS eligible clinicians. Therefore, under section 1115A(d)(1) of the Act, we propose to waive—for MIPS eligible clinicians participating in MIPS APMs other than the Shared Savings Program or the Next Generation ACO Model—the requirements under section under sections 1848(q)(2)(B)(ii) and 1848(q)(2)(A)(ii) of the Act to specify and use, respectively, resource use measures in calculating the MIPS CPS for such eligible clinicians.

Given the proposal to waive requirements of section 1848(q) of the Act to reduce the weight of the quality and resource use performance categories to zero, we must subsequently specify how those weights would be redistributed among the remaining CPIA and advancing care information categories in order to maintain a total weight of 100 percent. We propose to redistribute the quality and the resource use performance category weights as specified in Table 14.

We understand that as the resource use performance category evolves, the rationale we discussed earlier for establishing a weight of zero for this performance category might not be applicable in future years. We seek comment on whether and how we should incorporate the resource use performance category into the APM scoring standard under MIPS. We also understand that reducing the quality and resource use performance category weight to zero and redistributing the weight to the CPIA and advancing care information performance categories could, to the extent that CPIA and advancing care information scores are higher than the scores MIPS eligible clinicians would have received under resource use, result in higher average scores for MIPs eligible clinicians in APM Entity groups participating in MIPS APMs. We seek comment on the possible alternative of assigning a neutral score to MIPS eligible clinicians in APM Entity groups participating in MIPS APMs for the quality and resource use performance category in order to moderate APM Entity scores.

(15) MIPS APMs Other Than the Shared Savings Program and Next Generation ACO Model—CPIA and Advancing Care Information Performance Category Scoring Under the APM Scoring Standard

We propose that all MIPS eligible clinicians participating in a MIPS APM other than the Shared Savings Program or the Next Generation ACO would submit data for the CPIA and Advancing Care Information performance categories. We propose that these MIPS eligible clinicians would submit data for both the CPIA and advancing care information performance categories as individual MIPS eligible clinicians. MIPS eligible clinicians in these other APMs may belong to a billing TIN that includes MIPs eligible clinicians that do not participate in the APM. Therefore for both CPIA and the advancing care information performance category, we propose that these MIPS eligible clinicians submit individual level data to MIPS and not group level data.

For both the CPIA and advancing care information performance categories, the scores from all of the individual MIPS eligible clinicians in the APM Entity group would be aggregated to the APM Entity level and averaged for a mean score. Any individual MIPS eligible clinicians that do not submit data for the CPIA or advancing care information performance category would contribute a score of zero for that performance category in the calculation of the APM Entity score. All MIPS eligible clinicians in the APM Entity group would receive the same APM Entity group score.

Section 1848(q)(5)(C)(i) of the Act requires that MIPS eligible clinicians who are in a practice that is certified as a patient-centered medical home or comparable specialty practice, as determined by the Secretary, with respect to a performance period shall be given the highest potential score for the CPIA performance category. Accordingly, a MIPS eligible clinician in an APM Entity group that meets the definition of a patient-centered medical home or comparable specialty practice, as discussed in section II.E.5.f. of this proposed rule, will receive the highest potential score. Additionally, section 1848(q)(5)(C)(ii) of the Act requires that MIPS eligible clinicians participating in APMs that are not patient-centered medical homes for a performance period shall earn a minimum score of one-half of the highest potential score for CPIA. We acknowledge that using this increased weight for CPIA may make it easier in the first performance period to attain a higher MIPS score. We do not have historical data to assess the range of scores under CPIA because this is the first time such activities are being assessed in such a manner.

With respect to the advancing care information performance category, we believe that MIPS eligible clinicians participating in MIPS APMs would be using certified health IT and other health information technology to coordinate care and deliver better care to their patients. Most MIPS APMs encourage participants to use health IT to perform population management, monitor their own quality improvement activities and, better coordinate care for their patients in a way that aligns with the goals of the advancing care information performance category. We want to ensure that where we propose reductions in weights for other MIPS performance categories, such weights are appropriately redistributed to the advancing care information performance category.

Therefore, for the first MIPS performance period, we propose that the weights for the CPIA and advancing care information performance categories would be 25 percent and 75 percent, respectively. We seek comment on our proposals for reporting and scoring the CPIA and advancing care information performance categories under the APM scoring standard. In particular, we seek comment on the appropriate weight distributions in the first year and subsequent years when we anticipate incorporating assessment in the quality performance category for all MIPS eligible clinicians participating in MIPS APMs.

Table 14 shows the performance category scoring and weights for other APMs for which the APM scoring standard applies.

(14) APM Entity Data Submission Method

Presently, CMS requires MIPS APMs to either use the CMS Web Interface or another data submission mechanism for submitting data on the quality measures for purposes of the APM. We are not currently proposing to change the method used by APM Entities to submit their data on quality measures to CMS for purposes of MIPS. Therefore, we expect that APM Entities like the Shared Savings Program ACOs would continue to submit their data on quality measures using the CMS Web Interface data submission mechanism. Similarly, participants in the Comprehensive ESRD Care (CEC) Initiative would continue to submit their quality measures to CMS using the Quality Measures Assessment Tool (QMAT) for purposes of the CEC quality performance assessment under the APM. All eligible clinicians in APM Entities participating in MIPS APMs would be required to use one of the proposed MIPS data submission mechanisms to submit data for the CPIA and advancing care information performance categories.

(15) MIPS APM Performance Feedback

For the first MIPS performance feedback specified under section 1848(q)(12) of the Act to be published by July 1, 2017, we propose that all MIPS eligible clinicians participating in MIPS APMs would receive the same historical information prepared for all MIPS eligible clinicians except the report would indicate that the historical information provided to such MIPS eligible clinicians is for informational purposes only. MIPS eligible clinicians participating in APMs have been evaluated for performance only under the APM. Thus, historical information may not be representative of the scores that these MIPS eligible clinicians would receive under MIPS.

For MIPS eligible clinicians participating in MIPS APMs, we propose that the MIPS performance feedback would consist only of the scores applicable to the APM Entity group for the specific MIPS performance period. For example, the MIPS eligible clinicians participating in the Shared Savings Program and Next Generation ACO Model would receive performance feedback for the quality, CPIA, and advancing care information performance categories for the 2017 performance period. Because these MIPS eligible clinicians would not be assessed for the resource use performance category, information on MIPS performance scores for the resource use performance category would not be applicable to these MIPS eligible clinicians.

We also propose that, for the Shared Savings Program the performance feedback would be available to the eligible clinicians participating in the Shared Savings Program at the group billing TIN level. For the Next Generation ACO Model we propose that the performance feedback would be available to all MIPS eligible clinicians participating in the MIPS APM Entity.

We propose that in the first MIPS performance period, the MIPS eligible clinicians participating in MIPS APMs other than the Shared Savings Program or the Next Generation ACO Model would receive performance feedback for the CPIA and advancing care information only, as they would not be assessed under the quality or resource use performance categories. The information such as MIPS measure score comparisons for the quality and resource use performance categories would not be applicable to these MIPS eligible clinicians because no such comparative data would exist. We propose the performance feedback for all other MIPS eligible clinicians participating in APMs would be available for each MIPS eligible clinician that submitted MIPS data for these performance categories under their respective APM Entities. We invite comment on these proposals.

6. MIPS Composite Performance Score Methodology

By incentivizing quality and value for all eligible clinicians, MIPS creates a new mechanism for calculating eligible clinician payments. To implement this vision, we propose a scoring methodology that allows for accountability and alignment across the performance categories and minimizes burden on MIPS eligible clinicians. Further, we propose a scoring methodology that is meaningful, understandable and flexible for all MIPS eligible clinicians. Our proposed methodology allows for multiple pathways to success with flexibility for the variety of practice types and reporting options. First, we have proposed multiple ways that MIPS eligible clinicians may submit data to MIPS for the quality performance category. Second, we generally do not propose “all-or-nothing” reporting requirements for MIPS. Third, bonus points would be available for reporting high priority measures and electronic reporting of quality data. Recognizing that MIPS is a new program, we also outline proposals which we believe are operationally feasible for us to implement in the first year, while maintaining our longer-term vision, as well as Congress' vision.

Section 1848(q) of the Act requires the Secretary to: (1) Develop a methodology for assessing the total performance of each MIPS eligible clinician according to performance standards for a performance period for a year; (2) using the methodology, provide a composite performance score for each MIPS eligible clinician for each performance period; and (3) use the CPS of the MIPS eligible clinician for a performance period to determine and apply a MIPS adjustment factor (and, as applicable, an additional MIPS adjustment factor) to the MIPS eligible clinician for the MIPS payment year. Section II.E.5 of this rule proposes the measures and activities for each of the four MIPS performance categories: Quality, resource use, CPIA, and advancing care information. This section proposes the performance standards for the measures and activities for each of the four performance categories under section 1848(q)(3) of the Act, the methodology for determining a score for each of the four performance categories (referred to as a “performance category score”), and the methodology for determining a CPS under section 1848(q)(5) of the Act based on the scores determined for each of the four performance categories. The performance category score is defined at § 414.1305 as the assessment of each MIPS eligible clinician's performance on the applicable measures and activities for a performance category for a performance period based on the performance standards for those measures and activities. Section II.E.7. includes proposals for determining the MIPS adjustments factors based on the CPS.

As noted in section II.E.2., we propose to use multiple identifiers to allow MIPS eligible clinicians to be measured as individuals, or collectively as part of a group or an APM Entity group (an APM Entity participating in a MIPS APM). Further, in section II.E.5.a.2., we propose that data for all four MIPS performance categories would be submitted using the same identifier (either individual or group) and that the CPS would be calculated using the same identifier. The scoring proposals in this section II.E.6. would be applied in the same manner for either individual submissions, proposed as TIN/NPI, or for the group submissions using the TIN identifier. Unless otherwise noted, for purposes of this section, the term “MIPS eligible clinician” will refer to both individual and group reporting and scoring, but will not refer to an APM Entity group.

APM Entity group reporting and scoring for MIPS eligible clinicians participating in MIPS APMs are described in section II.E.5.h. of this proposed rule. All eligible clinicians that participate in APMs are considered MIPS eligible clinicians unless and until they are determined to be either QPs or Partial QPs who elect not to report under MIPS, and excluded from MIPS. For the APM scoring standard to apply to a MIPS eligible clinician, the eligible clinician must be listed as a participant in the APM Entity that participates in a MIPS APM as of December 31 of the performance period, as described in section II.E.5.h. CMS will publish a list of MIPS APMs on the CMS Web site in advance of the performance period.

MIPS eligible clinicians who participate in APMs that are not MIPS APMs would report to MIPS as an individual MIPS eligible clinician or group. Unless otherwise specified, the proposals in this section II.E.6 that relate to reporting and scoring of measures and activities do not affect the APM scoring standard.

Our rationale for our scoring methodology is grounded in the understanding that the MIPS scoring system is a complex system with numerous moving parts. Thus, we believe it is necessary to set up key parameters around scoring, including requiring MIPS eligible clinicians to report at the individual or group level across all performance categories and generally to submit information for a performance category using a single submission mechanism. Too many different permutations would create additional complexities that could create confusion amongst MIPS eligible clinicians as to what is and is not allowed.

a. Converting Measures and Activities Into Performance Category Scores

(1) Policies That Apply Across Multiple Performance Categories

The detailed policies for scoring the four performance categories are described in this section II.E.6.a. of this rule. However, as the four performance categories collectively create a single MIPS CPS, there are some cross-cutting policies that we propose to apply to multiple performance categories.

(a) Performance Standards

Section 1848(q)(3)(A) of the Act requires the Secretary to establish performance standards for the measures and activities in the four MIPS performance categories. Section 1848(q)(3)(B) of the Act requires the Secretary, in establishing performance standards for measures and activities for the four MIPS performance categories, to consider historical performance standards, improvement, and the opportunity for continued improvement. We propose to define the term, performance standards, at § 414.1305 as the level of performance and methodology that the MIPS eligible clinician is assessed on for a MIPS performance period at the measures and activities level for all MIPS performance categories. We define the term, MIPS payment year at § 414.1305 as the calendar year in which MIPS payment adjustments are be applied. Performance standards for each performance category are proposed in more detail later in this section, II.E.6. MIPS eligible clinicians would know the actual performance standards in advance of the performance period, when possible. Further, each performance category is unified under the principle that MIPS eligible clinicians would know, in advance of the performance period, the methodology for determining the performance standards and the methodology that would be used to score their performance. Table 16 summarizes the performance standards, which are proposed in more detail in section II.E.6.a.

(b) Unified Scoring System

Section 1848(q)(5)(A) of the Act requires the Secretary to develop a methodology for assessing the total performance of each MIPS eligible clinician according to performance standards for applicable measures and activities in each performance category applicable to the MIPS eligible clinician for a performance period. While MIPS has four different performance categories, we propose a unified scoring system that enables MIPS eligible clinicians, beneficiaries, and stakeholders to understand what is required for a strong performance in MIPS while being consistent with statutory requirements. We sought to keep the scoring as simple as possible, while providing flexibility for the variety of practice types and reporting options. We would incorporate the following characteristics into the proposed scoring methodologies for each of the four MIPS performance categories:

• For the quality and resource use performance categories, all measures would be converted to a 10-point scoring system which provides a framework to universally compare different types of measures across different types of MIPS eligible clinicians. A similar point framework has been successfully implemented in several other CMS quality programs including the Hospital Value-Based Purchasing Program (HVBP).
• The measure and activity performance standards would be published, where feasible, before the performance period begins, so that MIPS eligible clinicians can track their performance during the performance period. This transparency would make the information more actionable to MIPS eligible clinicians.
• Unlike the PQRS or the EHR Incentive Program, we generally would not include “all-or-nothing” reporting requirements for MIPS. The methodology would score measures and activities that meet certain standards defined in section II.E.5 and this section. However, section 1848(q)(5)(B)(i) of the Act provides that under the MIPS scoring methodology, MIPS eligible clinicians who fail to report on an applicable measure or activity that is required to be reported shall be treated as receiving the lowest possible score for the measure or activity. Therefore, MIPS eligible clinicians that fail to report specific measures or activities would receive zero points for each required measure or activity that they do not submit to MIPS.
• The scoring system would ensure sufficient reliability and validity, by only scoring the measures that meet certain standards (such as required case minimum). The standards are described later in this section.
• The scoring proposals provide incentives for MIPS eligible clinicians to invest and focus on certain measures and activities that meet high priority policy goals such as improving beneficiary health, improving care coordination through health information exchange, or encouraging APM Entity participation.
• Performance at any level would receive points towards the performance category scores.

For the first year of MIPS, there are some minor differences in the proposed performance category scoring methodologies to account for differences in the maturity of the data collection systems and the measures and activities; however, we anticipate that the scoring in future years would continue to align and simplify. We request comment on the characteristics of the proposed unified scoring system.

We also propose at § 414.1325 that MIPS eligible clinicians and groups may elect to submit information via multiple mechanisms; however, they must use the same identifier for all performance categories and they may only use one submission mechanism per performance category. For example, a MIPS eligible clinician could use one submission mechanism for sending quality measures and another for sending CPIA data, but a MIPS eligible clinician could not use two submission mechanisms for a single performance category, such as submitting three quality measures via claims and three quality measures via registry. We do intend to allow flexibility, for example, in rare situations where a MIPS eligible clinician submits data for a performance category via multiple submission mechanisms (for example, submits data for the quality performance category through a registry and QCDR), we would score all the options and use the highest performance category score for the eligible clinician.

In carrying out MIPS, section 1848(q)(1)(E) of the Act requires the Secretary to encourage the use of QCDRs under section 1848(m)(3)(E) of the Act. In addition, section 1848(q)(5)(B)(ii) of the Act provides that under the methodology for assessing the total performance of each MIPS eligible clinician, the Secretary shall encourage MIPS eligible clinicians to report on applicable measures under the quality performance category through the use of CEHRT and QCDRs. To encourage the use of QCDRs, we have created opportunities for QCDRs to report new and innovative quality measures. In addition, several CPIAs emphasize QCDR participation. Finally, we propose under section II.E.5.a. for QCDRs to be able to submit data on all MIPS performance categories. We believe these flexible options would allow MIPS eligible clinicians to meet the submission criteria for MIPS in a low burden manner, which in turn may positively affect their CPS.

In addition, section 1848(q)(5)(D) of the Act lays out the requirements for incorporating performance improvement into the MIPS scoring methodology beginning with the second MIPS performance period, if data sufficient to measure improvement is available. Section 1848(q)(5)(D)(ii) of the Act also provides that achievement may be weighted higher than improvement. Stated generally, we consider achievement to mean how a MIPS eligible clinician performs compared to other MIPS eligible clinicians for each applicable measure and activity in a performance category, and improvement to mean how a MIPS eligible clinician performs compared to the MIPS eligible clinician's own previous performance on measures and activities in a performance category. Improvement would not be scored for the first year of MIPS, but we seek comment on how best to incorporate improvement scoring for all performance categories.

(c) Baseline Period

In other Medicare quality programs, such as the HVBP, we have adopted a baseline period that occurs prior to the performance period for a program year to measure improvement and to establish performance standards. We view the MIPS Program as necessitating a similar baseline period for the quality performance category. We intend to establish a baseline period for each performance period for a MIPS payment year to measure improvement for the quality performance category and to enable us to calculate performance standards that we can establish and announce prior to the performance period. As with the HVBP, we intend to adopt baseline periods that are as close as possible in duration to the performance period specified for a MIPS payment year. In addition, evaluating performance compared to a baseline period may enable other payers to incorporate MIPS benchmarks into their programs. For each MIPS payment year, we propose at § 414.1380 that the baseline period would be two years prior to the performance period for the MIPS payment year. Therefore, for the first MIPS payment year (CY 2019 payment adjustments), for the quality performance category, we propose that the baseline period would be calendar year 2015 which is 2 years prior to the proposed calendar year 2017 performance period. As discussed in section II.E.6.a.2.a. we propose to use performance in the baseline period to set benchmarks for the quality performance category, with the exception of new measures for which we would set the benchmarks using performance in the performance period. For the resource use performance category, we propose to set the benchmarks using performance in the performance period and not the baseline period, as discussed in section II.E.6.a.3. For the resource use performance category, we also have included an alternative proposal to set the benchmarks using performance in the baseline period. We define the term “measure benchmark” for the quality and resource use performance categories at § 414.1305 as the level of performance that the MIPS eligible clinician will be assessed on for a performance period at the measures level.

(2) Scoring the Quality Performance Category

In section II.E.5.b.3, we proposed multiple ways that MIPS eligible clinicians may submit data for the quality performance category to MIPS; however, we propose that the scoring methodology would be consistent regardless of how the data is submitted. In summary, we propose at § 414.1380(b)(1) to assign 1-10 points to each measure based on how a MIPS eligible clinician's performance compares to benchmarks. Measures must have the required case minimum to be scored. If a MIPS eligible clinician fails to submit a measure required under the quality performance category criteria, then the MIPS eligible clinician would receive zero points for that measure. MIPS eligible clinicians would not receive zero points if the required measure is submitted (meeting the data completeness criteria as defined in section II.E.5.b.3.b.) but is unable to be scored for any of the reasons listed in this section II.E.6.a.2., such as not meeting the required case minimum or a measure lacks a benchmark). For example, if a MIPS eligible clinician reports a measure that meets the requirements specified in section II.E.5.b., but that measure does not meet the required case minimum criteria or lacks a benchmark, then the measure would not be scored under the MIPS quality performance category, whereas a MIPS eligible clinician that did not report this measure would have the measure scored as a zero. We describe in section II.E.6a.2.d. examples of how points would be allocated and how to compute the overall quality performance category score under these scenarios. Bonus points would be available for reporting high priority measures, defined as outcome, appropriate use, efficiency, care coordination, patient safety, and patient experience measures.

As discussed in section II.E.6.a.2.g., the quality performance category score would be the sum of all the points assigned for the scored measures required for the quality performance category plus the bonus points (subject to the cap) divided by the sum of total possible points. Since MIPS eligible clinicians would be generally required to submit six measures or six measures from a specialty measure set and we would also score MIPS eligible clinicians on up to three population-based measures calculated from administrative claims data as discussed in section II.5.b.6, the total possible points for the quality performance category would be 90 points (6 submitted measures × 10 points + 3 population-based measures × 10 points = 90). However, for eligible groups reporting via CMS Web Interface, the total possible points for the quality performance category would be 210 points (17 measures × 10 points + 3 population-based measures × 10 points = 200), subject to CMS Web Interface reporting criteria. Further, the total possible points for small groups of less than 10 would be 80 points (6 submitted measures × 10 points + 2 population-based measures × 10 points = 80) because under our proposals the all-cause hospital readmissions measure would not be applicable to groups of less than 10 MIPS eligible clinicians and MIPS eligible clinicians reporting as individuals due to reliability concerns. Therefore, small groups of less than 10 and MIPS eligible clinicians reporting as individuals would only be scored on two population-based measures.

In section II.E.6.b, we discuss how we would score MIPS eligible clinicians who do not have any scored measures in the quality performance category. The details of the proposed scoring methodology for the quality performance category are described below.

(a) Quality Measure Benchmarks

For the quality performance category, we propose at § 414.1380(b)(1) that the performance standard is measure-specific benchmarks. Benchmarks would be determined based on performance on measures in the baseline period. For quality performance category measures for which there are baseline period data, we would calculate an array of measure benchmarks based on performance during the baseline period, breaking baseline period measure performance into deciles. Then, a MIPS eligible clinician's actual measure performance during the performance period would be evaluated to determine the number of points that should be assigned based on where the actual measure performance falls within these baseline period benchmarks. If a measure does not have baseline period information, (for example, new measures) or if the measure specifications for the baseline period differ substantially from the performance period (for example, when the measure requirements change due to updated clinical guidelines), then we would determine the array of benchmarks based on performance on the measure in the performance period, breaking the actual performance on the measure into deciles. In addition, we propose to create separate benchmarks for submission mechanisms that do not have comparable measure specifications. For example, several electronic clinical quality measures have specifications that are different than the corresponding measure from registries. We propose to develop separate benchmarks for EHR submission options, claims submission options, Qualified Clinical Data Registries (QCDRs) and qualified registries submission options.

For CMS Web Interface reporting, we propose to use the benchmarks from the Shared Savings Program as described at https://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/sharedsavingsprogram/Quality-Measures-Standards.html . We would adopt the Shared Savings Program performance year benchmarks for measures that are reported through the CMS Web Interface for the MIPS performance period, but would apply the MIPS method of assigning 1 to 10 points to each measure. For example, for the 2017 MIPS performance year, we would use the benchmarks for the 2017 Shared Savings Program performance year, as both the MIPS performance period and the Shared Savings Program performance year use a calendar year for CMS Web Interface reporting. Because the Shared Savings Program does not create benchmarks below the 30th percentile, we would assign all scores below the 30th percentile a value of 2 points, which is consistent with the mid-cluster approach we are proposing for topped out measures. We believe using the same benchmarks for MIPS and the Shared Savings Program for the CMS Web Interface measures would be appropriate because, as is discussed in II.E.5.h., we propose to use the MIPS benchmarks to score the Shared Savings Program and the Next Generation ACO Model on the quality performance category and believe it is important to not have conflicting benchmarks. We would post the MIPS CMS Web Interface benchmarks with the other MIPS benchmarks.

As an alternative approach, we considered creating CMS Web Interface specific benchmarks for MIPS. This alternative would be restricted to CMS Web Interface reporters and would not include other MIPS data submission methods, which are currently used to create the Shared Saving Program benchmarks. This alternative would also apply the topped out cluster approach if any measures are topped out. While we see benefit in having CMS Web Interface methodology match the other MIPS benchmarks, we are also concerned about the Shared Saving Program and the Next Generation ACO Model participants having conflicting benchmark data. We request comments on building CMS Web Interface specific benchmarks.

All MIPS eligible clinicians, regardless of whether they report as an individual or group, and regardless of specialty, that submit data using the same submission mechanism would be included in the same benchmark. We propose to unify the calculation of the benchmark by using the same approach as the VM of weighting the performance rate of each MIPS eligible clinician and group submitting data on the quality measure by the number of beneficiaries used to calculate the performance rate so that group performance is weighted appropriately (77 FR 69321-69322). We would also include APM Entity submissions in the benchmark but would not score APM Entities using this methodology. For APM scoring, we refer to section II.E.5.h.

To ensure that we have robust benchmarks, we propose that each benchmark must have a minimum of 20 MIPS eligible clinicians who reported the measure meeting the data completeness requirement defined in section II.E.5.b.3, as well as meeting the required case minimum criteria for scoring that is defined later in this section. We selected a minimum of 20 because, as discussed below, our benchmarking methodology relies on assigning points based on decile distributions with decimals. A decile distribution requires at least 10 observations. We doubled the requirement to 20 so that we would be able to assign decimal point values and minimize cliffs between deciles. We did not want to increase the benchmark sample size requirement due to concerns that an increase could limit the number of measures with benchmarks.

We also propose that MIPS eligible clinicians who report measures with a performance rate of 0 percent would not be included in the benchmarks. In our initial analysis, we identified some measures that had a large cluster of eligible clinicians with a 0 percent performance rate. We are concerned that the 0 percent performance rate represents clinicians who are not actively engaging in that measurement activity. For example, it could be clinicians reporting the measures that are programmed into their EHR and that are submitted unintentionally, rather than measures the eligible clinician has actively selected for quality improvement. We do not want to inappropriately skew the distribution. We seek comment on whether or not to include 0 percent performance in the benchmark.

We propose at § 414.1380(b)(1)(i) to base the benchmarks on performance in the baseline period when possible, and to publish the numerical benchmarks when possible, prior to the start of the performance period. In those cases where we do not have comparable data from the baseline period, we propose to use information from the performance period to establish benchmarks. While the benchmark methodology would be established in a final rule in advance of the performance period, the actual numerical benchmarks would not be published until after the performance period for quality measures that do not have comparable data from the baseline period. The methodology for creating the benchmarks is discussed below in this section.

We considered not scoring measures that either are new to the MIPS program or do not have a historical benchmark based on performance in the baseline period. This policy would be consistent with the VM policy in which we do not score measures that have no benchmark (77 FR 69322). However, we are concerned that such a policy could stifle reporting on innovative new measures because it would take several years for the measure to be incorporated into the performance category score. We also believe that any issues related to reporting a new measure would not disproportionately affect the relative performance between MIPS eligible clinicians.

We also considered a variation on the scoring methodology that would provide a floor for a new MIPS measure. Under this variation, if a MIPS eligible clinician reports a new measure under the quality performance category, the MIPS eligible clinician would not score lower than 3 points for that measure. This would encourage reporting on new measures, but also prevent MIPS eligible clinicians from receiving the lowest scores for a new measure, while still measuring variable performance. Finally, we also considered lowering the weight of a new measure, so that new measures would contribute relatively less to the score compared to other measures. In the end, we are not proposing these alternatives we considered, because we want to encourage adoption and measured performance of new measures, however, we do request comment on these alternatives, including comments on what the lowest score should be for MIPS eligible clinicians who report a new measure under the quality performance category and protections against potential gaming related to reporting of new measures only. We also seek comments on alternative methodologies for scoring new measures under the quality performance category, which would assure equity in scoring between the methodology for measures for which there is baseline period data and for new measures which do not have baseline period data available.

Finally, we want to clarify that some PQRS reporting mechanisms have limited experience with all-payer data. For example, under PQRS, all-payer data was permitted only when reporting via registries for measure groups; reporting via registries for individual measures was restricted to Medicare only. Under MIPS however, we intend to have more robust data submissions, as described in section II.E.5.b.3. We recognize that comparing all-payer performance to a benchmark that is built, in part, on Medicare data is a limitation and would monitor the benchmarks to see if we need to develop separate benchmarks. This data issue would resolve in a year or two, as new MIPS data becomes the historical benchmark data in future years.

(b) Assigning Points Based on Achievement

We propose at § 414.1380(b)(1)(x) to establish benchmarks using a percentile distribution, separated by decile categories, because it translates measure-specific score distributions into a uniform distribution of MIPS eligible clinicians based on actual performance values. For each set of benchmarks, we propose to calculate the decile breaks for measure performance and assign points for a measure based on which benchmark decile range the MIPS eligible clinician's performance rate on the measure falls between. For example, MIPS eligible clinicians in the top decile would receive 10 points for the measure, and MIPS eligible clinicians in the next lower decile would receive points ranging from 9 to 9.9. We propose to assign partial points to prevent performance cliffs for MIPS eligible clinicians near the decile breaks. The partial points would be assigned based on the percentile distribution.

Table 17 illustrates an example of using decile points along with partial points to assign achievement points for a sample quality measure. The methodology in this example could apply to measures where the benchmark is based on the baseline period or for new measures where the benchmark is based on the performance period.

In the example above, a MIPS eligible clinician with a measure performance rate of 41 percent would receive 6.0 points based on the benchmark. MIPS eligible clinicians with measure performance rates of 85 percent or above would receive 10 points because they were in the top benchmark decile. We believe that MIPS eligible clinicians within the top decile in performance would warrant receiving the maximum number of points. This is a similar concept to the HVBP “benchmark” level. We note that 85 percent is solely illustrative. Any MIPS eligible clinician who reports some level of performance would receive a minimum of one point for reporting if the measure has the required case minimum, assuming the measure has a benchmark.

In Table 17 we described our scoring approach, using deciles. We do not propose to base scoring on decile distributions for the same measure ranges as described in Table 17 when performance is clustered at the high end (that is, “topped out” measures), as true variance cannot be assessed. MIPS eligible clinicians report on different measures and often elect to submit measures on which they expect to perform well. With MIPS eligible clinicians electing to report on measures where they expect to perform well, we anticipate many measures would have performance distributions clustered near the top. We propose to identify “topped out” measures by using a definition similar to the definition used in the HVBP: Truncated Coefficient of Variation is less than 0.10 and the 75th and 90th percentiles are within 2 standard errors; or median value for a process measure that is 95 percent or greater (80 FR 49550).

Using 2014 PQRS quality reported data measures, we modeled the proposed benchmark methodology and identified that approximately half of the measures proposed under the quality performance category are topped out. Several measures have a median score of 100 percent, which makes it difficult to assess relative performance needed for the quality performance category score.

However, we do not believe it would be appropriate to remove topped out measures at this time. As not all MIPS eligible clinicians would be required to report these measures under our proposals for the quality performance category in section II.E.5.b. it would be difficult to determine whether a measure is truly topped out or if only excellent performers are choosing to report the measure. We also believe removing such a large volume of measures would make it difficult for some specialties to have enough applicable measures to report. At the same time, we do not believe that the highest values on topped out measures convey the same meaning of relative quality performance as the highest values for measures that are not topped out. In other words, we do not believe that eligible clinicians electing to report topped out process measures should be able to receive the same maximum score as eligible clinicians electing to report preferred measures, such as outcome measures.

Therefore, we propose to modify the benchmark methodology for topped out measures. Rather than assigning up to 10 points per measure, we propose to limit the maximum number of points a topped out measure can achieve based on how clustered the scores are. We propose to identify clusters within topped out measures and would assign all MIPS eligible clinicians within the cluster the same value, which would be the number of points available at the midpoint of the cluster. That is, we would take the midpoint of the highest and lowest scores that would pertain if the measure was not topped out and the values were not clustered. We would only apply this methodology for benchmarks based on the baseline period. When we develop the benchmarks, we would identify the clusters and state the points that would be assigned when the measure performance rate is in a cluster. We would notify MIPS eligible clinicians when those benchmarks are published with regard to which measures are topped out.

Table 18 illustrates this hypothetical example. In developing the benchmark, we identified that the top five deciles (50 percent of eligible clinicians reporting the measure) of MIPS eligible clinicians are clustered at 100 percent. We would identify the middle of that cluster (in this example, the top 25 percent or the middle of the eighth decile) and then assign all MIPS eligible clinicians with performance rates in the cluster the same number of points for the measure. The decile points for the hypothetical topped out measure in Table 18 shows that the maximum a MIPS eligible clinician can receive for the topped out measure is 8.5 points in this example.

We propose this approach because we want to encourage MIPS eligible clinicians not to report topped out measures, but to instead choose other measures that are more meaningful. We also seek feedback on alternative ways and an alternative scoring methodology to address topped out measures so that topped out measures do not disproportionately affect a MIPS eligible clinician's quality performance category score. Other alternatives could include placing a limit on the number of topped out measures MIPS eligible clinicians may submit or reducing the weight of topped out measures. We also considered whether we should apply a flat percentage in building the benchmarks, similar to the Shared Savings Program, where MIPS eligible clinicians are scored on their percentage of their performance rate and not on a decile distribution and request comment on how to apply such a methodology without providing an incentive to report topped out measures. Under the Shared Savings Program, 42 CFR 425.502, there are circumstances when benchmarks are set using flat percentages. For some measures, benchmarks are set using flat percentages when the 60th percentile was equal to or greater than 80.00 percent, effective beginning with the 2014 reporting year (78 FR 74759-74763). For other measures benchmarks are set using flat percentages when the 90th percentile was equal to or greater than 95.00 percent, effective beginning in 2015 (79 FR 67925). Flat percentages allow those with high scores to earn maximum or near maximum quality points while allowing room for improvement and rewarding that improvement in subsequent years. Use of flat percentages also helps ensure those with high performance on a measure are not penalized as low performers. We also note that we anticipate removing topped out measures over time, as we work to develop new quality measures that will eventually replace these topped out measures. We request feedback on these proposals.

(c) Case Minimum Requirements and Measure Reliability and Validity

We seek to ensure that MIPS eligible clinicians are measured reliably; therefore, we propose at § 414.1380(b)(1)(v) to use for the quality performance category measures the case minimum requirements for the quality measures used in the 2018 VM (see § 414.1265): 20 cases for all quality measures, with the exception of the all-cause hospital readmissions measure, which has a minimum of 200 cases. We refer readers to Table 46 of the CY 2016 PFS final rule (80 FR 71282) which summarized our analysis of the reliability of certain claims-based measures used for the 2016 VM payment adjustment. MIPS eligible clinicians that report measures with fewer than 20 cases (and the measure meets the data completeness criteria) would receive recognition for submitting the measure, but the measure would not be included for MIPS quality performance category scoring. Since the all-cause hospital readmissions measure does not meet the threshold for what we consider to be moderate reliability for solo practitioners and groups of less than ten MIPS eligible clinicians for purposes of the VM (see Table 46 of the CY 2016 PFS final rule, referenced above), for consistency, we propose to not include the all-cause hospital readmissions measure in the calculation of the quality performance category for MIPS eligible clinicians who individually report, as well as solo practitioners or groups of two to nine MIPS eligible clinicians.

We also propose that if we identify issues or circumstances that would impact the reliability or validity of a measure score, we would also exclude those measures from scoring. For example, if we discover that there was an unforeseen data collection issue that would affect the integrity of the measure information, we would not want to include that measure in the quality performance category score. If a measure is excluded, we would recognize that the measure had been submitted and would not disadvantage the MIPS eligible clinicians by assigning them zero points for a non-reported measure. In this instance, if the MIPS eligible clinician, as a solo practitioner, scored 10 out of 10 on each of the remaining five measures submitted, and the two population-based measures applicable to solo practitioners, the MIPS eligible clinician would receive a perfect score in the quality performance category (5 measures × 10 points) + (2 population-based measures × 10 points) or 70 out of 70 possible points.

(d) Scoring for MIPS Eligible Clinicians that Do Not Meet Quality Performance Category Criteria

Section II.E.5.b. of this proposed rule outlines our proposed quality performance category criteria for the different reporting mechanisms. The criteria vary by reporting mechanism, but generally we propose to include a minimum of six measures with at least one cross-cutting measure (for patient facing MIPS eligible clinicians) (Table C) and an outcome measure if available. If an outcome measure is not available, then the eligible clinician would report one other high priority measure (appropriate use, patient safety, efficiency, patient experience, and care coordination measures) in lieu of an outcome measure. MIPS eligible clinicians and groups would have to select their measures from either the list of all MIPS Measures in Table A or a set of specialty specific measures in Table E.

We note that there are some special scenarios for those MIPS eligible clinicians who select their measures from the Specialty Sets (Table E) as discussed in section II.E.5.b.

For groups using the CMS Web Interface and MIPS APMs, we propose to have different quality performance category criteria described in sections II.E.5.b. and II.E.5.h. Additionally, as described in section II.E.5.b. we also propose to score MIPS eligible clinicians on up to three population-based measures.

Previously in PQRS, EPs had to meet all the criteria or be subject to a negative payment adjustment. We heard from numerous commenters a desire to move away from “all-or-nothing” scoring. Therefore, in MIPS, we propose that MIPS eligible clinicians receive credit for measures that they report, regardless of whether or not the MIPS eligible clinician meets the quality performance category submission criteria. Section 1848(q)(5)(B)(i) of the Act provides that under the MIPS scoring methodology, MIPS eligible clinicians who fail to report on an applicable measure or activity that is required to be reported shall be treated as receiving the lowest possible score for the measure or activity; therefore, for any MIPS eligible clinician who does not report a measure required to satisfy the quality performance category submission criteria, we propose that the MIPS eligible clinician would receive zero points for that measure. For example, a MIPS eligible clinician who is able to report on six measures, yet reports on four measures, would receive two “zero” scores for the missing measures. In another example, a patient facing MIPS eligible clinician reports more than six measures, but does not elect to report a cross-cutting measure and an outcome measure, or if one is not available, another high priority measure. The MIPS eligible clinician in that scenario would receive at least two “zero” scores for not reporting measures required by the quality performance category criteria.

However, MIPS eligible clinicians who report a measure that does not meet the required case minimum would not be scored on the measure but would also not receive a “zero” score. For example, a MIPS eligible clinician who submits six measures as part of a group with 10 or more clinicians, one of which does not meet the required case minimum, would be scored on the five remaining measures and the three population-based measures based on administrative claims data. If the MIPS eligible clinician scored 10 out of 10 on each of these measures, the MIPS eligible clinician would receive a perfect score in the quality performance category (5 measures × 10 points) + (3 population-based measures × 10 points) or 80 out of 80 possible points.

We also note that if MIPS eligible clinicians are able to submit measures that can be scored, we want to discourage them from continuing to submit the same measures year-after-year that cannot be scored due to not meeting the required case minimum. Rather, to the fullest extent possible, MIPS eligible clinicians should select measures that would have a required case minimum. We seek comment on any safeguards we should implement in future years to minimize any gaming attempts. For example, if the measures that a MIPS eligible clinician submits for a performance period are not able to be scored due to not meeting the required case minimum, we seek comment on whether we should require these MIPS eligible clinicians to submit different measures with sufficient cases for the next performance period (to the extent other measures are applicable and available to them).

MIPS eligible clinicians who report a measure where there is no benchmark due to less than 20 MIPS eligible clinicians reporting on the measure would not be scored on the measure but would also not receive a “zero” score. Instead, these MIPS eligible clinicians would be scored according to the following example: A MIPS eligible clinician who submits six measures through a group of 10 or more clinicians, with one measure lacking a benchmark, would be scored on the five remaining measures and the three population-based measures based on administrative claims data. If the MIPS eligible clinician scored 10 out of 10 on each of these measures, the MIPS eligible clinician would receive a perfect score in the quality performance category (5 measures × 10 points) + (3 population-based measures × 10 points) or 80 out of 80 possible points.

We intend to develop a validation process to review and validate a MIPS eligible clinician's inability to report on the quality performance requirements as proposed in section II.E.5.b. We anticipate that this process would function similar to the Measure Applicability Validity (MAV) process that occurred under PQRS, with a few exceptions. First, the MAV process under PQRS was a secondary process after an EP was determined to not be a satisfactory reporter. Under MIPS, we intend to build the process into our overall scoring approach to reduce confusion and burden on MIPS eligible clinicians by having a separate process. Second, as the requirements under PQRS are different than those proposed under MIPS, the process must be updated to account for different measures and different quality performance requirements. More information on the MAV process under PQRS can be found at http://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/PQRS/Downloads/2016_PQRS_MAV_ProcessforClaimsBasedReporting_030416.pdf . We request comments on these proposals.

(e) Incentives To Report High Priority Measures

Consistent with other CMS value-based payment programs, we propose that MIPS scoring policies would emphasize and focus on high priority measures that impact beneficiaries. These high priority measures are defined as outcome, appropriate use, patient safety, efficiency, patient experience and care coordination measures; see Tables A-D for these measures. We propose these measures as high priority measures given their critical importance to our goals of meaningful measurement and our measure development plan. We note that many of these measures are grounded in NQS domains. For patient safety, efficiency, patient experience and care coordination measures, we refer to the measures within the respective NQS domains and measure types. For outcomes measures, we include both outcomes measures and intermediate outcomes measures. For appropriate use measures, we have noted which measures fall within this category in Tables A-D and provided criteria for how we identified these measures in section II.E.5.b. For non-MIPS measures reported through QCDRs, we propose to classify which measures are high priority during the measure review process.

We are proposing scoring adjustments to create incentives for MIPS eligible clinicians to submit certain high priority measures and to allow these measures to have more impact on the total quality performance category score.

We propose to create an incentive for MIPS eligible clinicians to voluntarily report additional high priority measures. We propose to provide two bonus points for each outcome and patient experience measure and one bonus point for other high priority measures reported in addition to the one high priority measure (an outcome measure, but if one is not available, then another high priority measure) that would already be required under the proposed quality performance category criteria. For example, if a MIPS eligible clinician submitted two outcome measures, and two patient safety measures, the MIPS eligible clinician would receive two bonus points for the second outcome measure reported and two bonus points for the two patient safety measures. The MIPS eligible clinician would not receive any bonus points for the first outcome measure submitted since that is a required measure. We selected two bonus points for outcome measures given the statutory requirements under section 1848(q)(2)(C)(i) of the Act to emphasize outcome measures. We selected two bonus points for patient experience measures given the importance of patient experience measures to our measurement goals. We selected one bonus point for all other high priority measures given our measurement goals around each of those areas of measurement. We believe the number of bonus points provides extra credit for submitting the measure, yet would not mask poor performance on the measure. For example, a MIPS eligible clinician with poor outcomes receives only two points for performance for a particular high priority measure. The bonus points would increase the MIPS eligible clinician's points to three (or four if the measure is an outcome measure or patient experience measure), but that amount is far less than the ten points a top performer would receive. We note that population-based measures would not receive bonus points.

We note that a MIPS eligible clinician who submits a high priority measure but had a performance rate of 0 percent would not receive any bonus points. Eligible clinicians would only receive bonus points if the performance rate is greater than zero. Bonus points are also available for measures that are not scored (not included in the top 6 measures for the quality performance category score) as long as the measure has the required case minimum and data completeness. We believe these qualities would allow us to include the measure in future benchmark development.

For groups submitting data through the CMS Web Interface, including MIPS APMs that report through the CMS Web Interface, groups are required to submit a set of predetermined measures and groups are unable to submit additional measures. For that submission mechanism, we propose to apply bonus points based on the finalized set of measures. We would assign two bonus points for each outcome measure (after the first required outcome measure) and for each patient experience measure. We would also have one additional bonus point for each other high priority measure (patient safety, efficiency, appropriate use, care coordination). We believe MIPS eligible clinicians or groups should have the ability to receive bonus points for reporting high priority measures through all submission mechanisms, including the CMS Web Interface. In the final rule, we will publish how many bonus points the CMS Web Interface measure set would have available based on the final list of measures.

We propose to cap the bonus points for the high priority measures (outcome, appropriate use, patient safety, efficiency, patient experience, and care coordination measures) at 5 percent of the denominator of the quality performance category score. Tables 19 and 20 illustrate examples of how to calculate the bonus cap. We also propose an alternative approach of capping bonus points for high priority measures at 10 percent of the denominator of the quality performance category score. Our rationale for the 5 percent cap is that we do not want to mask poor performance by allowing an MIPS eligible clinician to perform poorly on a measure but still obtain a high quality performance category score by submitting numerous high priority measures in order to obtain bonus points; however, we are also concerned that 5 percent may not be enough incentive to encourage reporting. We request comment on the appropriate threshold for this bonus cap.

(f) Incentives To Use CEHRT To Support Quality Performance Category Submissions

Section 1848(q)(5)(B)(ii) of the Act provides that under the methodology for assessing the total performance of each MIPS eligible clinician, the Secretary shall: (I) Encourage MIPS eligible clinicians to report on applicable measures under the quality performance category through the use of CEHRT and QCDRs; and (II) with respect to a performance period for a year, for which a MIPS eligible clinician reports applicable measures under the quality performance category through the use of CEHRT, treat the MIPS eligible clinician as satisfying the clinical quality measures reporting requirement under section 1848(o)(2)(A)(iii) of the Act for such year. To encourage the use of CEHRT for quality improvement and reporting on measures under the quality performance category, we are proposing a scoring incentive to MIPS eligible clinicians who use their CEHRT systems to capture and report quality information.

We propose to allow one bonus point under the quality performance category score, up to a maximum of 5 percent of the denominator of the quality performance category score if:

• The MIPS eligible clinician uses CEHRT to record the measure's demographic and clinical data elements in conformance to the standards relevant for the measure and submission pathway, including but not necessarily limited to the standards included in the CEHRT definition proposed in 414.1305;
• The MIPS eligible clinician exports and transmits measure data electronically to a third party using relevant standards or directly to CMS using a submission method as defined at § 414.1325; and
• The third party intermediary (for example, a QCDR) uses automated software to aggregate measure data, calculate measures, perform any filtering of measurement data, and submit the data electronically to CMS using a submission method as defined at § 414.1325.

These requirements are referred to as “end-to-end electronic reporting.”

We note that this bonus would be in addition to the high priority bonus. MIPS eligible clinicians would be eligible for both this bonus option and the high priority bonus option with separate bonus caps for each option. We also propose an alternative approach of capping bonus points for this option at 10 percent of the denominator of the quality performance category score. Our rationale for the 5 percent cap is that we do not want to mask poor performance by allowing a MIPS eligible clinician to perform poorly on a measure but still obtain a high quality performance category score by submitting numerous measures in order to obtain bonus points; however, we are also concerned that 5 percent may not be enough incentive to encourage end-to-end electronic reporting. We seek comment on the appropriate threshold for this bonus cap. We propose the CEHRT bonus would be available to all submission mechanisms except claims submissions. This incentive would also be available for MIPS APMs reporting through the CMS Web Interface. Specifically, MIPS eligible clinicians who report via qualified registries, QCDRs, EHR submission mechanisms, and CMS Web Interface may receive one bonus point for each reported measure with a cap as described. We do not propose to allow this option for claims submission, because there is no mechanism for MIPS eligible clinicians to identify the information was pulled using an EHR.

This approach supports and encourages innovative approaches to measurement using the full array of standards ONC adopts, and the data elements MIPS eligible clinicians capture and exchange, to support patient care. Thus, approaches where a qualified registry or QCDR obtains data from a MIPS eligible clinician's CEHRT using any of the wide range of ONC-adopted standards and then uses automated electronic systems to perform aggregation, calculation, filtering, and reporting would qualify each such measure for the CEHRT bonus point. In addition, measures submitted using the EHR submission mechanism or the EHR submission mechanism through a third party would also qualify for the CEHRT bonus.

We request comment on this proposed approach.

(g) Calculating the Quality Performance Category Score

The next two subsections provide a detailed description of how the quality performance category score would be calculated under our proposals.

(i) Calculating the Quality Performance Category Score for Non-APM Entity, Non-CMS Web Interface Reporters

To calculate the quality performance category score, we propose at § 414.1380(b)(1)(xv) to sum the weighted points assigned for the measures required by the quality performance category criteria plus the bonus points and divide by the weighted sum of total possible points.

If a MIPS eligible clinician elects to report more than the minimum number of measures to meet the MIPS quality performance category criteria, then we would only include the scores for the measures with the highest number of assigned points. For example, if a patient facing MIPS eligible clinician's quality submission criteria is to report six measures with at least one cross-cutting measure and a high priority measure, and the MIPS eligible clinician reports eight process measures (three using CEHRT), one cross-cutting measure, and one outcome measure, then we propose to use the four process measures with the highest number of assigned points, plus the cross-cutting measure and the outcome measure, in addition to the two population-based measures (the all-cause readmission measure would not apply to an MIPS eligible clinician reporting individually), to calculate the quality performance category score. Allowing MIPS eligible clinicians to report additional measures without including them in the scoring allows MIPS eligible clinicians to become familiar with new measures and gain experience with those measures. It also provides the foundation for the MIPS eligible clinician to receive credit for improvement on those measures in future years.

If a MIPS eligible clinician has met the quality performance category submission criteria for reporting quality information, but does not have any scored measures as discussed in section II.E.6.b.2., then a quality performance category score would not be calculated. Refer to section II.E.6.a.2.d. for details on how we propose to address scenarios where a quality performance category score is not calculated for a MIPS eligible clinician.

The following example illustrates a sample scoring methodology. In this scenario, a MIPS eligible clinician submits individually via registry three process measures, one outcome measure, and one other high priority measure. Two of the process measures and one outcome measure qualify for the CEHRT bonus. The patient facing MIPS eligible clinician did not submit on an expected cross-cutting measure and therefore would receive zero points for that requirement. Measures that do not meet the required case minimum or do not have a benchmark are not used for scoring. We reiterate that a measure that is not scored due to not meeting the required case minimum or lack of a measure benchmark would be treated differently than a required measure that is not reported. Any required measure that is not reported, or reported in a way that does not meet the data completeness requirements, would receive a score of zero points and be considered a scored measure. Table 19 illustrates the example.

The total possible points for the eligible clinician is 70 points. The eligible clinician has 48.2 points based on performance. The eligible clinician also qualifies for one bonus point for reporting an additional high priority patient safety measure and three bonus points for end-to-end electronic reporting of quality measures. The bonus points for high priority measures and CEHRT reporting are both under two separate caps which is 5 percent of 70 possible points or 3.5 points per bonus category). The quality performance category score for this MIPS eligible clinician is (48.2 points + 4 bonus points = 52.2)/70 total possible points = 74.6 percent. The quality performance category score would be capped at 100 percent.

The following example in Table 20 illustrates how to calculate the bonus cap for the high priority measure bonus and the CEHRT bonus. In the scenario below, the MIPS eligible clinician has submitted six measures and would also be scored on two of the three population-based measures. The MIPS eligible clinician below successfully submitted five quality measures using end-to-end electronic reporting, and therefore, qualifies for the CEHRT bonus of one point for each of those measures. In addition to CEHRT bonus points, the MIPS eligible clinician reported outcome measures for high priority bonus points. The MIPS eligible clinician reported two outcome measures and receives two bonus points for the second outcome measure, given that no bonus points are given for the first required measure. However, both bonus categories are over the cap (which is 5 percent of 80 possible points or four points per bonus category). The quality performance category score for this MIPS eligible clinician is 68.8 (60.8 + 4 CEHRT bonus points after the cap + 4 high priority bonus points after the cap) or 86 percent (68.8/80). Note, in section II.E.5.b.(2), we propose to weight the quality performance category at 50 percent of the MIPS CPS, so an 86 percent quality performance category score would account for 50 percent of the CPS.

We request comment on our proposals to calculate the quality performance category score.

(ii) Calculating the Quality Performance Category for CMS Web Interface Reporters

CMS Web Interface reporters have different quality performance category submission criteria; therefore, we propose to modify our scoring logic slightly to accommodate this submission mechanism. CMS Web Interface users report on the entire set of measures specified for that mechanism. Therefore, rather than scoring the top six reported measures, we propose to score all measures. If a group does not meet the reporting requirements for one of the measures, then the group would receive zero points for that measure. We note that since groups reporting through the Web Interface are required to report on all measures, and since some of those measures are “high priority,” these groups would always have some bonus points for the quality performance category score if all the measures are reported. That is, the group would either report on less than all web interface measures, in which case the group would receive zeros for unreported measures, or the group would report on all measures, in which case the group would automatically be eligible for bonus points. The other proposals for scoring discussed in section II.E.6.a.2.g.i., including bonus points, would still apply for CMS Web Interface. We request comment on this proposal.

(h) Measuring Improvement

Section 1848(q)(3)(B) of the Act requires the Secretary, in establishing performance standards for measures and activities for the MIPS performance categories, to consider: Historical performance standards; improvement; and the opportunity for continued improvement. In addition, under section 1848(q)(5)(D) of the Act, beginning with the second year of the MIPS, if data sufficient to measure improvement are available, the CPS methodology shall take into account improvement of the MIPS eligible clinician in calculating the performance score for the quality and resource use performance categories and may take into account improvement for the CPIA and advancing care information performance categories.

We are soliciting public comments on potential ways to incorporate improvement into the scoring methodology moving forward. We are especially interested in feedback on the following three options, with the assumption that eligible clinicians would report the same measures year-to-year (where possible). We are also interested in feedback on how to score improvement given that a MIPS eligible clinician can change measures and submission mechanisms from year-to-year. In addition, a MIPS eligible clinician can elect to report as an individual or a member of a group and that election can vary from year to year. Finally, we seek feedback on whether to score improvement where MIPS eligible clinicians do not have the required case minimum for measures to be scored.

Option 1: We could adopt the approach for assessing improvement currently used for the HVBP, where we assign from 1-10 points for achievement and from 1-9 points for improvement for each measure. We would compare the achievement and improvement points for each measure in the quality performance category and score whichever is greater. Specifically, we would determine two scores for a MIPS eligible clinician at the measure level for the quality performance category. First, we would assess the MIPS eligible clinician's achievement score, which measures how the MIPS eligible clinician performed compared to benchmark performance scores for each applicable measure in the quality performance category. Second, we would assess the MIPS eligible clinician's improvement score, which measures how much a MIPS eligible clinician has improved compared to the MIPS eligible clinician's own previous performance during a baseline period for each applicable measure in the quality performance category. Under this methodology, we would compare the achievement and improvement scores for each measure and only use whichever is greater, but only those eligible clinicians with the top achievement would be able to receive the maximum number of points. If a MIPS eligible clinician's practice was not open during the baseline period but was open during the performance period, points would be awarded based on achievement only for that performance period. For a more detailed description of the HVBP methodology, we refer readers to § 412.160 and § 412.165.

Option 2: We could adopt the approach for assessing improvement currently used in the Shared Savings Program, where eligible clinicians or groups would receive a certain number of bonus points for the quality performance category for improvement, although the total points received for the performance may not exceed the maximum total points for the performance category in the absence of the quality improvement points. Under this methodology, we would score individual measures and determine the corresponding number of points that may be earned based on the MIPS eligible clinician's performance. We would add the points earned for the individual measures within the quality performance category and divide by the total points available for the performance category to determine the quality performance category score. MIPS eligible clinicians that demonstrate quality improvement on established quality measures from year-to-year would be eligible for up to four bonus points for the quality performance category. Bonus points would be awarded based on a MIPS eligible clinician's net improvement in measures within the quality performance category, which would be calculated by determining the total number of significantly improved measures and subtracting the total number of significantly declined measures. Up to four bonus points would be awarded based on a comparison of the MIPS eligible clinician's net improvement in performance on the measures to the total number of individual measures in the quality performance category. When bonus points are added to points earned for the quality measures in the quality performance category, the total points received for the quality performance category may not exceed the maximum total points for the performance category in the absence of the quality improvement points. For a more detailed description of the Shared Savings Program methodology, we refer readers to § 425.502, as well as CY 2015 PFS final rule with comment (79 FR 67928-67931) for a discussion of how CMS will determine whether the improvement or decline is significant.

Option 3: We could adopt the approach similar to that for assessing improvement for the Medicare Advantage 5-star rating methodology. Under this approach, we would identify an overall “improvement measure score” by comparing the underlying numeric data for measures from the prior year with the data from measures for the performance period. To obtain an “improvement measure score” MIPS eligible clinicians would need to have data for both years in at least half of the required measures for the quality performance category. The numerator for the overall “improvement measure” would be the net improvement, which is a sum of the number of significantly improved measures minus the number of significantly declined measures. The denominator is the number of measures eligible for improvement since to qualify for use in the “improvement measure” calculation, a measure must exist in both years and not have had a significant change in its specification. This “improvement measure” would be included in the quality performance category. We recognize that high performing MIPS eligible clinicians may have less room for improvement and consequently may have lower scores on the overall “improvement measure”. Therefore, under this option we would propose the following rule, which is similar to how the 5-star rating methodology treats highly rated plans in connection with the improvement measure to avoid penalizing consistently high-performing eligible clinicians: We would calculate a MIPS eligible clinician's score with the “improvement measure” and without, and use the MIPS eligible clinician's best score. We request comments on these proposals.

(3) Scoring the Resource Use Performance Category

As we described in section II.E.6.a.1. of this rule, we proposed to align scoring across the MIPS performance categories. For the resource use performance category, we propose to score the resource use measures similarly to the quality performance category. Specifically, we propose at § 414.1380(b)(2) to assign one to ten points to each measure based on a MIPS eligible clinician's performance compared to a benchmark. However, we note that for the resource use performance category (unlike the quality performance category), the benchmark is based on the performance period, rather than the baseline period. The details of the scoring for resource use measures are described below.

(a) Resource Use Measure Benchmarks

For the resource use performance category, we propose at § 414.1380(b)(2) that the performance standard is measure-specific benchmarks. We would calculate an array of measure benchmarks based on performance. Then, a MIPS eligible clinician's actual measure performance during the performance period would be evaluated to determine the number of points that should be assigned based on where the actual measure performance falls within these benchmarks.

We propose at § 414.1380(b)(2) to create benchmarks for the resource use measures based on the performance period. Changes in payment policies, including changes in relative value units, and changes that affect how hospitals, clinicians and other health care providers are paid under Medicare Parts A and B, can make it challenging to compare resource use in a performance period with a historical baseline period. In addition, for HVBP and VM, we use the performance period to establish the benchmarks for scoring HVBP's efficiency measures and VM's cost measures (80 FR 49562, 80 FR 71280). If we use the performance period, we would publish the benchmark methodology in a final rule, but would not be able to publish the actual numerical benchmarks in advance of the performance period. We believe that it is important for MIPS eligible clinicians to know in advance how they might be scored and can track their performance so we would continue to provide performance feedback with information on the MIPS eligible clinician's relative performance.

We considered an alternative to base the resource use performance category measure benchmarks on the baseline period proposed in section II.E.6.a.1.c., rather than the performance period. This option would further align the resource use performance category benchmark methodology with the quality performance category benchmark methodology. This option would also allow us to publish the numerical benchmarks before the performance period ends; however, we believe the benefits of earlier published benchmarks are more limited for resource use measures. MIPS eligible clinicians would not be able to track their daily progress because they would not have all the necessary information to determine the attribution, price standardization, and otherwise adjust the measures. We believe the relative performance that we provide through feedback reports would provide MIPS eligible clinicians the information they need to track performance and to learn about their resource utilization. In addition, we believe that using benchmarks based in the performance period is a better approach than using benchmarks based in the baseline period because different payment policies could apply during the baseline period than during the performance period which could affect a MIPS' eligible clinician's resource use. We would also have to identify the baseline benchmark and trend it forward so that the dollars in the baseline period are comparable to the performance period, whereas we would not have to make a trending adjustment for benchmarks based on the performance period. For these reasons, we elected to propose to base the benchmarks on the performance period rather than the baseline period.

We propose to create a single set of benchmarks for each measure specified for the resource use performance category. All MIPS eligible clinicians that are attributed sufficient cases for the measure would be included in the same benchmark. In addition, we would require a minimum of 20 MIPS eligible clinicians or groups to be attributed the case minimum in order to develop the benchmark. If a measure does not have enough eligible clinicians or groups that are attributed enough cases to create a benchmark, then we would not include that measure in the scoring for the resource use performance category.

We request comment on the proposal to establish resource use measure benchmarks based on the performance period as well as the alternative proposal.

(b) Assigning Points Based on Achievement

For each set of benchmarks, we propose to calculate the decile breaks based on measure performance during the performance period and assign points for a measure based on which benchmark decile range the MIPS eligible clinician's performance on the measure is between. We propose that for resource use measures, lower costs represent better performance. In other words, MIPS eligible clinicians in the top decile would have the lowest resource use. We propose to use a methodology generally consistent with the methodology proposed for the quality performance category. We refer readers to Tables 21 and 22 for details on assigning points based on decile distribution. We request comments on the methodology for assigning points based on performance period deciles for the resource use performance category and solicit comments on alternative methodologies for assigning points for performance under this performance category for future rulemaking.

Table 21 illustrates an example of using decile points along with partial points to assign achievement points for a sample resource use measure.

(c) Case Minimum Requirements

We seek to ensure that MIPS eligible clinicians are measured reliably; therefore, we proposed in section II.E.5.e.3. to establish a 20 case minimum for each resource use measure. We note that this would include the Medicare Spending Per Beneficiary (MSPB) measure. In the CY 2016 PFS final rule, we finalized a policy that increases the required case minimum for MSPB from 20 to 125 cases (80 FR 71295-71296). However, due to the proposed changes to the MSPB measure, discussed in section II.E.5.e.(3)(a)., we believe we can appropriately use a required case minimum of 20 for the revised MSPB measure. Refer to section II.E.5.e.(3) for our rationale for this proposal.

(d) Calculating the Resource Use Performance Category Score

To calculate the resource use performance category score, we propose at § 414.1380(b)(2)(iii) to average all the scores of all the resource use measures attributed to the MIPS eligible clinician. All measures in the resource use performance category as described in section II.E.5.e would be weighted equally. If a MIPS eligible clinician has only one resource use measure with a required case minimum to be scored, we would score that measure accordingly, and the MIPS eligible clinician's resource use performance category score would consist of the score for that one measure. We note that MIPS eligible clinicians cannot receive a zero score for any resource use measure for failure to submit the measure since none of the resource use performance category measures are submitted by MIPS eligible clinicians. Rather, these measures are attributed to MIPS eligible clinicians through claims data. However, if a MIPS eligible clinician is not attributed any resource use measures (for example, because the case minimum requirements have not been met for any measure or there is not a sufficient number of MIPS eligible clinicians to create a benchmark for any measure), then a resource use performance category score would not be calculated. Refer to section II.E.6.b for details on how we propose to address scenarios where a performance category score is not calculated for a MIPS eligible clinician. MIPS eligible clinicians would receive performance feedback as required under section 1848(q)(12) of the Act and discussed in section II.E.8.a of this proposed rule. Over time, performance feedback may include a list of attributed cases for each measure by MIPS eligible clinician. We request comment on our proposals to calculate the resource use performance category score.

Table 22 illustrates a sample scoring methodology for a limited set of measures. A MIPS eligible clinician is attributed resource use measures as described above and receives a score for measures where the eligible clinician has a sufficient number of cases attributed.

The MIPS eligible clinician described in Table 22 did not have the required case minimum for Measure 4 (Episode 2), and therefore is not scored on this measure. Similarly, the MIPS eligible clinician was not attributed any cases for Measure 5 (Episode 3) and was not scored on the measure. Measures that do not meet the required case minimum are not used for scoring.

In the example above, making the assumption that all measures listed have a median performance falling between the fifth and sixth deciles and would provide a score of six points, the MIPS eligible clinician with a value above the median would receive a score lower than six points. For example, Measure 1 has a performance of $15,000 which is higher than the median performance of $13,000, therefore the number of points assigned (4.0) is lower than six points.

Based on the resource use measures available for scoring, the MIPS eligible clinician is scored against the total number of points available. The resource use performance category score for this eligible clinician is (22.3 performance points/40 possible points) = 55.8 percent.

Unlike the quality performance category score, we are not proposing bonus points as part of the resource use performance category score.

(4) Scoring the CPIA Performance Category

Section 1848(q)(5)(C) of the Act outlines specific scoring rules for the CPIA performance category. Section 1848(q)(5)(C)(i) of the Act provides that a MIPS eligible clinician who is in a practice that is certified as a patient-centered medical home or comparable specialty practice with respect to a performance period shall receive the highest potential score for the CPIA performance category for such period. Section 1848(q)(5)(C)(ii) of the Act provides that MIPS eligible clinicians participating in an APM with respect to a performance period shall earn a minimum score of one-half of the highest potential score for the CPIA performance category for such period. We refer readers to section II.E.5.h of this preamble for a description of the APM scoring standard. Section 1848(q)(5)(C)(iii) of the Act states that MIPS eligible clinicians are not required to perform activities in each subcategory or participate in an APM in order to receive the highest possible score for the CPIA performance category. Based on these criteria, we propose a scoring methodology that assigns points for the CPIA performance category (based on patient-centered medical home participation and the CPIAs reported by the MIPS eligible clinician). A MIPS eligible clinician's performance would be evaluated by comparing the reported CPIAs to the highest possible score.

(a) Assigning Points to Reported CPIAs

CPIA is a new performance category that has not been implemented in our previous programs. Therefore, in year 1, we cannot assess how well the MIPS eligible clinician has performed on the activity against data from a baseline year. We can only assess whether the MIPS eligible clinician has participated sufficiently to receive credit in the CPIA performance category. Therefore, we propose at § 414.1380(b)(3) to assign points for each reported activity within two categories: Medium-weighted and high-weighted activities. Medium-weighted activities are worth 10 points. High-weighted activities are worth 20 points. Table 23 lists all of the proposed CPIAs that are high-weighted. All other activities not listed as high-weighted activities would be considered medium activities. Table H in the Appendices provides the CPIA Inventory of all activities, both medium-weighted and high-weighted. Consistent with our unified scoring system principles, MIPS eligible clinicians would know in advance how many potential points they could receive for each CPIA.

Activities are proposed to be weighted as high based on the extent to which they align with activities that support the patient-centered model home, since that is the standard under section 1848(q)(5)(C)(i) of the Act for achieving the highest potential score for the CPIA performance category, as well as with CMS priorities for transforming clinical practice. Additionally, activities that require performance of multiple actions, such as participation in the Transforming Clinical Practice Initiative, participation in a MIPS eligible clinician's state Medicaid program, or an activity identified as a public health priority (such as emphasis on anticoagulation management or utilization of prescription drug monitoring programs) are justifiably weighted as high. We seek comment on which activities should receive a high weight as opposed to a medium weight.

We also considered an approach of equal weighting for all CPIAs. We seek comment on a multi-tier weighting approach such as low, medium and high activity categories for future years of MIPS.

(b) CPIA Performance Category Highest Potential Score

Although there is variability in the level that each MIPS eligible clinician would perform a CPIA, we currently do not have a standard way of measuring that variability. In future years, we plan to capture data to begin to develop a baseline for measuring CPIA improvement. Because we cannot measure variable performance within a CPIA, we propose at § 414.1380(b)(3)(v) to compare the points associated with the reported activities against the highest potential score. We propose the highest potential score to be 60 points for the CY 2017 performance period given the following rationale.

Based on discussions with several high performing organizations, we believe that MIPS eligible clinicians would be able to report on as many as six activities of medium weight. Examples of these organizations include one that led a major redesign of patient workflow after Hurricane Katrina, implementing clinical practice improvements to ensure patients receive faster treatment in the event of future disasters, ranked nationally in 6 adult specialties and high-performing in 6 adult specialties; a second that was recognized by a leading medical association that achieved: 6.7 percent 30-day all cause readmissions, 42 percent fewer ED visits with implementation of a 60-day intensive home care program, costs of 15 percent-28 percent below regional average and significant improvement in patient surveys from CAHPS; and a third recognized as a leader in rural health with the highest award for excellence from the National Rural Primary Care Association.

We also believe that a top performing small practice (consisting of 15 or fewer professionals) or practice in a rural or health professional shortage area, or a non-patient facing MIPS eligible clinician would be able to report on at least two activities. In consideration of special circumstances for these small practices, as well as practices located in rural areas and in Health Professional Shortage Areas (HPSAs) or non-patient facing MIPS eligible clinicians, we propose that the weight for any activity selected would be 30 points. For any MIPS eligible clinician, the maximum total points achievable in this performance category is 60 points. Based on the above rationale, we believe it is reasonable to expect all MIPS eligible clinicians to be able to report CPIAs, and as such, a MIPS eligible clinician reporting no CPIA would receive a zero score for the CPIA performance category. We believe this proposal allows us to capture variation in reporting the CPIA performance category.

(c) Points for Certified Patient-Centered Medical Home or Comparable Specialty Practice

Section 1848(q)(5)(C)(i) of the Act specifies that a MIPS eligible clinician who is in a practice that is certified as a patient-centered medical home or comparable specialty practice, as determined by the Secretary, with respect to a performance period must be given the highest potential score for the CPIA performance category for the performance period. We propose that patient-centered medical home practices are those that have received accreditation from any of the following four nationally recognized accreditation organizations (the Accreditation Association for Ambulatory Health Care, the National Committee for Quality Assurance (NCQA), The Joint Commission, and the Utilization Review Accreditation Commission (URAC)); or are a Medicaid Medical Home Model or Medical Home Model. We propose that CMS's proposed comparable specialty practices are those that include the NCQA Patient-Centered Specialty Recognition. We refer readers to section II.F. of this proposed rule for further description of the Medicaid Medical Home Model or Medical Home Model. The four accreditation organizations listed above all have evidence of being used by a large number of medical organizations as the model for their patient-centered medical home and are national in scope. No other criteria are required for receiving recognition as a certified patient patient-centered medical home or comparable specialty practice except for being recognized by one of the above organizations.

Section II.E.5.f. of this rule outlines the policy for certified patient-centered medical homes. The organizations identified above maintain a list of certified patient-centered medical homes, including the Medicaid Medical Home and Medical Home Models, that would be used to determine whether a MIPS eligible clinician qualifies for the highest potential score for the CPIA performance category because the MIPS eligible clinician is in a certified patient-centered medical home. NCQA maintains a list of practices that have received the Patient-Centered Specialty Recognition which would be used to determine whether a MIPS eligible clinician qualifies for the highest potential score for the CPIA performance category because the MIPS eligible clinician is in a comparable specialty practice.

We propose at § 414.1380(b)(3) that a MIPS eligible clinician who is in a practice that is certified as a patient-centered medical home, including a Medicaid Medical Home or Medical Home Model, or comparable specialty practice in accordance with those proposals would receive the highest potential score (in accordance with section 1848(q)(5)(C)(i) of the Act) of 60 points for the CPIA performance category.

(1) Section II.E.5.f. of this rule presents the CMS Study on CPIA and Measurement. Given the burden for participants completing the year-long study and the value of collectively examining innovation and practice activities to improve clinical quality data submissions and further reduce time requirements for eligible clinicians and groups to report, we propose that MIPS eligible clinicians and groups that successfully participate and submit data to fulfill study requirements would receive the highest potential score of 60 points for the CPIA performance category.

(d) Calculating the CPIA Performance Category Score

To determine the CPIA performance category score, we propose to sum the points for all of the MIPS eligible clinician's reported activities and divide by the proposed CPIA performance category highest potential score of 60. A perfect score would be 60 points divided by 60 possible points, which equals 100 percent. If MIPS eligible clinicians have more than 60 CPIA points, then we propose to cap the resulting CPIA performance category score at 100 percent.

Table 24 illustrates a sample scoring methodology for the CPIA performance category. The MIPS eligible clinician below was not an APM participant and does not immediately earn the minimum score of one-half of the highest potential score or 30 points that are available for APM participation. The MIPS eligible clinician below completed two high-weighted activities worth 20 points each and two medium-weighted activities for 10 points each in order to receive the maximum 60 points available in the performance category for a CPIA performance category score of 100 percent.

Alternatively, the MIPS eligible clinician could have selected three high-weighted activities for 20 points each, six medium-weighted activities for ten points each, or some combination to reach 60 points. The score however is capped at 100 percent (60/60). This means that a MIPS eligible clinician who selects four high-weight activities (80 possible points) would still be given a score of 100 percent (60/60).

Section 1848(q)(2)(B)(iii) of the Act requires the Secretary to give consideration to the circumstances of small practices (consisting of 15 or fewer professionals) and practices located in rural areas and in geographic health professional shortage areas (HPSAs) (as designated under section 332(a)(1)(A) of the Public Health Service Act) in defining activities. Section 1848(q)(2)(C)(iv) of the Act also requires the Secretary to give consideration to non-patient-facing MIPS eligible clinicians. Further, section 1848(q)(F)(5) of the Act allows the Secretary to assign different scoring weights for measures, activities, and performance categories, if there are not sufficient measures and activities applicable and available to each type of eligible clinician.

For MIPS eligible clinicians and groups that are small practices (consisting of 15 or fewer professionals), practices located in rural areas, practices located in geographic HPSAs, or non-patient facing MIPS eligible clinicians or non-patient facing MIPS eligible clinician groups, we propose alternative scoring requirements for the CPIA performance category. The rationale for this alternative scoring is grounded in the resource constraints these MIPS eligible clinicians face which was further discovered during listening sessions with small, rural and geographic HPSAs and medical societies for non-patient facing MIPS eligible clinicians and groups. We believe that while non-patient facing MIPS eligible clinicians and non-patient facing groups could select activities from some sub-categories (such as care coordination and patient safety), for other sub-categories (such as beneficiary engagement and population management) non-patient facing MIPS eligible clinicians and groups will need to consider novel practice activities that are within their scope and can improve beneficiary care. We will continue to work with non-patient facing MIPS eligible clinician professional organizations to further develop activities relevant for these clinicians in future years. Our rationale for small practices and practices located in rural areas and in HPSAs is grounded in the resource constraints that these MIPS eligible clinicians face. This rationale is especially compelling given that each activity requires at least 90 days and may not necessarily be conducted in parallel, with time allocated to pre-planning and post-planning, which would impact the practice's limited resources.

All MIPS eligible clinicians would be allowed to self-identify as part of an APM, a patient-centered medical home or comparable specialty practice, a Medicaid Medical Home or Medical Home Model, a non-patient facing professional, a small practice (consisting of 15 or fewer professionals), a practice located in a rural area, or a practice in a geographic HPSA or any combination thereof as applicable during attestation following the performance period. We refer readers to https://innovation.cms.gov/Medicare-Demonstrations/Medicare-Medical-Home-Demonstration.html for more information on the Medical Home Model.

We would validate these self-identifications as appropriate. We propose that the following scoring would apply to MIPS eligible clinicians who are a non-patient facing professional, a small practice (consisting of 15 or fewer professionals), a practice located in a rural area, or practice in a geographic HPSA or any combination thereof:

• Reporting of one medium-weighted or high-weighted activity would result in 50 percent of the highest potential score.
• Reporting of two medium-weighted or high-weighted activities would result in 100 percent of the highest potential score.

In future years, we may adjust the weighting of activities at the MIPS eligible clinician level based on initial patterns of CPIA reporting. For example, if a MIPS eligible clinician reports on the same medium-weighted activity over several performance periods, in a subsequent year that MIPS eligible clinician may not be allowed to continue to select that same activity. This is because the intent of the CPIA performance category is to demonstrate improvement over time and not just demonstrate same benefit from year to year. For example, continuing to provide expanded practice access does not demonstrate improvement over time. Further, should the weighting of activities change in future years, we may also adjust the CPIA performance category point target accordingly. We request comment on our proposed approach to score the CPIA performance category. We also seek comment on alternative methodologies for the CPIA performance category. We seek to assure equity in scoring MIPS eligible clinicians while still considering activity variation, impact and burden.

(5) Scoring the Advancing Care Information Performance Category

We refer readers to section II.E.5.g.6. for our proposed methodology for scoring the advancing care information performance category. We reiterate that this methodology has many of the features of the unified scoring system described above. Specifically, we are moving away from the “all-or-nothing” scoring approach of the Medicare EHR Incentive Program. In addition, MIPS eligible clinicians would know in advance what they have to do to achieve points under the advancing care information performance category in MIPS. We provide a brief summary of our proposed scoring methodology here.

In the advancing care information performance category, we propose to score for both participation and performance. We refer to these scoring methods as the “base score” and the “performance score”.

To earn points toward the base score, a MIPS eligible clinician or group must report the numerator and denominator (or yes/no statement as applicable) for certain measures adopted by the EHR Incentive Programs in the 2015 EHR Incentive Programs Final Rule to achieve 50 percent of the total advancing care information performance category score. For measures that previously included a percentage-based threshold, we are not requiring MIPS eligible clinicians or groups to meet those thresholds. Instead we propose to require eligible clinicians and groups to report the numerator (of at least one) and denominator (or a yes/no statement for applicable measures) for each measure being reported.

For the base score, MIPS eligible clinicians or groups must meet Objective 1: Protect Patient Health Information and its associated measure in 2015 EHR Incentive Programs Final Rule. Additionally, eligible clinicians would be required to report the numerator and denominator, or a yes/no statement as appropriate, for each measure for Electronic Prescribing, Patient Electronic Access to Health Information, Coordination of Care Through Patient Engagement, Health Information Exchange, and Public Health and Clinical Data Registry Reporting— as adopted in the 2015 EHR Incentive Programs Final Rule. Failure to meet any of the objectives would result in a base score of zero and an advancing care information performance category score of zero.

For the Public Health and Clinical Data Registry Reporting objective, an eligible clinician or group is only required to report on the Immunization Registry Reporting measure. Completing any additional measures under the objective would earn one additional bonus point after calculation of the performance score.

The performance score is then determined in addition to the base score. The performance score methodology would implement a decile scale for the application of additional points based on performance in the objectives and measures for Patient Electronic Access, Coordination of Care through Patient Engagement, and Health Information Exchange. There are eight associated measures under these three objectives; each has a maximum of ten percentage points available. The total available performance score would be 80 percent which is, in combination with the base score of 50 percent, greater than the total possible performance category score of 100 percent. We have taken this approach in order to provide flexibility toward achieving the maximum score in the advancing care information performance category—however, a MIPS eligible clinician or group's score is capped at 100 percent.

This summary only represents the primary advancing care information performance category scoring proposal. For full details on the advancing care information performance category scoring and an explanation of alternatives considered, as well as accommodation for eligible clinicians planning to report Modified Stage 2 or use 2014 Edition CEHRT in 2017 please refer to II.E.5.g.4.

b. Calculating the Composite Performance Score (CPS)

Section II.E.6.a. of this rule describes our proposed methodology for assessing and scoring MIPS eligible clinician performance for each of the four performance categories. In this section, we propose the methodology to determine the CPS based on the scores for each of the four performance categories. We define at § 414.1305 the CPS as a composite assessment (using a scoring scale of 0 to 100) for each MIPS eligible clinician for a specific performance period determined using the methodology for assessing the total performance of each MIPS eligible clinician according to the performance standards with respect to the applicable measures and activities for each applicable performance category. The CPS is the sum of the products of each performance category score and each performance category's assigned weight multiplied by 100.

(1) Formula To Calculate the CPS

Section 1848(q)(5)(A) of the Act requires the Secretary to develop a methodology for assessing the total performance of each MIPS eligible clinician according to the performance standards with respect to the applicable measures and activities with respect to each performance category applicable to such clinician for a performance period, and using the methodology, provide for a CPS (using a scoring scale of 0 to 100) for each MIPS eligible clinician for the performance period. Additionally, sections 1848(q)(5)(E) and (F) of the Act address the weights for each of the performance categories in the CPS.

To create a CPS from 0-100 based on the individual performance category scores, we propose to multiply the score for each performance category by the assigned weight for the performance category. We provide in Table 25 the weights for each performance category for the 2019, 2020 and 2021 MIPS payment years. The resulting weighted performance category scores would be summed to create a single CPS. As described in section II.E.2 of this preamble, we propose that the identifier for MIPS performance would be the same for all four performance categories, and therefore, the methodology to calculate a CPS would be the same for both individual and group performance.

The following equation summarizes the proposed CPS calculation at § 414.1380(c):

CPS = [(quality performance category score × quality performance category weight) + (resource use performance category score × resource use performance category weight) + (CPIA performance category score × CPIA performance category weight) + (advancing care information performance category score × advancing care information performance category weight)] × 100.

(a) Accounting for Risk Factors

Section 1848(q)(1)(G) of the Act requires us to consider risk factors in our scoring methodology. Specifically, that section provides that the Secretary, on an ongoing basis, shall, as the Secretary determines appropriate and based on individuals' health status and other risk factors, assess appropriate adjustments to quality measures, resource use measures and other measures used under MIPS and assess and implement appropriate adjustments to payment adjustments, CPSs, scores for performance categories or scores for measures or activities under the MIPS. In doing this, the Secretary is required to take into account the relevant studies conducted and recommendations made in reports under section 2(d) of the Improving Medicare Post-Acute Transformation (IMPACT) Act of 2014 and, as appropriate, other information, including information collected before completion of such studies and recommendations. HHS' Office of the Assistant Secretary for Planning and Evaluation (ASPE) is conducting studies and making recommendations on the issue of risk adjustment for socioeconomic status on quality measures and resource use as required by section 2(d) of the IMPACT Act and expects to issue a report to Congress by October 2016. We will closely examine the recommendations issued by ASPE and incorporate them as feasible and appropriate through future rulemaking. We also note that several MIPS measures, as appropriate, include risk adjustment in their measure specifications. For example, outcome measures in the quality performance category generally have risk adjustment embedded in the measure calculation specification, while process measures generally do not. Similarly, in the resource use performance category, the proposed total per capita costs for all attributed beneficiaries measure is adjusted for demographic and clinical factors. That measure also has a specialty adjustment that is applied after the measure calculation to account for differences in specialty mix within a practice. The MSPB measure and other resource use measures have different risk adjustments that are specific to the individual measure. For the first year of MIPS, for the quality and resource use performance categories, we propose to use the measure-specific risk adjustment for all measures (where applicable), as well as the additional specialty adjustment for the total per capita costs for all attributed beneficiaries.

We invite public comments on this proposal.

(2) CPS Performance Category Weights

(a) General Weights

Section 1848(q)(5)(E)(i) of the Act specifies weights for the performance categories included in the MIPS CPS: In general, 30 percent for the quality performance category, 30 percent for the resource use performance category, 25 percent for the advancing care information performance category, and 15 percent for the CPIA performance category. However, that section also specifies different weightings for the quality and resource use performance categories for the first and second years for which the MIPS applies to payments. Section 1848(q)(5)(E)(i)(II)(bb) of the Act specifies that for year 1, not more than 10 percent of the CPS will be based on the resource use performance category and for year 2, not more than 15 percent will be based on resource use performance category. Under section 1848(q)(5)(E)(i)(I)(bb) of the Act, the weight of the quality performance category for each of the first two years will increase by the difference of 30 percent minus the weight specified for the resource use performance category for the year.

In previous sections of this rule, we have proposed the performance category weights for the first MIPS payment year of 2019. In section II.E.5.e.2., we propose to set the resource use performance category weight at 10 percent for the 2019 payment year and 15 percent for the 2020 payment year. Correspondingly, in section II.E.5.b.2., we propose to set the quality performance category weight to 50 percent for the 2019 payment year and 45 percent for the 2020 payment. The quality performance category weight proposal is based on the 30 percent required by statute for the quality performance category plus 30 percent minus the weight of the resource use performance category, as required by section 1848(q)(5)(E)(i)(I)(bb) of the Act. As specified in section 1848(q)(5)(E)(i) of the Act, the weights for the other performance categories are 25 percent for the advancing care information performance category; and 15 percent for the CPIA performance category. Section 1848(q)(5)(E)(ii) of the Act provides that in any year in which the Secretary estimates that the proportion of eligible professionals (as defined in section 1848(o)(5) of the Act) who are meaningful EHR users (as determined under in section 1848(o)(2) of the Act) is 75 percent or greater, the Secretary may reduce the applicable percentage weight of the advancing care information performance category in the CPS, but not below 15 percent, and adjust the weighting of the other performance categories. We refer readers to our proposals concerning section 1848(q)(5)(E)(ii) of the Act in section II.E.5.g.(6)(e).

Table 25 summarizes the weights specified for each performance category under section 1848(q)(5)(E)(i) of the Act and in accordance with our proposals.

(b) Flexibility for Weighting Performance Categories

Under section 1848(q)(5)(F) of the Act, if there are not sufficient measures and activities applicable and available to each type of eligible clinician involved, the Secretary shall assign different scoring weights (including a weight of zero) for each performance category based on the extent to which the category is applicable and for each measure and activity based on the extent to which the measure or activity is applicable and available to the type of eligible clinician involved.

In section II.E.6.a and section II.E.5.g.8., we describe scenarios where certain MIPS eligible clinicians might not receive a performance category score in the quality, resource use, or advancing care information performance categories. We propose that in such scenarios we would use the authority under section 1848(q)(5)(F) of the Act to assign a weight of zero to the performance category and redistribute the weight for that performance category or categories as described in the next section.

For the quality and resource use performance categories, we believe having sufficient measures applicable and available means that we are able to reliably calculate a score for the measures that adequately captures and reflects the performance of the MIPS eligible clinician. For the quality and resource use performance categories, we propose in sections II.E.6.a.2.d., II.E.6.3.a., and II.E.6.a.3.d. that we would not calculate a performance category score if a MIPS eligible clinician does not have any measures with the required case minimum or any measures with a sufficient number of MIPS eligible clinicians to create a benchmark. Measures that do not meet the required case minimum or a sufficient number of MIPS eligible clinicians to create a benchmark would be excluded from scoring, and the MIPS eligible clinician would not receive a quality or resource use performance category score. (Note, this situation is different from a MIPS eligible clinician who elects not to submit any quality measures. A MIPS eligible clinician who elects not to submit any quality measures would receive a quality performance category score of zero.) We believe MIPS eligible clinicians who would have no scored measures for a performance category under our proposals would not have sufficient measures applicable and available for that performance category.

For the quality performance category, we anticipate that most MIPS eligible clinicians would select the measures most relevant to their practice and that in most cases, the measures they select would meet the required case minimum. We plan to monitor measure selection trends under the performance category and will revise this policy if it appears MIPS eligible clinicians are reporting measures that are not relevant to their practice or measures that do not meet the required case minimum. In the resource use performance category, we believe MIPS eligible clinicians who are not attributed enough cases to be reliably measured should not be scored for the performance category. We have proposed to include many resource use measures that we believe are sufficiently developed and ready for evaluating resource use by MIPS eligible clinicians; however, if a MIPS eligible clinician is not attributed any (or very few) cases for the measure, then we do not believe the MIPS eligible clinician should be measured on performance.

We refer readers to section II.E.5.g.8. of this proposed rule for a detailed discussion of the scenarios in which a MIPS eligible clinician may not have sufficient measures applicable and available under the advancing care information performance category. For the CPIA performance category, however, we envision that all MIPS eligible clinicians would have sufficient activities applicable and available and do not propose any scenario where a MIPS eligible clinician would not receive a CPIA performance category score.

In addition to scenarios where a MIPS eligible clinician would have no scored measures for a performance category, we believe there may be scenarios in which a MIPS eligible clinician would have too few scored measures under the quality performance category for us to reliably calculate a performance category score that is worth half the weight of the CPS for the 2019 MIPS payment year. We propose that if a MIPS eligible clinician has fewer than three scored quality measures (either submitted measures or measures calculated from administrative claims data) for a performance period, we would consider the MIPS eligible clinician not to have a sufficient number of measures applicable and available for the 2019 MIPS payment year quality performance category weight and would therefore lower the weight of the quality performance category. In this situation, the MIPS eligible clinician has a quality performance category score, but has data for only one or two scored measures, which is not a sufficient number of measures for the quality performance category because the quality performance category would constitute half of the CPS for the 2019 MIPS payment year. In addition, as described in the next section, for MIPS eligible clinicians that are not scored on the resource use or advancing care information performance category, we propose to increase the weight of the quality performance category. For these reasons, we believe that for the first year of MIPS, the quality performance category requires a sufficient number of measures to justify its weight in the CPS. We will reconsider this policy in future years as the weights for the performance categories change. We may consider implementing a similar policy for the resource use performance category for future years, but not for the first year of MIPS based upon the lower weighting of the resource use performance category.

In section II.E.5.b., we are proposing for the quality performance category, generally, that MIPS eligible clinicians submit a minimum of six measures for scoring in MIPS. In addition, we propose to include up to three population-based measures derived from claims data. As described in section II.E.6.a.2., a MIPS eligible clinician may submit a measure that is not scored, either because the measure did not meet the required case minimum to be reliably measured or because fewer than 20 MIPS eligible clinicians with sufficient volume submitted a measure through a similar reporting mechanism and a benchmark could not be created for the performance or baseline period. We reiterate that a measure that is not scored due to not meeting the required case minimum or lack of a measure benchmark, is different than a required measure that is not reported. Any required measure that is not reported or reported with in a way that does not meet the data completeness requirements would receive a score of zero points and would be considered a scored measure.

We are concerned that if a large percentage of the expected measures are not able to be scored due to not meeting the required case minimums or a missing benchmark, then just one or two measures would contribute disproportionately to the CPS because the quality performance category score is worth 30 to 50 percent (depending on the year) of the CPS under section 1848(q)(5)(E)(i) of the Act. We do not believe a score for one or two quality measures can capture all the elements of quality performance during a performance period. We believe the lack of a sufficient number of measures for scoring limits the value of quality performance measurement toward the CPS. Therefore, we propose that if a MIPS eligible clinician has only two scored measures (including both submitted measures and measures derived from administrative claims data) to reduce the weight of the quality performance category by one-fifth (for example, from 50 percent to 40 percent in year 1) and redistribute the weight (for example, 10 percent in year 1) proportionately to the other performance categories for which the MIPS eligible clinician did receive a performance category score. If a MIPS eligible clinician has only one scored quality measure, then we propose to reduce the weight of the quality performance category by two-fifths (for example, from 50 percent to 30 percent in year 1) and redistribute the weight (for example, 20 percent in year 1) proportionately to the other performance categories for which the MIPS eligible clinician did receive a performance category score. Lowering the weight of the quality performance category would be consistent with the relatively low percentage of expected quality measures that are able to be scored.

We request comment on these proposals to identify MIPS eligible clinicians without sufficient measures and activities applicable and available and our proposals to reweight those performance categories. We also seek comment on alternative methods for reweighting performance categories for MIPS eligible clinicians without sufficient measures and activities in certain performance categories. We seek to ensure that reweighting would not cause an eligible clinician to be either advantaged or disadvantaged due to a lack of sufficient measures and activities applicable and available, and a corresponding inability to generate a score for a certain performance category.

(c) Redistributing Performance Category Weights

We propose at § 414.1380(c)(3) to reweight the performance categories for MIPS eligible clinicians when there are not sufficient measures and activities applicable and available to them. We propose to reweight the performance categories in the following situations.

If the MIPS eligible clinician does not receive a resource use or advancing care information performance category score, and has at least three scored measures (either submitted measures or those calculated from administrative claims) in the quality performance category, then we propose to reassign the weights of the performance categories without a score to the quality performance category. We believe this policy is appropriate for several reasons. First, section 1848(q)(5)(E)(i)(I)(bb) of the Act redistributes weight from the resource use performance category to the quality performance category in the first two years of MIPS. This proposal is consistent with that redistribution logic. In addition, MIPS eligible clinicians have experience reporting quality measures through the PQRS program and measurement in this performance category is more mature. Finally, for the 2019 MIPS payment year, quality performance would be worth at least half of the CPS. By requiring the MIPS eligible clinician to have at least three scored quality measures, we believe the quality performance category would be robust enough to support more weight reassigned to it than other performance categories. We may revisit this policy in future years as the weight for the resource use performance category increases and the weight for the quality performance category decreases.

We also propose an alternative that does not reassign all the weight to the quality performance category, but rather reassigns the weight proportionately to each of the other performance categories for which the MIPS eligible clinician has received a performance category score.

We request public comments on the proposal to reassign the weights to the quality performance category, as well as the alternate proposal to redistribute proportionately to other performance categories.

If the MIPS eligible clinicians have fewer than three scored measures in the quality performance category score, then we propose to reassign the weights for the performance categories without scores proportionately to the other performance categories for which the MIPS eligible clinician has received a performance category score. We request comment on this proposal.

Finally, because the CPS is a composite score, we believe the intention of section 1848(q)(5) of the Act is for MIPS eligible clinicians to be scored based on multiple performance categories. Basing a CPS on a single performance category, even a robust and familiar performance category like quality, would frustrate that intent. In our proposals, CPIA is the only performance category which would always have a performance category score. We are particularly concerned about the possibility that a MIPS eligible clinician might, for the reasons discussed above, not have sufficient measures applicable and available for the quality, resource use, and advancing care information performance categories, and would only receive a score for the CPIA performance category. The CPIA performance category is based on activities that are reported by attestation, not on measured performance. In addition, because CPIA is not as mature as the other performance categories, each of which include certain aspects of existing CMS programs, we are unsure how much variation we will have in the CPIA performance category. We do not think it would be equitable to allow MIPS eligible clinicians that attest to receive the maximum points for that performance category and then base the CPS solely on the CPIA performance category. Such a scenario may result in higher CPS and payment adjustment factors for some MIPS eligible clinicians based solely on the CPIA performance category, while other MIPS eligible clinicians are measured based on their performance under the other performance categories. Therefore, we propose that if a MIPS eligible clinician receives a score for only one performance category, we would assign the MIPS eligible clinician a CPS that is equal to the performance threshold described in section II.E.5., which means the eligible clinician would receive a MIPS adjustment factor of 0 percent for the year. We anticipate this proposal would affect very few MIPS eligible clinicians in year 1 and even fewer in future years as more eligible clinicians are able to report on and receive scores for more of the performance categories.

We welcome public comment on this proposal.

7. MIPS Payment Adjustments

a. Payment Adjustment Identifier and CPS Used in Payment Adjustment Calculation

i. Payment Adjustment Identifier

As we describe in section II.E.2 of this preamble, we propose to allow MIPS eligible clinicians to measure performance as an individual, as a group defined by TIN, or as an APM Entity group using the APM scoring standard, yet for purposes of the application of the MIPS adjustment factors to payments in accordance with section 1848(q)(6)(E) of the Act (referred to as the payment adjustment), we are proposing to use a single identifier, TIN/NPI, for all MIPS eligible clinicians, regardless of whether the TIN/NPI was measured as an individual, group or APM Entity group. In other words, a TIN/NPI may receive a CPS based on individual, group, or APM Entity group performance, but the payment adjustment would be applied at the TIN/NPI level.

We are proposing to use the single identifier, TIN/NPI, for the payment adjustment for a few reasons. First, the final eligibility status of some clinicians would not be known until after the performance period ends. For example, the calculations to determine which clinicians would be excluded from MIPS, such as identifying clinicians that are QPs or are below the low-volume threshold, occur after the performance period ends. Using TIN/NPI would allow us to correctly identify which TIN/NPIs are still MIPS eligible clinicians after the exclusion criteria have been applied.

Second, the identifiers for measurement are not mutually exclusive and using TIN/NPI to apply the payment adjustment would allow us to resolve any inconsistencies that arise from the measurement identifiers. For example, a TIN may have 40 percent of its eligible clinicians participating in a MIPS APM and the remaining 60 percent are not participating in any APM. The TIN elects to submit performance information for all the eligible clinicians in the TIN, including those that are participating in the MIPS APM, so that it can ensure all of its eligible clinicians are being measured in MIPS. We cannot simply use the APM Entity and TIN identifiers because we either have eligible clinicians with duplicative data and overlapping scores, or we have portions of the measurement identifier carved out if we eliminate the overlap. In our example, the eligible clinicians participating in the MIPS APM would have data for two CPSs (one based on the APM Entity group performance and one based on the group TIN performance). The eligible clinicians not participating in the MIPS APM would have only one CPS (one based on the group TIN performance). Applying the payment adjustment at the TIN/NPI level provides us the flexibility to correctly identify and resolve the conflicts emerging when measurement identifiers overlap. The TIN/NPI identifier is mutually exclusive on all of our measurement identifier options; therefore, we believe this identifier can be consistently used for individual, group, or APM scoring standard identifiers. We refer readers to section II.E.2 for a discussion of identifiers and our proposals related to them.

ii. CPS Used in Payment Adjustment Calculation

Because we are proposing to use only TIN/NPI to apply the MIPS payment adjustments and because there is a gap between the performance period and the MIPS payment year, we believe we should assign the historical CPS to each TIN/NPI that is subject to MIPS for the payment year.

In general, we propose to use the CPS associated with the TIN/NPI combination in the performance period. For groups submitting data using the TIN identifier, we propose to apply the group CPS to all the TIN/NPI combinations that bill under that TIN during the performance period. For individual MIPS eligible clinicians submitting data using TIN/NPI, we propose to use the CPS associated with the TIN/NPI that is used during the performance period. For eligible clinicians in MIPS APMs, we propose to assign the APM Entity group's CPS to all the APM Entity Participant Identifiers that are associated with the APM Entity on December 31 of the performance period. We refer readers to section II.E.5.h for more information about the process to identify participating APM Entities. For eligible clinicians that participate in APMs for which the APM scoring standard does not apply, we propose to assign a CPS using either the individual or group data submission assignments described above.

In the case where a MIPS eligible clinician starts working in a new practice or otherwise establishes a new TIN that did not exist during the performance period, there would be no corresponding historical performance information or CPS for the new TIN/NPI. Because we want to connect actual performance to the individual MIPS eligible clinician as often as possible, in cases where there is no CPS associated with a TIN/NPI from the performance period, we propose to use the NPI's performance for the TIN(s) the NPI was billing under during the performance period. If the MIPS eligible clinician has only one CPS associated with the NPI from the performance period, then we propose to use that CPS. For example, if a MIPS eligible clinician worked in one practice (TIN A) in the performance period, but is working at a new practice (TIN B) during the payment year, then we would use the CPS for the old practice (TIN A/NPI) to apply the MIPS payment adjustment for the NPI in the new practice (TIN B/NPI). This proposal most closely links the MIPS eligible clinician's performance during the performance period to the payment adjustment. It also ensures that MIPS eligible clinicians who qualify for a positive payment adjustment are able to keep it, even if they change practices. For those who have a negative payment adjustment, this proposal also ensures MIPS eligible clinicians are still accountable for their performance.

In scenarios where the MIPS eligible clinician billed under more than one TIN during the performance period, and the MIPS eligible clinician starts working in a new practice or otherwise establishes a new TIN that did not exist during the performance period, we propose to use a weighted average CPS based on total allowed charges associated with the NPI from the performance period. This proposal would provide a CPS that is based on all the services the NPI billed to Medicare during the performance period. Table 26 presents an example of how this proposed approach would work. In this example, a MIPS eligible clinician (NPI) was assigned a CPS for two unique TIN/NPI combinations from the performance period (TIN A/NPI and TIN B/NPI). In the MIPS payment year, the eligible clinician is now billing for Medicare services under a third TIN/NPI combination without a previously calculated CPS (TIN C/NPI). In this case, the eligible clinician's MIPS adjustment for payments made to TIN C/NPI would be based on a weighted average of CPSs for TIN A/NPI and TIN B/NPI.

If an NPI did not have any allowed charges in the performance period, then the clinician would not be included in MIPS due to the low-volume exclusion.

We also propose an alternative proposal where in lieu of taking the weighted average, we take the highest CPS from the performance period, which would be a CPS of 67.5 in the above example which is the CPS for TIN A/NPI. We believe the alternative approach rewards eligible clinicians for their prior performance and may be easier to implement in year 1 of MIPS. Our concern with this approach is that the highest CPS may represent a relatively small portion of the eligible clinician's practice during the performance period.

We request comment on the proposal to use the CPSs associated with the TIN(s) the NPI was billing under during the performance period when the TIN/NPI does not have a CPS from the performance period. We also request comment on our proposal to use a weighted average, and the alternative proposal to select the highest CPS from the performance period.

We also considered, but are not proposing, a policy to have the performance follow the group (TIN) rather than the individual (NPI). In other words, the MIPS eligible clinician's performance would be based on the historical performance of the new TIN that the MIPS eligible clinician moved to after the performance period, even though the MIPS eligible clinician was not part of this group during the performance period. This policy is consistent with the policy for the VM and would create incentives for MIPS eligible clinicians to move to higher performing practices (77 FR 69308). We also believe this policy would provide a lower burden for practice administrators as all MIPS eligible clinicians in the TIN would have the same payment adjustment. On the other hand, having performance follow the TIN creates some challenges. We are concerned that MIPS eligible clinicians who earned a positive adjustment based on their performance during the performance period would not retain the positive adjustment if the new TIN had a lower CPS. Finally, we believe that having performance follow the TIN could create some unanticipated issues with budget neutrality if high-performing TINs expand. For all of these reasons, we are not proposing to have performance follow the TIN, but rather have performance follow the NPI; however, we seek comment on this option.

In some cases, a TIN/NPI could have more than one CPS associated with it from the performance period, if the eligible clinician submitted duplicative data sets. In this situation, the MIPS eligible clinician has not changed practices, rather for example, a MIPS eligible clinician has a CPS for an APM Entity and a CPS for a group TIN. If a MIPS eligible clinician has multiple CPSs, we propose a multi-pronged approach to select the CPS that would be used to determine the MIPS payment adjustment. First, we propose that if a MIPS eligible clinician is a participant in MIPS APM, then the APM Entity CPS would be used instead of any other CPS (such as a group TIN CPS or individual CPS). We propose that if a MIPS eligible clinician has more than one APM Entity CPS for the same TIN (by participating in multiple MIPS APMs), we would apply the highest APM Entity CPS to the eligible clinician. Second, if a MIPS eligible clinician reports as a group and as an individual, we would calculate a CPS for the group and individual identifier and use the highest CPS for the TIN/NPI. We request comment on this proposed approach.

b. MIPS Adjustment Factors

Section 1848(q)(6)(A) of the Act requires the Secretary to specify a MIPS adjustment factor for each MIPS eligible clinician for a year determined by comparing the CPS of the MIPS eligible clinician for such year to the performance threshold established under paragraph (D)(i) for such year, in a manner such that the adjustment factors specified for a year result in differential payments. Section 1848(q)(6)(A)(iii) of the Act provides that MIPS eligible clinicians with CPS at or above the performance threshold receive a zero or positive adjustment factor on a linear sliding scale such that an adjustment factor of 0 percent is assigned for a CPS at the performance threshold and an adjustment factor of the applicable percent is assigned for a CPS of 100. Section 1848(q)(6)(A)(iv) of the Act provides that MIPS eligible clinicians with CPS below the performance threshold receive a negative payment adjustment factor on a linear sliding scale such that an adjustment factor of 0 percent is assigned for a CPS at the performance threshold and an adjustment factor of the negative of the applicable percent is assigned for a CPS of 0; further, MIPS eligible clinicians with CPS that are equal to or greater than zero, but not greater than one-fourth of the performance threshold, receive a negative payment adjustment factor that is equal to the negative of the applicable percent.

Section 1848(q)(6)(B) of the Act defines the applicable percent for each year as follows: (i) For 2019, 4 percent; (ii) for 2020, 5 percent; (iii) for 2021, 7 percent; and (iv) for 2022 and subsequent years, 9 percent.

Section 1848(q)(6)(C) of the Act provides for an additional positive MIPS adjustment factor for exceptional performance, for each of the years 2019 through 2024, for each MIPS eligible clinician with a CPS for a year at or above the additional performance threshold under paragraph (D)(ii) for such year. The additional MIPS adjustment factor shall be in the form of a percent and determined in a manner such that eligible clinicians having higher CPS above the additional performance threshold receive higher additional MIPS adjustment factors.

c. Determining the Performance Thresholds

(1) Establishing the Performance Threshold

Under section 1848(q)(6)(D)(i) of the Act, for each year of the MIPS, the Secretary shall compute a performance threshold with respect to which the CPS of MIPS eligible clinicians are compared for purposes of determining the MIPS adjustment factors under section 1848(q)(6)(A) of the Act for a year. The performance threshold for a year must be either the mean or median (as selected by the Secretary, which may be reassessed every three years) of the CPS for all MIPS eligible clinicians for a prior period specified by the Secretary. Section 1848(q)(6)(D)(iii) of the Act outlines a special rule for the initial two years of MIPS, which requires the Secretary, prior to the performance period for such years, to establish a performance threshold for purposes of determining the MIPS adjustment factors under paragraph (A) and an additional performance threshold for purposes of determining the additional MIPS adjustment factors under paragraph (C), each of which shall be based on a period prior to the performance periods and take into account data available with respect to performance on measures and activities that may be used under the performance categories and other factors determined appropriate by the Secretary.

We define the term performance threshold at § 414.1305, as the level of performance that is established for a performance period at the CPS level. CPSs above the performance threshold receive a positive MIPS adjustment factor and CPSs below the performance threshold receive a negative MIPS adjustment factor. CPSs that are equal to or greater than 0, but not greater than one-fourth of the performance threshold receive the maximum negative MIPS adjustment factor for the MIPS payment year. CPSs at the performance threshold receive a neutral MIPS adjustment factor.

To establish the performance threshold for the 2019 MIPS payment year, we propose to model 2014 and 2015 Part B allowed charges, 2014 and 2015 PQRS data submissions, 2014 and 2015 QRUR and sQRUR feedback data, and 2014 and 2015 Medicare and Medicaid EHR Incentive Program data to inform where the performance threshold should be. We would use this data to estimate the impact of the quality and resource use scoring proposals. We would also use the EHR Incentive Program information to estimate which MIPS eligible clinicians are likely to receive points for the advancing care information performance category. Because of the lack of historical data for the CPIA performance category, we would apply some sensitivity analyses to help inform where the performance threshold should be.

For the 2019 MIPS payment year, we propose to set the performance threshold at a level where approximately half of the eligible clinicians would be below the performance threshold and half would be above the performance threshold, which we believe is consistent with the intent of section 1848(q)(6)(D)(i) of the Act which requires the performance threshold in year 3 and beyond to be equal to the mean or median of CPS from a prior period. We also considered other policy options when setting the performance threshold. For example, we considered setting the performance threshold so that the scaling factor (which is described in section II.E.7.b) is 1.0. We could set the performance threshold based on policy goals to ensure a minimum number of points are earned before an eligible clinician is able to receive a positive adjustment factor and potentially an additional adjustment factor for exceptional performance. We seek comment on the policy options for setting the performance threshold.

We would determine the performance threshold in accordance with the methodology established in the final rule. We intend to publish the performance threshold on the CMS Web site prior to the performance period.

(2) Additional Performance Threshold for Exceptional Performance

In addition to the performance threshold, section 1848(q)(6)(D)(ii) of the Act requires the Secretary to compute, for each year of the MIPS, an additional performance threshold for purposes of determining the additional positive MIPS adjustment factors for exceptional performance under paragraph (C). For each such year, the Secretary shall apply either of the following methods for computing the additional performance threshold: (1) The threshold shall be the score that is equal to the 25th percentile of the range of possible CPS above the performance threshold determined under section 1848(q)(6)(D)(i) of the Act; or (2) the threshold shall be the score that is equal to the 25th percentile of the actual CPS for MIPS eligible clinicians with CPS at or above the performance threshold with respect to the prior period described in section 1848(q)(6)(D)(i) of the Act.

We define at § 414.1305 the additional performance threshold as an additional level of performance, in addition to the performance threshold, for a performance period at the CPS level at or above which a MIPS eligible clinician may receive an additional positive MIPS adjustment factor. For each year of the MIPS, we will compute an additional performance threshold for purposes of determining the additional MIPS adjustment factors under section 1848(q)(6)(C) of the Act. We propose at § 414.1405(e) the following methods for computing the additional performance threshold: the threshold shall be equal to the 25th percentile of the range of possible CPS above the performance threshold; or it shall be equal to the 25th percentile of the actual CPS for MIPS eligible clinicians with CPS at or above the performance threshold with respect to the prior period used to determine the performance threshold.

As discussed above, section 1848(q)(6)(D)(iii) of the Act outlines a special rule for establishing the additional performance threshold for the initial two years of MIPS. Because 2019 is the first MIPS payment year, we do not have any actual CPS for MIPS eligible clinicians to use for purposes of defining an additional performance threshold under the methodology proposed above. Therefore, we propose to establish the additional performance threshold at the 25th percentile of the range of possible CPS above the performance threshold. For example, if the performance threshold is 60, then the range of possible CPS above the performance threshold would be 61-100. The 25th percentile of those possible values is 70. We intend to publish the exceptional performance threshold with the performance threshold prior to the performance period.

d. Scaling/Budget Neutrality

Section 1848(q)(6)(F)(i) of the Act provides, with respect to positive MIPS adjustment factors for eligible clinicians whose CPS is above the performance threshold under paragraph (D)(i) for such year, the Secretary shall increase or decrease such adjustment factors by a scaling factor (not to exceed 3.0) in order to ensure that the budget neutrality requirement of clause (ii) is met. Stated generally, budget neutrality as required by section 1848(q)(6)(F)(ii) of the Act means the estimated increase in the aggregate allowed charges resulting from the application of positive MIPS adjustment factors under paragraph (A) (after application of the scaling factor) is equal to the estimated decrease in the aggregate allowed charges resulting from the application of negative MIPS adjustment factors under paragraph (A). Under section 1848(q)(6)(F)(iii) of the Act, budget neutrality requirements shall not apply if all MIPS eligible clinicians receive CPS for a year that are below the performance threshold under paragraph (D)(i) for such year, or if the maximum scaling factor (3.0) is applied for a year.

e. Additional Adjustment Factors

Section 1848(q)(6)(C) of the Act requires, for each of the years 2019 through 2024, the Secretary to specify an additional positive MIPS adjustment factor for each MIPS eligible clinician whose CPS for a year is at or above the additional performance threshold established under paragraph (D)(ii) for that year. This additional adjustment factor is required to take the form of a percentage and to be determined by the Secretary such that MIPS eligible clinicians with higher CPS above the additional performance threshold receive higher additional MIPS adjustment factors. Section 1848(q)(6)(F)(iv)(I) of the Act provides, in specifying the additional adjustment factors under paragraph (C) for each applicable MIPS eligible clinician for a year, the Secretary shall ensure that the estimated aggregate increase in payments under Part B resulting from the application of such additional adjustment factors shall be equal to $500,000,000 for each year beginning with 2019 and ending with 2024. We refer to the $500,000,000 increase in payments as aggregate incentive payments. Section 1848(q)(6)(F)(iv)(II) of the Act provides that the additional adjustment factor for each applicable MIPS eligible clinician shall not exceed 10 percent, which may result in an aggregate increase in payments that is less than $500,000,000 as described in subclause (I).

To be consistent with the MIPS adjustment factors under section 1848(q)(6)(A) of the Act, we propose to apply a linear sliding scale where MIPS eligible clinicians with a CPS at the additional performance threshold would receive 0.5 percent additional adjustment factor and MIPS eligible clinicians with a CPS equal to 100 would receive a 10 percent maximum additional adjustment factor. Similar to the adjustment factor, we would apply a scaling factor that is greater than 0 and less than or equal to 1.0 if needed to ensure distribution of the $500,000,000 increase in payments. The scaling factor must be greater than 0 to ensure that MIPS eligible clinicians with higher CPS receive a higher additional adjustment factor. The scaling factor cannot exceed 1.0; the 10 percent maximum additional adjustment factor could only decrease and not increase because section 1848(q)(6)(F)(iv)(II) of the Act provides that the additional adjustment factor shall not exceed 10 percent. We are proposing the starting point for the additional adjustment factor at 0.5 percent for a CPS at the additional performance threshold because this would provide a large enough incentive for MIPS eligible clinicians to strive for the additional performance threshold, while still providing the opportunity for a positive slope on the linear sliding scale. If we are unable to achieve a linear sliding scale starting at 0.5 percent (because the estimated aggregate increase in payments for a year would exceed $500 million), then we propose to lower the starting percentage for a CPS at the additional performance threshold until we are able to create the linear sliding scale with a scaling factor greater than 0 and less than or equal to 1.0. A MIPS eligible clinician with a CPS that is below the additional performance threshold would not be eligible for an additional adjustment factor. We request comments on these proposals.

f. Application of the MIPS Adjustment Factors

Section 1848(q)(6)(E) of the Act provides that for items and services furnished by a MIPS eligible clinician during a year (beginning with 2019), the amount otherwise paid under Part B with respect to such items and services and MIPS eligible clinician for such year, shall be multiplied by 1 plus the sum of the MIPS adjustment factor determined under paragraph (A) divided by 100, and as applicable, the additional MIPS adjustment factor determined under paragraph (C) divided by 100. We would apply the adjustment factors in accordance with section 1848(q)(6)(E) of the Act.

We request comment on our proposals.

g. Example of Adjustment Factors

Figure A provides an example of how various CPS would be converted to an adjustment factor and potentially an additional adjustment factor, using the statutory formula. In this example, the performance threshold is 60. The applicable percentage is 4 percent for 2019. The adjustment factor is determined on a linear sliding scale from zero to 100, with zero being the lowest negative applicable percentage (negative 4 percent for 2019), and 100 being the highest positive applicable percentage. However, there are two modifications to this linear sliding scale. First, there is an exception for a CPS between 0 and 1/4 of the performance threshold (0-15 in our example). All MIPS eligible clinicians with a CPS in this range would receive the lowest negative applicable percentage (negative 4 percent for 2019). Second, the linear sliding scale line for the positive adjustment factor is adjusted by the scaling factor (which is determined by the formula described in section II.E.7.c.) If the scaling factor is greater than 0 and less than or equal to 1.0, then the adjustment factor for a CPS of 100 would be less than or equal to 4 percent. If the scaling factor is above 1.0, but less than or equal to 3.0, then the adjustment factor for a CPS of 100 would be higher than 4 percent. Only those MIPS eligible clinicians with a CPS equal to 60 (which is the performance threshold in this example) would receive no adjustment. In Figure A, the scaling factor for the adjustment factor is 1.37. MIPS eligible clinicians with a CPS equal to 100 would have an adjustment of 5.5 percent (4.0 percent × 1.37).

For the performance threshold of 60, the additional performance threshold for exceptional performance is 70. A CPS of 70 would have an additional adjustment factor of 0.5 percent, and the amount of the additional adjustment factor would increase to 10 percent times a scaling factor that is greater than 0 and less than or equal to 1.0. In Figure A, the scaling factor for the additional adjustment factor is 0.32. Therefore, MIPS eligible clinicians with a CPS of 100 would have an additional adjustment of 3.2 percent (10 percent × 0.32). The total adjustment for a MIPS eligible clinician with a CPS equal to 100 would be 1 + 0.055 + 0.032 = 1.087, for a total positive adjustment of 8.7 percent.

Note:

The adjustment factor for CPS values above the performance threshold is illustrative. For MIPS eligible clinicians with a CPS of 100, the adjustment factor would be 4 percent times a scaling factor greater than 0 and less than or equal to 3.0. The scaling factor is intended to ensure budget neutrality, but cannot be higher than 3.0. The additional adjustment factor is also illustrative. The additional adjustment factor starts at 0.5 percent and cannot exceed 10 percent.

The final MIPS payment adjustments would be determined by the distribution of CPS across MIPS eligible clinicians and the performance threshold. More MIPS eligible clinicians above the performance threshold means the scaling factors would decrease because more MIPS eligible clinicians receive a positive adjustment. More MIPS eligible clinicians below the performance threshold means the scaling factors would increase because more MIPS eligible clinicians would have negative adjustments and relatively fewer MIPS eligible clinicians receive positive adjustments.

We request comment on our proposals.

8. Review and Correction of MIPS Composite Performance Score

a. Feedback and Information To Improve Performance

Through the MIPS and APMs RFI, we solicited comment on various questions related to performance feedback under section 1848(q)(12) of the Act, such as what type of information should be contained in the performance feedback data, how often the feedback should be made available, and who should be able to access the data. Several commenters stated that it would be beneficial if the performance feedback under MIPS contained all the data that contributes to an EP's CPS and any MIPS adjustment. Further, several commenters suggested that performance feedback allow for interactive use of the data. Commenters supported frequent availability of such data and many noted that a minimum of quarterly feedback data would be preferred. Commenters also noted that access to PQRS Feedback Reports currently was a challenge and some suggested that the EPs should be able to control who can access the feedback reports.

(1) Performance Feedback

(a) MIPS Eligible Clinicians

Under section 1848(q)(12)(A)(i) of the Act, as added by section 101(c)(1) of the MACRA, we are at a minimum required to provide MIPS eligible clinicians with timely (such as quarterly) confidential feedback on their performance under the quality and resource use performance categories beginning July 1, 2017, and we have discretion to provide such feedback regarding the CPIA and advancing care information performance categories.

Beginning July 1, 2017, we propose to include information on the quality and resource use performance categories in the performance feedback. Within these performance categories, we propose to use fields similar (that is, quality and resource use) to those currently available in the Quality and Resource Use Reports (QRURs). Since the QRURs already provide information on quality and resource use we believe this is a good starting point for the data fields to be included in the performance feedback. Additional information on the current QRURs can be found at https://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/PhysicianFeedbackProgram/Obtain-2013-QRUR.html .

The first performance feedback is due on July 1, 2017. As this is prior to us having received any MIPS data, we propose to initially provide feedback to MIPS eligible clinicians who are participating in MIPS using historical data set(s), as available and applicable. For example, these historical data set(s) could be a baseline report, using data based off performance that occurred in CY 2015 or CY 2016 for applicable and available quality and resource use data. In the event that 2017 is the first MIPS performance period (as proposed in section II.E.4. of this rule), we would not anticipate receiving the first set of data for MIPS until 2018 (as proposed in section II.E.5. of this rule). At a minimum for the first year, we propose to provide performance feedback on an annual basis since the first performance feedback, required on July 1, 2017 would be based on historic data set(s). As the program evolves, and we can operationally assess/analyze the MIPS data, we may consider in future years providing performance feedback on a more frequent basis, such as quarterly. Section 1848(q)(12)(A)(i) of the Act requires the performance feedback to be provided “timely” (such as quarterly), which is our goal as MIPS evolves. In addition, we seek comments on whether we should include first year measures in the performance feedback, meaning new measures that have been in use for less than 1 year, regardless of submission methods. The reasoning behind first-year measures potentially not being reported is we need to review the data from the measure before this data is incorporated into performance feedback, as we want to ensure the data we are providing in the performance feedback is useful and has usability for our stakeholders. We request comments on these proposals.

In future years and as the program evolves, we intend to seek comment on the template, including but not limited to the data fields, for performance feedback. While section 1848(q)(12)(A)(i) of the Act only requires us to provide performance feedback for the quality and resource use performance categories, we understand that the CPIA and advancing care information performance categories are important MIPS data. Commenters to the MIPS and APMs RFI noted that CMS should consult with stakeholders to ensure this performance feedback is useful before this data is provided to MIPS eligible clinicians. Therefore, we may consider including feedback on the performance categories of CPIA and advancing care information in future years. Further, before we consider adding CPIA and advancing care information data to the performance feedback we would like to engage in stakeholder outreach to understand what data fields might be helpful and usable to MIPS eligible clinicians. Regarding the MIPS CPS, this is something we are targeting to provide annually as part of the performance feedback as the program evolves. As technically feasible, we are also planning to provide data fields such as the CPS and each of the four performance categories in future performance feedback once MIPS data becomes available. In addition, we plan to explore the possibility of including the MIPS adjustment factor (and, as applicable, the additional MIPS adjustment factor) in future performance feedback. We seek comment on the frequency with which this performance feedback should be provided, considerations for including CPIA and advancing care information, and data fields that should be included in the performance feedback as this program evolves.

(b) APM Entities

We proposed in section II.E.5.h.(15) of this rule that MIPS eligible clinicians who participate in APM Entities would receive performance feedback, as technically feasible.

(2) Mechanisms

Under section 1848(q)(12)(A)(ii) of the Act, the Secretary may use one or more mechanisms to make performance feedback available, which may include use of a web-based portal or other mechanisms determined appropriate by the Secretary. For the quality performance category, described in section 1848(q)(2)(A)(i) of the Act, the feedback shall, to the extent an eligible clinician chooses to participate in a data registry for purposes of MIPS (including registries under sections 1848(k) and (m)) of the Act, be provided based on performance on quality measures reported through the use of such registries. With respect to any other performance category (that is, resource use, CPIA, or advancing care information), the Secretary shall encourage provision of feedback through qualified clinical data registries (QCDRs) as described in sections 1848(m)(3)(E) of the Act.

We understand that the PQRS and VM programs have employed various communication strategies to notify health care providers of the availability of their PQRS Feedback Reports and QRURs, respectively, through the CMS portal. However, many health care providers are still unaware of these reports and/or have difficulty accessing their reports in the portal. Further, we are aware that some health care providers perceive the current reports as complex and often difficult to understand; while others find the QRURs, and the drill down data included in them on the Medicare beneficiaries they serve, very useful. We are continuing to work with stakeholders to improve the usability of these reports. As we transition to MIPS, we are committed to ensuring that eligible clinicians are able to access their performance feedback, and that the data are easy to understand while providing information that will help drive quality improvement. We propose to initially make performance feedback available using a CMS designated system, such as a web-based portal; if technically feasible perhaps an interactive dashboard. As further discussed in section II.E.7.e. of this proposed rule, we also propose to leverage additional mechanisms such as health IT vendors, registries, and QCDRs to help disseminate data/information contained in the performance feedback to eligible clinicians, where applicable. At this time, we believe that these additional mechanisms will only be able to provide information on the quality performance category for MIPS in regard to performance feedback.

We plan to coordinate with third party intermediaries such as health IT vendors and QCDRs as MIPS evolves to enable additional feedback to be sent on the resource use, advancing care information and CPIA performance categories. We seek comment on this for future rulemaking.

Comments received through the MIPS and APMs RFI noted issues associated with access to the current Feedback Reports for PQRS. Specifically, comments were received noting issues with Enterprise Identity Management (EIDM) and access to the portal to view PQRS Feedback Reports. Commenters also noted the need for a mechanism to be put in place to notify EPs when their PQRS Feedback Report is available. We propose to use the information contained in the provider or supplier's Medicare enrollment records, and stored in the Provider Enrollment, Chain, and Ownership System (PECOS), as the system of records for eligible clinicians' contact information that should be used when the MIPS performance feedback is available. It is therefore critical that eligible clinicians ensure that their Medicare enrollment records (especially in regard to phone and email contact information) are updated, meaning current, on a consistent basis in PECOS. If more than one email address is listed, then the email address that should be used for communication should be designated. We also intend to provide education and outreach on how to access performance feedback. We seek comment on additional means that could be used to notify or contact MIPS eligible clinicians and groups when their performance feedback is available.

(3) Use of Data

Under section 1848(q)(12)(A)(iii) of the Act, for purposes of providing performance feedback, the Secretary may use data, for a MIPS eligible clinician, from periods prior to the current performance period and may use rolling periods in order to make illustrative calculations about the performance of such professional. We believe “illustrative calculations” means an interim, snap shot in time of performance, or perhaps a “dry-run” of the data including measure rates. This would provide an indication of how a MIPS eligible clinician might be performing, but would not be conclusive. Since MIPS will not likely have comparable data until year 3 of the program, these “illustrative calculations” could be based on historical data sets available to CMS until actual data for MIPS is available.

(4) Disclosure Exemption

As stated under section 1848(q)(12)(A)(iv) of the Act, feedback made available under section 1848(q)(12)(A) of the Act shall be exempt from disclosure under 5 U.S.C. 552 (the Freedom of Information Act).

(5) Receipt of Information

Section 1848(q)(12)(A)(v) of the Act, states that the Secretary may use the mechanisms established under section 1848(q)(12)(A)(ii) of the Act to receive information from professionals. This allows for expanded use of the feedback mechanism to not only provide feedback on performance to eligible clinicians, but to also receive information from professionals.

We intend to explore the possibility of adding this feature to the CMS designated system, such as a portal, in future years under MIPS. This feature could be a mechanism where eligible clinicians can send their feedback (that is, if they are experiencing issues accessing their data, technical questions about their data, etc.) to CMS. We appreciate that eligible clinicians may have questions regarding the information contained in their performance feedback. In order to assist eligible clinicians, we intend to establish resources, such as a helpdesk or offer technical assistance, to help address questions with the goal of linking these resource features to the CMS designated system, such as a portal.

Additionally, we seek comment on the types of information eligible clinicians would like to send to CMS via this mechanism.

(6) Additional Information—Type of Information

Section 1848(q)(12)(B)(i) of the Act, states that beginning July 1, 2018, the Secretary shall make available to MIPS eligible clinicians information about the items and services for which payment is made under Title 18 that are furnished to individuals who are patients of MIPS eligible clinicians by other suppliers and providers of services. This information may be made available through mechanisms determined appropriate by the Secretary, such as the proposed CMS designated system that would also provide performance feedback. Section 1848(q)(12)(B)(ii) of the Act specifies that the type of information provided may include the name of such providers, the types of items and services furnished, and the dates items and services were furnished. Historical data regarding the total, and components of, allowed charges (and other figures as determined appropriate by the Secretary) may also be provided. We seek comment on the type of information MIPS eligible clinicians would find useful and the preferred mechanisms to provide such information, as well as, arrangements that should be in place regarding this data (that is, eligible clinicians sharing data). We also seek comment as to whether additional information regarding beneficiaries attributed to a MIPS eligible clinician under the resource use performance category or information about which MIPS eligible clinician(s) beneficiaries to whom a given MIPS eligible clinician provides services were attributed would be useful feedback in regards to quality improvement efforts.

(7) Performance Feedback Template

The performance feedback under section 1848(q)(12)(A) of the Act is meant to be meaningful and usable to eligible clinicians. In an effort to ensure these data are tailored to the needs of eligible clinicians, we solicited comment through the MIPS and APMs RFI and received numerous comments regarding overall format of the performance feedback template. Suggestions were made on what this feedback should include for MIPS. We intend to collaborate with stakeholders outside of notice-and-comment rulemaking on how the performance feedback should look for MIPS; as well as, what data elements would be useful for eligible clinicians. We seek comment on the fields that should be included in the performance feedback template for MIPS eligible clinicians.

b. Announcement of Result of Adjustments

Section 1848(q)(7) of the Act requires that under the MIPS, the Secretary shall, not later than 30 days prior to January 1 of the year involved, make available to MIPS eligible clinicians the MIPS adjustment factor (and, as applicable, the additional MIPS adjustment factor) applicable to the eligible clinician for items and services furnished by the professional for such year. The Secretary may include such information in the confidential feedback under section 1848(q)(12) of the Act.

If technically feasible, we propose to include the MIPS adjustment factor (and, as applicable, the additional MIPS adjustment factor) in the performance feedback for eligible clinicians provided under section 1848(q)(12)(A) of the Act. If it is not technically feasible to provide this information in the performance feedback, we propose to make it available through another mechanism as determined appropriate by the Secretary (such as a portal or a CMS designated Web site) and seek comment on mechanisms that might be appropriate. The first announcement will be available no later than December 1, 2018 to meet statutory requirements. We request comment on these proposals.

c. Targeted Review

Section 1848(q)(13)(A) of the Act requires the establishment of a process under which a MIPS eligible clinician may seek an informal review of the calculation of the MIPS adjustment factor (or factors) applicable to such MIPS eligible clinician for a year.

We recognize that a principled approach to requesting and conducting a targeted review is required under the MACRA in order to minimize burdens on MIPS eligible clinicians and ensure transparency under MIPS. We also believe it is important to retain the flexibility to modify MIPS eligible clinicians' CPS or payment adjustment based on the results of targeted review. This will lend confidence to the determination of the CPS and payment adjustments, as well as, providing finality for the MIPS eligible clinician after the targeted review is completed. It will also minimize the need for claims reprocessing. We are proposing an approach below that outlines the factors that we would use to determine if a targeted review may be conducted. In keeping with the statutory direction that this process be “informal,” we have attempted to minimize the associated burden on the MIPS eligible clinician to the extent possible.

In accordance with section 1848(q)(13)(A) of the Act, we propose at § 414.1385 to adopt a targeted review process under MIPS wherein a MIPS eligible clinician may request that we review the calculation of the MIPS adjustment factor under section 1848(q)(6)(A) of the Act and, as applicable, the calculation of the additional MIPS adjustment factor under section 1848(q)(6)(C) of the Act applicable to such MIPS eligible clinician for a year. Because this review will be limited to the calculation of the MIPS adjustment factor and, as applicable, the additional MIPS adjustment factor, we anticipate we may find it necessary to review data related to the measures and activities and the calculation of the CPS according to the defined methodology. The following are examples of circumstances under which a MIPS eligible clinician may wish to request a targeted review. This is not a comprehensive list of circumstances:

• The MIPS eligible clinician believes that measures or activities submitted to CMS during the submission period and used in the calculations of the CPS and determination of the adjustment factors have calculation errors or data quality issues. These submissions could be with or without the assistance of a third party intermediary; or
• The MIPS eligible clinician believes that there are certain errors made by CMS, such as performance category scores were wrongly assigned to the MIPS eligible clinician (for example, the MIPS eligible clinician should have been subject to the low-volume threshold exclusion and should not have received a performance category score).

We believe that a fair targeted review request process requires accessibility to all MIPS eligible clinicians within a reasonable period of time and provides electronic and telephonic communication for questions regarding the targeted review process, as well as for the actual request for review and receipt of the decision on that request. The targeted review process will use the same help desk support mechanism as is provided for MIPS as a whole.

We further propose at § 414.1385 to adopt the following general process for targeted reviews under section 1848(q)(13)(A):

• A MIPS eligible clinician electing to request a targeted review may submit their request within 60 days (or a longer period specified by us) after the close of the data submission period. All requests for targeted review must be submitted by July 31 after the close of the data submission period or by a later date that we specify in guidance.
• We will provide a response with our decision on whether or not a targeted review is warranted. If a targeted review is warranted, the timeline for completing that review may be dependent on the number of reviews requested (for example, multiple reviews versus a single review by one MIPS eligible clinician) and general nature of the review.
• As this process is informal and the statute does not require a formal appeals process, we will not include a hearing process. The MIPS eligible clinician may submit additional information to assist in their targeted review at the time of request. If we or our contractors request additional information from the MIPS eligible clinician, the supporting information must be received from the MIPS eligible clinician by us or our contractors within 10 calendar days of the request. Non-responsiveness to the request for additional information will result in the closure of that targeted review request, although another review request may be submitted if the targeted review submission deadline has not passed.
• Since this is an informal review process and given the limitations on review under section 1848(q)(13)(B) of the Act, decisions based on the targeted review will be final, and there will be no further review or appeal.

If a request for targeted review is approved, the outcome of such review may vary. For example, we may determine that the clinician should have been excluded from MIPS, re-distribute the weights of certain performance categories within the CPS (for example, if a performance category should have been weighted at zero), or recalculate a performance category score in accordance with the scoring methodology for the affected category, if technically feasible.

We request comments on these proposals.

d. Review Limitation

Section 1848(q)(13)(B) of the Act, as added by section 101(c)(1) of the MACRA, provides there shall be no administrative or judicial review under sections 1869 and 1878 of the Act, or otherwise of the following:

• The methodology used to determine the amount of the MIPS adjustment factor and the amount of the additional MIPS adjustment factor and the determination of such amounts;
• The establishment of the performance standards and the performance period;
• The identification of measures and activities specified for a MIPS performance category and information made public or posted on our Physician Compare Web site; and
• The methodology developed that is used to calculate performance scores and the calculation of such scores, including the weighting of measures and activities under such methodology.

We propose at § 414.1385 to implement these provisions as written in the statute.

We would reject any requests for targeted review under section 1848(q)(13)(A) of the Act that focus on the areas precluded from review under section 1848(q)(13)(B) of the Act. We request comments on this proposal.

e. Data Validation and Auditing

Our experience with the PQRS, VM and Medicare EHR Incentive Programs, has demonstrated the value of data validation and auditing as an important part of program integrity, which is necessary to ensure valid, reliable data. The current voluntary data validation process for PQRS and the audit process for the Medicare EHR Incentive Program are multi-step processes. We communicate the types of data elements that may be included for data validation across multiple Web sites and our documents. This includes defining specific data that may be abstracted from the certified EHR technology, as well as other documented records.

As we begin the MIPS, our strategy is to combine our past program integrity processes of the data validation process used in PQRS, and the auditing process used in the Medicare EHR Incentive Program into one set of requirements for MIPS eligible clinicians and groups, which we refer to as “data validation and auditing.” Based on our need for valid and reliable data on which to base a MIPS eligible clinician's or group's payment, we propose certain requirements for MIPS eligible clinicians and groups submitting data for the 2017 performance period (see section II.E.4) under MIPS. Further, we propose at § 414.1390 to selectively audit MIPS eligible clinicians on a yearly basis, and that if a MIPS eligible clinician or group is selected for audit, the MIPS eligible clinician or group would be required to do the following in accordance with applicable law:

• Comply with data sharing requests, providing all data as requested by us or our designated entity. All data must be shared with CMS or our designated entity within 10 business days or an alternate time frame that is agreed to by CMS and the MIPS eligible clinician or group. Data would be submitted via email, facsimile, or an electronic method via a secure Web site maintained by CMS.
• Provide substantive, primary source documents as requested. These documents may include: Copies of claims, medical records for applicable patients, or other resources used in the data calculations for MIPS measures, objectives and activities. Primary source documentation also may include verification of records for Medicare and non-Medicare beneficiaries where applicable.

We propose that we would monitor MIPS eligible clinicians and groups on an ongoing basis for data validation, auditing, program integrity issues and instances of non-compliance with MIPS requirements. If a MIPS eligible clinician or group is found to have submitted inaccurate data for MIPS, we propose that we would reopen, revise, and recoup any resulting overpayments in accordance with the rules set forth at § 405.980 (re-opening rules), § 450.982 and § 450.984 (revising rules); and § 405.370 and § 405.373 (recoupment rules). It is important to note that at § 405.980(b)(3) there is an exception whereby we have the authority to re-open at any time for fraud or similar fault. If we re-open the initial determination we must revise it, and send out a notice of the revised determination under § 450.982. We also propose that we would recoup any payments from the MIPS eligible clinician by the amount of any debts owed to us by the MIPS eligible clinician and likewise, we would recoup any payments from the group by the amount of any debts owed to us by the group. We also note that we would need to limit each such data validation and audit request to the minimum data necessary to conduct validation.

We propose all MIPS eligible clinicians and groups that submit data to CMS electronically must attest to the accuracy and completeness to the best of their knowledge of any data submitted to us. This attestation will occur prior to any electronic data submissions, via a Web site maintained by CMS.

We request comments on these proposals.

9. Third Party Data Submission

One of our strategic goals in developing MIPS includes developing a program that is meaningful, understandable, and flexible for participating MIPS eligible clinicians. One way we believe this will be accomplished is through flexible reporting options to accommodate different practices and make measurement meaningful. We believe this goal can be accomplished by allowing MIPS eligible clinicians the flexibility of using third party intermediaries to collect or submit data on their behalf. Specifically, qualified registries, QCDRs, health IT vendors that obtain data from an eligible clinician's certified EHR technology, and CMS-approved survey vendors as discussed in the following proposed policies. In this section, we are specifying the requirements that must be met to become a third party intermediary.

In the PQRS program, quality measures data may be collected or submitted by third party vendors on behalf of an individual EP or group by: (1) A registry; (2) a QCDR; or (3) an EHR vendor that obtains data from an EP's certified EHR technology; or (4) a CMS-approved survey vendor. We propose at § 414.1400(a)(1) that MIPS data may be submitted by third party intermediaries on behalf of a MIPS eligible clinician or group by: (1) A qualified registry; (2) a QCDR; (3) a health IT vendor; or (4) a CMS-approved survey vendor. Furthermore, we propose at § 414.1400(a)(3) that third party intermediaries must meet all the requirements designated by CMS as a condition of their qualification or approval to participate in MIPS as a third party intermediary. As proposed at § 414.1400(a)(3)(ii), all submitted data must be submitted in the form and manner specified by CMS.

In the MIPS and APMs RFI, we solicited feedback on how we should address data integrity, testing and standards, and review and qualification processes for QCDRs. Subsequently, we also met with several organizations that were either a QCDR or are in the process of becoming a QCDR. Commenters agreed that data quality is a critical issue for QCDRs. To address some of the data quality concerns, some commenters suggested having processes in place in advance of reporting that could mitigate data errors. For example, this could include a process to reconcile TIN and NPI combinations. Several commenters also suggested limiting submission mechanisms to one submission mechanism per performance category to the extent possible. Commenters generally agreed that QCDRs should be required to submit data using uniform submission standards, with several suggesting the use of the Quality Reporting Document Architecture (QRDA) standard, which certified EHR technology is required to support.

Most commenters noted that uniform standards would ease participation by MIPS eligible clinicians and reduce barriers to entry. Others noted that we should work with ONC and the standards development organization Health Level Seven (HL7) to improve the QRDA standard for current submissions, and that in the future, we should prepare to support emerging standards such as Fast Healthcare Interoperability Resources. Commenters also noted that use of QRDA will align CMS requirements and ONC certification requirements as ONC's 2015 Edition Certification requires that all health information technology (IT) modules used for the submission of CQM data must at least be certified to the QRDA standard. Requiring QCDRs to use QRDA could help reduce vendor interface costs for MIPS eligible clinicians already using certified EHR technology and who desire to participate in registry reporting. Commenters also directed our attention towards the 2015 Edition Certification for additional information on improved test methods and to address historic issues and inaccuracies observed with past calculation and reporting of quality and performance data. With regard to testing, commenters were divided about whether we should require QCDR-specific testing. Several noted that certified EHR technology that support QCDRs have been tested already and that onerous testing may discourage participation. Commenters in favor of testing recommended a degree of flexibility in the early years of the program. Suggestions for testing included the use of comprehensive specifications and accurate testing tools far enough in advance of the performance period to allow developers and implementers to conduct robust testing. These specifications could be included in an Implementation Guide. Opportunities for early testing, using sample data was also emphasized. Commenters did express concern on the amount of time needed for troubleshooting and fixing errors early enough in the testing process such as format, content, and measure accuracy. Commenters suggested several ways we might implement testing, recommending that we:

• Test the accuracy, completeness, and reliability of measure calculations for specific, individual measures.
• Test the feasibility of data collection requirements.
• Pilot new CQMs before release; establish a regular schedule of CQM revisions, and ensure adequate time is allowed for implementation of the revisions.
• Align the ONC Health IT Certification program and CMS testing requirements for data submission.
• Expand the test data sets used by the Cypress Testing Tool. More information on the Cypress Testing Tool is available at: http://projectcypress.org/about.html .

There was a strong consensus that MIPS eligible clinicians should not be penalized for signing up with an entity that purported to offer reliable services but then was unable to accurately submit data to us. Several commenters suggested that entities that do not meet standards move to a probationary phase and eventually be prohibited from periods of future participation until standards are met. However, commenters also cautioned us not to move too quickly in moving entities to a probationary phase because many QCDRs are run by medical specialty societies and if they were to be disqualified to the detriment of physicians participating, it would also diminish physician enthusiasm for future submission of data.

Commenters had mixed responses regarding how to resolve inaccurate data submission problems when time did not allow for continued review. Commenters felt we should use a “trust but validate” methodology, allowing the QCDR to recalculate the performance rate or authorizing us to do so, but also that we should have validation processes in place as well once the recalculation of the performance rate occurs. Ultimately, we would need to be able to calculate all rates based on a submitted numerator and denominator. Commenters suggested that MIPS eligible clinicians should be assessed an average score or a “pass” for the MIPS quality performance category if data problems cannot be resolved in a timely manner or at the least not be penalized due to data errors outside their control. One commenter suggested use of a Data Quality Management (DQM) program for MIPS eligible clinicians that includes early data qualification evaluation processes to take advantage of feedback and assessments with thresholds for acceptance of data. MIPS eligible clinicians who demonstrate effort toward achieving high quality data submissions but were not able to meet the threshold should be chaperoned to that target and provided with guidance.

Commenters were also divided about our review and qualification of QCDRs to ensure our form and manner requirements are met. Several commenters were concerned with a CMS process in addition to an ONC certification process and recommended we work with ONC to align their certification to address our requirements for QCDRs. Commenters suggested that we also develop more robust implementation guides, and enhance our submission engine validation tool (SEVT).

a. Qualified Clinical Data Registries (QCDRs)

Section 1848(q)(1)(E) of the Act requires the Secretary to encourage the use of QCDRs under section 1848(m)(3)(E) of the Act in carrying out MIPS. Section 1848(q)(5)(B)(ii)(I) of the Act requires the Secretary, under the CPS methodology, to encourage MIPS eligible clinicians to report on applicable measures with respect to the quality performance category through the use of certified EHR technology and QCDRs. Section 1848(q)(2)(B)(iii)(II) of the Act requires that the CPIA subcategories specified by the Secretary include population management, such as monitoring health conditions of individuals to provide timely health care interventions or participation in a QCDR. Section 1848(q)(12)(A)(ii) of the Act requires the Secretary to encourage the provision of performance feedback through QCDRs.

Section 1848(m)(3)(E)(i) of the Act requires the Secretary to establish requirements for an entity to be considered a QCDR, which must include a requirement that the entity provide the Secretary with such information, at such times, and in such manner, as the Secretary determines necessary to carry out section 1848(m) of the Act. Section 1848(m)(3)(E)(iv) of the Act requires the Secretary to consult with interested parties in carrying out section 1848(m)(3)(E) of the Act.

Currently, the QCDR reporting mechanism provides a method to satisfy PQRS requirements based on satisfactory participation. We propose that entities interested in becoming a QCDR for MIPS go through a qualification process. This includes the QCDR meeting the definition of a QCDR, self-nomination requirements, and the requirements of a QCDR, including the deadlines listed below. This qualification process allows us to ensure that the entity has the capability to successfully report MIPS eligible clinicians' data to us and allows for review and approval of the QCDR's proposed non-MIPS quality measures. We intend to compile and post a list of entities that we “qualify” to submit data to us as a QCDR for purposes of MIPS on a Web site maintained by CMS.

Section 1848(q)(1)(E) of the Act encourages the use of QCDRs in carrying out the MIPS. Although section 1848(q)(5)(B)(ii)(I) of the Act specifically requires the Secretary to encourage MIPS eligible clinicians to use QCDRs to report on applicable measures with respect to the quality performance category and section 1848(q)(12)(A)(ii) of the Act requires the Secretary to encourage the provision of performance feedback through QCDRs, the statute does not specifically address usage of QCDRs for the other MIPS performance categories. Although we could limit the usage of QCDRs to assessing the quality performance category under MIPS and providing performance feedback, we believe it would be less burdensome for MIPS eligible clinicians if we expand the QCDRs capabilities. By allowing QCDRs to report on the quality, advancing care information, and CPIA performance categories we would alleviate the need for individual MIPS eligible clinicians and groups to use a separate mechanism to report data for these performance categories. It is important to note that no data will need to be reported for the resource use performance category since these measures are administrative claims-based. Therefore, we are proposing at § 414.1400(a)(2) to expand QCDRs' capabilities by allowing QCDRs to submit data on measures, activities, or objectives for any of the following MIPS performance categories:

(i) Quality;

(ii) CPIA; or

(iii) Advancing care information, if the MIPS eligible clinician or group is using certified EHR technology.

We believe this approach would permit a single QCDR to report on the quality, advancing care information, and CPIA performance category requirements for MIPS and should mitigate the risks, costs, and burden of MIPS eligible clinicians having to report multiple times to meet the requirements of MIPS.

We propose to define a QCDR at § 414.1305 as a CMS-approved entity that has self-nominated and successfully completed a qualification process to determine whether the entity may collect medical and/or clinical data for the purpose of patient and disease tracking to foster improvement in the quality of care provided to patients. Examples of the types of entities that may qualify as QCDRs include, but are not limited to, regional collaboratives and specialty societies using a commercially available software platform, as appropriate.

(1) Establishment of an Entity Seeking To Qualify as a QCDR

We propose at § 414.1400(c) the establishment of a QCDR entity is required as follows: for an entity to become qualified for a given performance period as a QCDR, the entity must be in existence as of January 1 of the performance period for which the entity seeks to become a QCDR (for example, January 1, 2017, to be eligible to participate for purposes of performance periods beginning in 2017). The QCDR must have at least 25 participants by January 1 of the performance period. These participants do not need to be using the QCDR to report MIPS data to us; rather, they need to be submitting data to the QCDR for quality improvement.

(2) Self-Nomination Period

For the 2017 performance period we propose at § 414.1400(b) a self-nomination period from November 15, 2016 until January 15, 2017. For future years of the program, starting with the 2018 performance period, we propose to establish the self-nomination period from September 1 of the prior year until November 1 of the prior year. Entities that desire to qualify as a QCDR for the purposes of MIPS for a given performance period would need to self-nominate for that year and provide all information requested by CMS at the time of self-nomination. Having qualified as a QCDR in a prior year does not automatically qualify the entity to participate in MIPS as a QCDR in subsequent performance periods. For example, a QCDR may choose not to continue participation in the program in future years, or the QCDR may be precluded from participation in a future year due to multiple data or submission errors as noted below. Finally, QCDRs may want to update or change the measures or services or performance categories they intend to provide. As such, CMS believes an annual self-nomination process is the best process to ensure accurate information is conveyed to MIPS eligible clinicians and accurate data is submitted to MIPS.

We propose to require other information (described below) of QCDRs at the time of self-nomination. If an entity becomes qualified as a QCDR, they will need to sign a statement confirming this information is correct prior to listing it on their Web site. Once we post the QCDR on our Web site, including the services offered by the QCDR, we will require the QCDR to support these services/measures for its clients as a condition of the entity's qualification as a QCDR for purposes of MIPS. Failure to do so will preclude the QCDR from participation in MIPS in the subsequent year.

(3) Information Required at the Time of Self-Nomination

We propose that a QCDR must provide the following information to us at the time of self-nomination to ensure that QCDR data is valid:

• Organization Name (Specify Sponsoring Organization name and software vendor name if the two are different. For example, a specialty society in collaboration with a software vendor).
• MIPS performance categories (that is, categories for which the entity is self-nominating. For example, quality, advancing care information, and/or CPIA).
• Performance Period.
• Vendor Type (for example, qualified clinical data registry).
• Provide the method(s) by which the entity obtains data from its customers for each performance category for which it is approved: Claims, web-based tool, practice management system, certified EHR technology, other (please explain). If a combination of methods (Claims, web-based tool, Practice Management System, certified EHR technology, and/or other) is utilized, the entity should state which method(s) it utilizes to collect data (for example, performance numerator and denominator).
• Indicate the method the entity will use to verify the accuracy of each TIN/NPI it is intending to submit (for example, National Plan and Provider Enumeration System (NPPES), CMS claims, tax documentation).
• Describe the method that the entity will use to accurately calculate performance rates for quality measures based on the appropriate measure type and specification. For composite measures or measures with multiple performance rates, the entity must provide us with the methodology the entity uses to calculate these composite measures and measures with multiple performance rates. The entity should be able to report to us a calculated composite measure rate if applicable.
• Describe the method that the entity will use to accurately calculate performance data for CPIA and advancing care information based on the appropriate parameters or activities.
• Describe the process that the entity will use for completion of a randomized audit of a subset of data prior to the submission to us (for all performance categories the QCDR is submitting data on, that is, quality, CPIA, and advancing care information, as applicable). Periodic examinations may be completed to compare patient record data with submitted data and/or ensure MIPS quality measures or other performance category (CPIA, advancing care information) activities were accurately reported and performance calculated based on the appropriate measure specifications (that is, accuracy of numerator, denominator, and exclusion criteria) or performance category requirements.
• Provide information on the entity's process for data validation for both individual MIPS eligible clinicians and groups within a data validation plan. For example, for individuals it is encouraged that 3 percent of the TIN/NPIs submitted to us by the QCDR be sampled with a minimum sample of 10 TIN/NPIs or a maximum sample of 50 TIN/NPIs. For each TIN/NPI sampled, it is encouraged that 25 percent of the TIN/NPI's patients (with a minimum sample of five patients or a maximum sample of 50 patients) should be reviewed for all measures applicable to the patient.
• Provide the results of the executed data validation plan by May 31 of the year following the performance period. If the results indicate the QCDR's validation reveals inaccuracy or low compliance provide to CMS an improvement plan. Failure to implement improvements may result in the QCDR being placed in a probationary status or disqualification from future participation.
• For non-MIPS quality measures, if the measure is risk-adjusted, the QCDR is required to provide details to CMS on their risk adjustment methodology (risk adjustment variables, and applicable calculation formula) at the time of the QCDR's self-nomination. The QCDR must submit the risk adjusted results to CMS when submitting a risk-adjusted measure on behalf of the QCDR's MIPS eligible clinicians for the performance period.

(4) QCDR Requirements for Data Submission

In addition, we propose that a QCDR must perform the following functions:

• For measures under the quality performance category and as proposed at § 414.1400(a)(4)(i), if the data is derived from certified EHR technology, the QCDR must be able to indicate this data source.
• QCDRs must provide complete quality measure specifications including data elements to us for non-MIPS quality measures intended for reporting from certified EHR technology.
• QCDRs must provide a plan to risk adjust (if appropriate for the measure) the non-MIPS quality measures data for which it collects and intends to transmit to us and must submit the risk-adjusted results (not the non-risk adjusted rates), to CMS. The risk adjustment methodology (formula and variables) must be integrated with the complete quality measure specifications. Specifically, for risk-adjusted non-MIPS quality measures, a QCDR is required to provide details to CMS on their risk adjustment methodology. The data elements used for risk adjustment may vary by measure and measure type. The risk adjustment methodology, including the risk adjustment variables, must be posted along with the measure's specifications on the QCDR's Web site. CMS believes risk-adjustment for certain outcomes measures is important to account for the differences in the complexities of care provided to different patients. That is, some patients may have additional comorbidities which could affect their response to treatment and subsequently their outcome. Risk adjustment will help offset potential poorer outcomes for those MIPS eligible clinicians caring for sicker patients.
• QCDRs submitting MIPS quality measures that are risk-adjusted (and have the risk-adjusted variables and methodology listed in the measure specifications) must submit the risk-adjusted measure results to CMS when submitting the data for these measures.
• Submit quality, advancing care information, or CPIA data and results to us in the applicable MIPS performance categories for which the QCDR is providing data.
• A QCDR must have in place mechanisms for the transparency of data elements and specifications, risk models, and measures. That is, we expect that the non-MIPS measures and their data elements (that is, specifications) comprising these measures be listed on the QCDR's Web site unless the measure is a MIPS measure, in which case the specifications will be posted by us.
• Submit to us data on measures, activities, and objectives for all patients, not just Medicare patients.
• Provide timely feedback, at least 6 times a year, on all of the MIPS performance categories that the QCDR will report to us. That is, if the QCDR will be reporting on data for the CPIA, advancing care information, or quality performance category, all results as of the feedback report date should be included in the information sent back to the MIPS eligible clinician. The feedback should be given to the individual MIPS eligible clinician or group (if participating as a group) at the individual participant level or group level, as applicable, for which the QCDR reports. The QCDR is only required to provide feedback based on the MIPS eligible clinician's data that is available at the time the feedback report is generated.
• Possess benchmarking capacity (for non-MIPS quality measures) that compares the quality of care a MIPS eligible clinician provides with other MIPS eligible clinicians performing the same quality measures. For non-MIPS measures the QCDR must provide us, if available, data from years prior (for example, 2015 data for the 2017 MIPS performance period) before the start of the performance period. In addition, the QCDR must provide us, if available, with the entire distribution of the measure's performance broken down by deciles. As an alternative to supplying this information to us, the QCDR may post this information on their Web site prior to the start of the performance period, to the extent permitted by applicable privacy laws.
• QCDRs must comply with any request by us to review the data submitted by the QCDR for purposes of MIPS in accordance with applicable law. Specifically, data requested would be limited to the minimum necessary for us to carry out, for example, health care operations or health oversight activities.
• Mandatory participation in ongoing support conference calls hosted by us (approximately one call per month), including an in-person QCDR kick-off meeting (if held) at our headquarters in Baltimore, MD. More than one unexcused absence could result in the QCDR being precluded from participation in the program for that year. If a QCDR is precluded from participation in MIPS, the individual MIPS eligible clinician or group would need to find another QCDR or utilize another data submission mechanism to submit their MIPS data.
• Agree that data inaccuracies including (but not limited to) TIN/NPI mismatches, formatting issues, calculation errors, data audit discrepancies affecting in excess of 3 percent of the total number of MIPS eligible clinicians submitted by the QCDR may result in notations on our qualified QCDR posting of low data quality and would place the QCDR on probation (if they decide to self-nominate for the next program year). If the QCDR does not reduce their data error rate below 3 percent in the subsequent year, they would continue to be on probation and have their listing on the CMS Web site continue to note the poor quality of the data they are submitting for MIPS. Data errors affecting in excess of 5 percent of the MIPS eligible clinicians submitted by the QCDR may lead to the disqualification of the QCDR from participation in the following year's program. As we gain additional experience with QCDRs, we intend to revisit and enhance these thresholds in future years.
• Be able to submit results for at least six quality measures including one cross-cutting measure and one outcome measure. If an outcome measure is not available, be able to submit results for at least one other high priority measure (appropriate use, patient safety, efficiency, patient experience, and care coordination measures). If no outcome measure is available, then the QCDR must provide a justification for not including an outcome measure.
• QCDRs may request to report on up to 30 quality measures not in the annual list of MIPS quality measures. Full specifications will need to be provided to us at the time of self-nomination. CMS will review the quality measures and determine if they are appropriate for QCDR reporting.
• Enter into and maintain with its participating clinicians an appropriate Business Associate agreement that provides for the QCDR's receipt of patient-specific data from an individual MIPS eligible clinician or group, as well as the QCDR's disclosure of quality measure results and numerator and denominator data and/or patient specific data on Medicare and non-Medicare beneficiaries on behalf of MIPS eligible clinicians and groups.
• Obtain and keep on file signed documentation that each holder of an NPI whose data are submitted to the QCDR, has authorized the QCDR to submit quality measure results, CPIA measure and activity results, advancing care information objective results and numerator and denominator data and/or patient-specific data on Medicare and non-Medicare beneficiaries to CMS for the purpose of MIPS participation. This documentation must be obtained at the time the MIPS eligible clinician or group signs up with the QCDR to submit MIPS data to the QCDR and must meet the requirements of any applicable laws, regulations, and contractual business associate agreements. Groups participating in MIPS via a QCDR may have their group's duly authorized representative grant permission to the QCDR to submit their data to us. If submitting as a group, each individual MIPS eligible clinician does not need to grant their individual permission to the QCDR to submit their data to us.
• Not be owned and managed by an individual locally owned single specialty group (for example, single specialty practices with only one practice location or solo practitioner practices are prohibited from self-nominating to become a qualified QCDR).
• Be able to separate out and report on all payers including Medicare Part B FFS patients and non-Medicare patients.
• Provide the measure numbers for the MIPS quality measures on which the QCDR is reporting.
• Provide the measure title for the MIPS quality measures and CPIAs (if applicable) on which the QCDR is reporting.
• Report the number of eligible instances (reporting denominator).
• Report the number of instances a quality service is performed (performance numerator).
• Report the number of performance exclusions, meaning the quality action was not performed for a valid reason as defined by the measure specification.
• Comply with a CMS-specified secure method for data submission, such as submitting the QCDR's data in an XML file.
• Sign a document verifying the QCDR's name, contact information, cost for MIPS eligible clinicians or groups to use the QCDR, services provided, and the measures and specialty-specific measure sets the QCDR intends to report. Once posted, on the QCDR's or CMS Web site, the QCDR will need to support the measures/measure sets confirmed by the QCDR. Failure to do so will preclude the QCDR from participation in MIPS in the subsequent year.
• Must provide attestation statements during the data submission period that all of the data (quality measures, CPIAs, and advancing care information measures and objectives, if applicable) and results are accurate and complete.
• For purposes of distributing feedback reports to MIPS eligible clinicians, collect a MIPS eligible clinician's email addresses and have documentation from the MIPS eligible clinician authorizing the release of his or her email address.
• Be able to calculate and submit measure-level reporting rates or, upon request, the data elements needed to calculate the reporting and performance rates by TIN/NPI and/or TIN.
• Be able to calculate and submit, by TIN/NPI and/or TIN, a performance rate (that is the percentage of a defined population who receive a particular process of care or achieves a particular outcome based on a calculation of the measures' numerator and denominator specifications) for each measure on which the TIN/NPI and/or TIN reports or, upon request the Medicare beneficiary data elements needed to calculate the performance rates.
• Provide the performance period start date the QCDR will cover.
• Provide the performance period end date the QCDR will cover.
• Report the number of reported instances, performance not met, meaning the quality actions was not performed for no valid reason as defined by the measure specification.
• For data validation purposes, provide information on the entity's sampling methodology. For example, it is encouraged that 3 percent of the MIPS eligible clinicians be sampled with a minimum sample of 10 MIPS eligible clinicians or a maximum sample of 50 MIPS eligible clinicians. For each MIPS eligible clinicians sampled, it is encouraged that 25 percent of the MIPS eligible clinicians' patients (with a minimum sample of five patients or a maximum sample of 50 patients) should be reviewed for all measures applicable to the patient.
• Submit all of the measures (MIPS measures and non-MIPS measures) including specifications for the non-MIPS measures to CMS on a designated Web page. The measures must address a gap in care. Outcome or other high priority types of measures are preferred. Simple documentation or “check box” measures are discouraged.

(5) QCDR Measure Specifications Requirements

A QCDR must provide specifications for each measure, activity, or objective the QCDR intends to submit to CMS. We propose at § 414.1400(f) the QCDR must provide the following information:

• Provide descriptions and narrative specifications for, each measure activity, or objective for which it will submit to us by no later than January 15 of the applicable performance period for which the QCDR wishes to submit quality measures or other performance category (CPIA and advancing care information) data. In future years, starting with the 2018 performance period, those specifications must be provided to us by no later than November 1 prior to the applicable performance period for which the QCDR wishes to submit quality measures or other performance category (CPIA and advancing care information) data.
• For non-MIPS quality measures, the quality measure specifications must include: Name/title of measures, NQF number (if NQF-endorsed), descriptions of the denominator, numerator, and when applicable, denominator exceptions, denominator exclusions, risk adjustment variables, and risk adjustment algorithms. The narrative specifications provided must be similar to the narrative specifications we provide in our measures list. CMS will consider all non-MIPS measures submitted by the QCDR but the measures must address a gap in care and outcome or other high priority measures are preferred. Documentation or “check box” measures are discouraged. Measures that have very high performance rates already or address extremely rare gaps in care (thereby allowing for little or no quality distinction between MIPS eligible clinicians) are also unlikely to be approved for inclusion.
• For MIPS measures, the QCDR only needs to submit the MIPS measure numbers and/or the specialty-specific measure sets (if applicable).
• The QCDR must publicly post the measure specifications (no later than 15 days following our approval of these measure specifications) for each non-MIPS quality measure it intends to submit for MIPS. The QCDR may use any public format it prefers. Immediately following posting of the measures specification information, the QCDR must provide CMS with the link to where this information is posted. CMS will then post this information when it provides its list of QCDRs for the year.

(6) Identifying Non-MIPS Quality Measures

To clarify the definition of a non-MIPS quality measures for purposes of QCDRs submitting data for the MIPS quality performance category, we propose at § 414.1400(e) to consider the following types of quality measures to be non-MIPS quality measures:

• A measure that is not contained in the annual list of MIPS quality measures for the applicable performance period.
• A measure that may be in the annual list of MIPS quality measures but has substantive differences in the manner it is submitted by the QCDR. For example, if a MIPS quality measure is only reportable via the CMS Web Interface and a QCDR wishes to report this quality measure on behalf of its MIPS eligible clinicians, the quality measure would be considered a non-MIPS quality measure. This is because we would have only extracted the data collected from this quality measure using the CMS Web Interface, in which we utilize a claims-based assignment and sampling methodology to inform the groups on which patients they are to report, and the reporting of this quality measure would require changes to the way that the quality measure is calculated and reported to us via a QCDR instead of through the CMS Web Interface. Therefore, due to the substantive changes needed to report this quality measure via a QCDR, this CMS Web Interface quality measure would be considered a non-MIPS quality measure. CMS would not be able to directly compare MIPS eligible clinicians submitting the quality measure using the CMS Web Interface to those submitting the quality measure using the QCDR. Thus, this would be considered a non-MIPS quality measure.
• In addition, the CAHPS for MIPS survey currently could be submitted only using a CMS-approved survey vendor. Although the CAHPS for MIPS survey is proposed for inclusion in the MIPS measure set, we consider the changes that will need to be made available for reporting by individual MIPS eligible clinicians (and not as a part of a group) significant enough as to treat the CAHPS for MIPS survey as a non-MIPS quality measure for purposes of reporting the CAHPS for MIPS survey via a QCDR. To the extent that further clarification on the distinction between a MIPS and a non-MIPS measure is necessary, we will provide additional guidance on our Web site.

(7) Collaboration of Entities To Become a QCDR

In the CY 2016 PFS final rule (80 FR 71136 through 71138) we finalized our proposal to allow collaboration of entities to become a QCDR based on our experience with the qualifying entities wishing to become QCDRs for performance periods. We received feedback from organizations who expressed concern that the entity wishing to become a QCDR may not meet the requirements of a QCDR solely on its own. We believe this policy supporting entity collaboration should be continued under MIPS. Therefore, we are proposing at § 414.1400 that an entity that may not meet the requirements of a QCDR solely on its own but could do so in conjunction with another entity, would be eligible for qualification through collaboration with another entity.

We propose to allow that an entity that uses an external organization for purposes of data collection, calculation, or transmission may meet the definition of a QCDR provided the entity has a signed, written agreement that specifically details the relationship and responsibilities of the entity with the external organization effective as of September 1 the year prior to the year for which the entity seeks to become a QCDR (for example, September 1, 2016, to be eligible to participate for purposes of the 2017 performance period). Entities that have a mere verbal, non-written agreement to work together to become a QCDR by September 1 the year prior to the year for which the entity seeks to become a QCDR would not fulfill this proposed requirement. We request comments on these proposals.

b. Health IT Vendors That Obtain Data From MIPS Eligible Clinician's Certified EHR Technology

Currently, EHR-based systems are required to be considered certified EHR technology for multiple CMS quality programs. The Office of the National Coordinator for Health Information Technology (ONC) certification process has established standards and other criteria for structured data that EHRs must use. We propose to maintain this standard and require EHR-based data submission (whether transmitted directly from the EHR or from a data intermediary) to be certified EHR technology to submit quality measures, advancing care information, and CPIA data for MIPS. In addition, we propose at § 414.1400(a)(4) that health IT vendors that obtain data from a MIPS eligible clinician's certified EHR technology, like other third party intermediaries, would have to meet all requirements designated by CMS as a condition of their qualification or approval to participate in MIPS as a third party intermediary. This includes submitting data in the form and manner specified by CMS as proposed at § 414.1400(a)(4)(ii). We anticipate that for the initial years of MIPS the form and manner requirements will be similar to what was used in the PQRS program however, at a minimum these will be modified to address the four performance categories under MIPS and MIPS data calculation needs. As we gain experience under MIPS we anticipate that these form and manner requirements may change in future years to ease reporting burden. Historical form and manner requirements under the PQRS program are available here: https://www.qualitynet.org/imageserver/pqrs/registry2015/index.htm or https://www.cms.gov/Regulations-and-Guidance/Legislation/EHRIncentivePrograms/Downloads/QRDA_2016_CMS_IG.pdf . In addition, health IT vendors must comply with our QRDA Implementation Guides if submitting data from a certified EHR technology, which we anticipate will be similar to the one noted above. We anticipate providing further subregulatory guidance that would identify the certified EHR technology data formats that providers must submit. In addition, we propose at § 414.1325(b)(2) and (c)(2) to allow individual MIPS eligible clinicians and groups to submit data using certified EHR technology for the quality, CPIA, or advancing care information performance categories.

Although section 1848(q)(5)(B)(ii)(I) of the Act specifically requires the Secretary to encourage MIPS eligible clinicians to report on applicable measures using EHR technology with respect to the quality performance category, the statute does not specifically address allowing a third party intermediary—such as a health IT vendor to submit on a MIPS eligible clinician's behalf for the other performance categories. Although we could limit the usage of health IT vendors assessing the quality performance category under MIPS, we believe it would be less burdensome for MIPS eligible clinicians if we expand the health IT vendors' capabilities. By allowing health IT vendors to report on the quality, advancing care information, and CPIA performance categories we would alleviate the need for individual MIPS eligible clinicians and groups to use a separate mechanism to report data for these performance categories. Our intention is to encourage health IT vendors to design systems to be able to accept new types of EHR data (for example, CPIA and advancing care information) from MIPS eligible clinicians and groups—this would be in addition to the quality measure data that we already can accept. Therefore, we are proposing at § 414.1400(a)(2) to expand health IT vendors' capabilities by allowing health IT vendors to submit data on measures, activities, or objectives for any of the following MIPS performance categories:

(i) Quality;

(ii) CPIA; or

(iii) Advancing care information.

As proposed at § 414.1400(a)(1), health IT vendors submitting data on behalf of a MIPS eligible clinician or group would be required to obtain data from the MIPS eligible clinician's certified EHR technology. We believe this approach would permit a single health IT vendor to report on quality, advancing care information, and CPIA performance category requirements for MIPS and should mitigate the risks, costs, and burden of MIPS eligible clinicians having to report multiple times to meet the requirements of MIPS.

Health IT Vendors Data Requirements

We further propose that health IT vendors must be able to do the following:

• For measures, activities, and objectives under the quality, advancing care information, and CPIA performance categories, and as proposed at § 414.1400(a)(4)(i); if the data is derived from certified EHR technology, the health IT vendor must be able to indicate this data source.
• Either transmit data from the certified EHR technology or through a data intermediary in the CMS-specified form and manner, or have the ability for the individual MIPS eligible clinician and group to be able to submit data directly from their certified EHR technology, in the CMS-specified form and manner.

For MIPS eligible clinicians who choose to electronically submit quality, advancing care information, and CPIA data extracted from their certified EHR technology to an intermediary, the intermediary would then submit the measure and activity data to CMS in a CMS-specified form and manner on the MIPS eligible clinician's behalf for the respective performance period. In addition to meeting the appropriate data submission criteria for the quality, advancing care information, and CPIA performance categories for the MIPS EHR submission mechanism, MIPS eligible clinicians who choose the EHR submission mechanism would be required to have certified EHR technology meeting the proposed definition at § 414.1305. We request comments on these proposals.

c. Qualified Registries

We propose to define a qualified registry at § 414.1305 as a medical registry, a maintenance of certification program operated by a specialty body of the American Board of Medical Specialties or other data intermediary that, with respect to a particular performance period, has self-nominated and successfully completed a vetting process (as specified by CMS) to demonstrate its compliance with the MIPS qualification requirements specified by CMS for that performance period. The registry must have the requisite legal authority to submit MIPS data (as specified by CMS) on behalf of a MIPS eligible clinician or group to CMS. In addition, we are proposing at § 414.1400(a)(2) to expand a qualified registry's capabilities by allowing qualified registries to submit data on measures, activities, or objectives for any of the following MIPS performance categories:

(i) Quality;

(ii) CPIA; or

(iii) Advancing care information, if the MIPS eligible clinician or group is using certified EHR technology.

(1) Establishment of an Entity Seeking To Qualify as a Registry

We propose at § 414.1400(h) that in order for an entity to become qualified for a given performance period as a qualified registry, the entity must be in existence as of January 1 of the performance period for which the entity seeks to become a qualified registry (for example, January 1, 2017, to be eligible to participate for purposes of performance periods beginning in 2017). The qualified registry must have at least 25 participants by January 1 of the performance period. These participants do not necessarily need to be using the qualified registry to report MIPS data to us; rather, they need to be submitting data to the qualified registry for quality improvement. We also propose a qualified registry must provide attestation statements from the qualified registry/MIPS eligible clinicians during the data submission period that all of the data (quality measures, CPIAs, and advancing care information measures and objectives, if applicable) and results are accurate and complete.

(2) Self-Nomination Period

For the 2017 performance period, we propose at § 414.1400(g) a self-nomination period from November 15, 2016 until January 15, 2017. For future years of the program, starting with the 2018 performance period, we propose to establish the self-nomination period from September 1 of the prior year until November 1 of the year in which the qualified registry seeks to be qualified. Entities that desire to qualify as a qualified registry for purposes of MIPS for a given performance period would need to provide all requested information to CMS at the time of self-nomination and would need to self-nominate for that performance period. Having qualified as a qualified registry does not automatically qualify the entity to participate in subsequent MIPS performance periods. For example, a qualified registry may choose not to continue participation in the program in future years, OR the qualified registry may be precluded from participation in a future year, due to multiple data or submission errors as noted below. As such, CMS believes an annual self-nomination process is the best process to ensure accurate information is conveyed to MIPS eligible clinicians and accurate data is submitted to MIPS.

We propose to require further information (described below) of qualified registries at the time of self-nomination. If an entity becomes qualified as a qualified registry, they will need to sign a statement confirming this information is correct prior to us listing their qualifications on their Web site. Once we post the qualified registry on our Web site, including the services offered by the qualified registry, we will require the qualified registry to support these services/measures for its clients as a condition of the entity's qualification as a qualified registry for purposes of MIPS. Failure to do so will preclude the qualified registry from participation in MIPS in the subsequent performance year.

(3) Information Required at the Time of Self-Nomination

We propose that a qualified registry must provide the following information to us at the time of self-nomination:

• Organization Name (Specify Sponsoring Organization name and software vendor name if the two are different. For example, a specialty society in collaboration with a software vendor).
• MIPS performance categories (that is, categories for which the entity is self-nominating to report. For example, quality measures, advancing care information, and/or CPIA).
• Performance Period.
• Vendor Type (for example, qualified registry).
• Provide the method(s) by which the entity obtains data from its customers for each performance category for which it is approved: Claims; web-based tool; practice management system; certified EHR technology; other (please explain). If a combination of methods (Claims, web-based tool, Practice Management System, certified EHR technology, and/or other) is utilized, please state which method(s) the entity utilizes to collect data (performance numerator and denominator).
• Indicate the method the entity will use to verify the accuracy of each TIN/NPI and/or TIN it is intending to submit (for example; National Plan and Provider Enumeration System (NPPES), CMS claims, tax documentation).
• Describe the method the entity will use to accurately calculate performance rates for quality measures based on the appropriate measure type and specification. For composite measures or measures with multiple performance rates, the entity must provide us with the methodology the entity uses to calculate these composite measures and measures with multiple performance rates. The entity should be able to report to us a calculated composite measure rate, if applicable.
• Describe the method that the entity will use to accurately calculate performance data for CPIA and advancing care information performance categories based on the appropriate parameters or activities.
• Describe the process that the entity will use for completion of a randomized audit of a subset of data prior to the submission to us (for all performance categories the qualified registry is submitting data on; that is, quality, CPIA, and advancing care information, as applicable). Periodic examinations may be completed to compare patient record data with submitted data and/or ensure MIPS quality measures or other performance category (CPIA and advancing care information) activities, measures, or objectives were accurately reported and performance calculated based on the appropriate measure specifications (that is, accuracy of numerator, denominator, and exclusion criteria) or performance category requirements.
• Provide information on the entity's process for data validation for both individual MIPS eligible clinicians and groups within a data validation plan. For example, for individuals, it is encouraged that 3 percent of the MIPS eligible clinicians submitted to CMS by the qualified registry be sampled with a minimum sample of 10 TIN/NPIs or a maximum sample of 50 MIPS eligible clinicians. For each MIPS eligible clinician sampled, it is encouraged that 25 percent of the MIPS eligible clinicians' patients (with a minimum sample of five patients or a maximum sample of 50 patients) should be reviewed for all measures applicable to the patient.
• Provide the results of the executed data validation plan by May 31st of the year following the performance period. If the results indicate the qualified registry's validation reveals inaccuracy or low compliance provide to us an improvement plan. Failure to implement improvements may result in the qualified registry being placed in a probationary status or disqualification from future participation.

(4) Qualified Registry Requirements for Data Submission

Further, we propose that a qualified registry must perform the following functions:

• For measures, activities, and objectives under the quality, advancing care information, and CPIA performance categories and as proposed at § 414.1400(a)(4)(i); if the data is derived from certified EHR technology, the qualified registry must be able to indicate this data source.
• A qualified registry submitting MIPS quality measures that are risk-adjusted (and have the risk-adjusted variables and methodology listed in the measure specifications) must submit the risk-adjusted measure results to CMS when submitting the data for these measures.
• Submit to us, quality measures and activities data on all patients, not just Medicare patients.
• Submit quality measures, advancing care information, or CPIA performance categories data and results to us in the applicable MIPS performance categories for which the qualified registry is providing data.
• Provide timely feedback, at least 4 times a year, on all of the MIPS performance categories that the qualified registry will report to us. That is, if the qualified registry will be reporting on data for the CPIA, advancing care information, or quality performance category, all results as of the feedback report date should be included in the information sent to the MIPS eligible clinician. The feedback should be given to the individual MIPS eligible clinician or group (if participating as a group) at the individual participant level or group level, as applicable, for which the qualified registry reports. The qualified registry is only required to provide feedback based on the MIPS eligible clinician's data that is available at the time the feedback report is generated.
• A qualified registry must comply with any request by us to review the data submitted by the qualified registry for purposes of MIPS in accordance with applicable law. Specifically, data requested would be limited to the minimum necessary for us to carry out, for example, health care operations or health oversight activities.
• Mandatory participation in ongoing support conference calls hosted by us (approximately one call per month), including an in-person qualified registry kick-off meeting (if held) at our headquarters in Baltimore, MD. More than one unexcused absence could result in the qualified registry being precluded from participation in the program for that year. If a qualified registry is precluded from participation in MIPS, the individual MIPS eligible clinician or group would need to find another entity to submit their MIPS data.
• Agree that data inaccuracies including (but not limited to) TIN/NPI mismatches, formatting issues, calculation errors, data audit discrepancies affecting in excess of 3 percent of the total number of MIPS eligible clinicians submitted by the qualified registry may result in notations on our qualified registry posting of low data quality and would place the qualified registry on probation (if they decide to self-nominate for the next program year). If the qualified registry does not reduce their data error rate below 3 percent in the subsequent year, they would continue to be on probation and have their listing on the CMS Web site continue to note the poor quality of the data they are submitting for MIPS. Data errors affecting in excess of 5 percent of the MIPS eligible clinicians submitted by the qualified registry may lead to the disqualification of the qualified registry from participation in the following year's program. As we gain additional experience with qualified registries, we intend to revisit and enhance these thresholds in future years.
• Be able to report at least six quality measures including one cross-cutting measure and one outcome measure. If an outcome measure is not available, be able to report another high priority measure (appropriate use, patient safety, efficiency, patient experience, and care coordination measures).
• Enter into and maintain with its participating clinicians an appropriate Business Associate agreement that provides for the qualified registry's receipt of patient-specific data from an individual MIPS eligible clinician or group, as well as the qualified registry's disclosure of quality measure results and numerator and denominator data and/or patient specific data on Medicare and non-Medicare beneficiaries on behalf of MIPS eligible clinicians or group.
• Obtain and keep on file signed documentation that each holder of an NPI whose data are submitted to the qualified registry, has authorized the qualified registry to submit quality measure results, CPIA measure and activity results, advancing care information objective results and numerator and denominator data and/or patient-specific data on Medicare and non-Medicare beneficiaries to us for the purpose of MIPS participation. This documentation must be obtained at the time the MIPS eligible clinician or group signs up with the qualified registry to submit MIPS data to the qualified registry and must meet any applicable laws, regulations, and contractual business associate agreements. Groups participating in MIPS via a qualified registry may have their group's duly authorized representative grant permission to the qualified registry to submit their data to us. If submitting as a group each individual MIPS eligible clinician does not need to grant their individual permission to the qualified registry to submit their data to us.
• Not be owned and managed by an individual locally-owned single specialty group (for example, single specialty practices with only one practice location or solo practitioner practices are prohibited from self-nominating to become a MIPS qualified registry).
• Be able to separate out and report on all payers, including Medicare Part B FFS patients and non-Medicare patients.
• Provide the measure numbers for the MIPS quality measures on which the qualified registry is reporting.
• Provide the measure title (and specialty-specific measure set title, if applicable) for the MIPS quality measures and CPIAs (if applicable) on which the qualified registry is reporting.
• Indicate if the qualified registry will be reporting the advancing care information component measures and objectives.
• Report the number of eligible instances (reporting denominator).
• Report the number of instances a quality service is performed (performance numerator).
• Report the number of performance exclusions, meaning the quality action was not performed for a valid reason as defined by the measure specification.
• Comply with a CMS-specified secure method for data submission, such as submitting the qualified registry's data in an XML file.
• Sign a document verifying the qualified registry's name, contact information, cost for MIPS eligible clinicians or groups to use the qualified registry, services provided, and the specialty-specific measure sets the qualified registry intends to report. Once posted on the qualified registry's CMS Web site, the qualified registry will need to support the measures/measure sets confirmed by the qualified registry. Failure to do so will may preclude the qualified registry from participation in MIPS in the subsequent year.
• Must provide attestation statements during the data submission period that all of the data (quality measures, CPIAs, and advancing care information measures and objectives, if applicable) and results are accurate and complete.
• For purposes of distributing feedback reports to MIPS eligible clinicians, collect a MIPS eligible clinician's email address(es) and have documentation from the MIPS eligible clinician authorizing the release of his or her email address.
• Be able to calculate and submit measure-level reporting rates or, upon request, the data elements needed to calculate the reporting and performance rates by TIN/NPI and/or TIN.
• Be able to calculate and submit, by TIN/NPI and/or TIN, a performance rate (that is the percentage of a defined population who receive a particular process of care or achieves a particular outcome based on a calculation of the measures' numerator and denominator specifications) for each measure on which the TIN/NPI and/or TIN reports or, upon request the Medicare and non-Medicare level data elements needed to calculate the performance rates.
• Provide the performance period start date the qualified registry will cover.
• Provide the performance period end date the qualified registry will cover.
• Report the number of instances in which the applicable submission criteria were not met, for example, the quality measure was not reported and a performance exclusion did not apply.
• For data validation purposes, provide information on the entity's sampling methodology. For example, if is encouraged that 3 percent of the MIPS eligible clinicians be sampled with a minimum sample of 10 MIPS eligible clinicians or a maximum sample of 50 MIPS eligible clinicians. For each MIPS eligible clinician sampled, it is encouraged that 25 percent of the MIPS eligible clinicians' patients (with a minimum sample of five patients or a maximum sample of 50 patients) should be reviewed for all measures applicable to the patient.

We request comments on these proposals.

d. CMS-Approved Survey Vendors

As discussed in the section II.E.5.b. we propose to allow groups to report CAHPS for MIPS survey measures. We propose the data collected on the CAHPS for MIPS survey measures would be transmitted to us via a CMS-approved survey vendor.

For purposes of MIPS, we propose to define a CMS-approved survey vendor at § 414.1305 as a survey vendor that is approved by CMS for a particular performance period to administer the CAHPS for MIPS survey and transmit survey measures data to CMS. We propose at § 414.1400(i) that vendors are required to undergo the CMS approval process for each year in which the survey vendor seeks to transmit survey measures data to CMS. We anticipate retaining the same policies and procedures we currently follow for a CMS-approved survey vendor for PQRS and apply them to a MIPS CMS-approved survey vendor. We propose the following requirements for a CMS-approved survey vendor for the CAHPS for MIPS survey. A CMS-approved survey vendor for CAHPS for MIPS must:

(1) Comply with and complete the Vendor Participation Form—We anticipate retaining the same application process and Vendor Participation Form that was required for the CAHPS for PQRS survey. Please refer to http://www.pqrscahps.org/en/participation-form/ for further details. Therefore, we are proposing at § 414.1400(i) that all CMS-approved survey vendor applications and materials will be due April 30 of the performance period. However, we do seek comments on whether the deadline for CMS-approved survey vendor applications and materials should be earlier, such as prior to the beginning of the performance period. In addition, we propose the following items will be required for your organization to be a CMS-approved survey vendor of the CAHPS for MIPS Survey:

• Meet all of the Minimum Survey Vendor Business Requirements at the time of the submission of the Vendor Participation Form; and
• Complete the Vendor Participation Form.

(2) Comply with the Minimum Survey Vendor Business Requirements—We anticipate retaining the same minimum survey business requirements that were required for the CAHPS for PQRS survey. Please refer to http://www.pqrscahps.org/en/business-requirements/ for further details. We propose Applicant Organizations (survey vendor and subcontractors) must possess all required facilities and systems to implement the CAHPS for MIPS Survey. Subcontractors will be subject to the same requirements as the applicant vendor. Organizations that are approved to administer the CAHPS for MIPS Survey must conduct all their CAHPS for MIPS business operations within the United States. This requirement applies to all staff and subcontractors. In addition, we propose to request information regarding:

• Relevant organization and survey experience.
• Survey capability and capacity.
• Adherence to quality assurance guidelines and participation in quality assurance activities.
• Documentation requirements.
• Adhere to all protocols and specifications, and agree to participate in training sessions.

Specifically, to obtain our approval, we propose that survey vendors would be required to undergo training, meet our standards on how to administer the survey, and submit a quality assurance plan. We would provide the identified survey vendor with an appropriate sample frame of beneficiaries from each group that has contracted with the survey vendor and elected to participate in the CAHPS for MIPS survey. The survey vendor would also be required to administer the survey according to established protocols to ensure valid and reliable results. More information on quality assurance and protocols can be reviewed at http://www.pqrscahps.org/en/quality-assurance-guidelines/ . CMS-approved survey vendors would be supplied with mail and telephone versions of the survey in electronic form, and text for beneficiary pre-notification and cover letters. CAHPS for MIPS surveys can be administered in English, Spanish, Cantonese, Mandarin, Korean, Russian and/or Vietnamese. Survey vendors would be required to use appropriate quality control, encryption, security and backup procedures to maintain survey response data. The data would then be securely sent back to us for scoring and/or validation in accordance with applicable law. To ensure that a survey vendor possesses the ability to transmit survey measures data for a particular performance period, we propose to require survey vendors to undergo this approval process for each year in which the survey vendor seeks to transmit survey measures data to us. We request comments on this proposal.

e. Probation and Disqualification of a Third Party Intermediary

We propose at § 414.1400(k) a process for placing third party intermediaries on probation and for disqualifying such entities for failure to meet certain standards established by CMS. Specifically, we propose that if at any time we determine that a third party intermediary (that is, a QCDR, health IT vendor, qualified registry, or CMS-approved survey vendor) has not met all of the applicable requirements for qualification, we may place the third party intermediary on probation for the current performance period and/or the following performance period, as applicable.

In addition, we propose CMS requires a corrective action plan from the third party intermediary to address any deficiencies or issues and prevent them from recurring. We propose the corrective action plan must be received and accepted by CMS within 14 days of the CMS notification to the third party intermediary of the deficiencies or probation. Failure to comply with this would lead to disqualification from MIPS for the subsequent performance period.

We propose probation to mean that, for the applicable performance period, the third party intermediary would not be allowed to miss any meetings or deadlines and would need to submit a corrective action plan for remediation or correction of deficiencies identified that resulted in the probation.

In addition, we propose that if the third party intermediary has data inaccuracies including (but not limited to) TIN/NPI mismatches, formatting issues, calculation errors, data audit discrepancies affecting in excess of 3 percent (but less than 5 percent) of the total number of MIPS eligible clinicians or groups submitted by the third party intermediary, CMS would annotate on the CMS qualified posting that the third party intermediary furnished data of poor quality and would place the entity on probation for the subsequent MIPS performance period with the opportunity to go on probation for a year to correct their deficiencies.

Further, we propose if the third party intermediary does not reduce their data error rate below 3 percent for the subsequent performance period, the third party intermediary would continue to be on probation and have their listing on the CMS Web site continue to note the poor quality of the data they are submitting for MIPS for one additional performance year. After two years on probation, the third party intermediary would be disqualified for the subsequent performance year. Data errors affecting in excess of 5 percent of the MIPS eligible clinicians or groups submitted by the third party intermediary may lead to the disqualification of the third party intermediary from participation for the following performance period. In placing the third party intermediary on probation; we would notify the third party intermediary of the identified issues, at the time of discovery of such issues.

Finally, we propose if the third party intermediary does not submit an acceptable corrective action plan within 14 days of notification of the deficiencies and correct the deficiencies within 30 days or before the submission deadline—whichever is sooner, we may disqualify the third party intermediary from participating in MIPS for the current performance period and/or the following performance period, as applicable. We request comments on these proposals.

(f) Auditing of Third Party Intermediaries Submitting MIPS Data

We propose at § 414.1400(j) that any third party intermediary (that is, a QCDR, health IT vendor, qualified registry, or CMS-approved survey vendor) must comply with certain auditing requirements as a condition of their qualification or approval to participate in MIPS as a third party intermediary. Specifically, we propose the entity must make available to CMS the contact information of each MIPS eligible clinician or group on behalf of whom it submits data. The contact information will include, at a minimum, the MIPS eligible clinician or group's practice phone number, address, and, if available, email. Further, we propose the entity must retain all data submitted to CMS for MIPS for a minimum of 10 years. We request comments on this proposal.

10. Public Reporting on Physician Compare

This section contains the proposed approach for publicly reporting on Physician Compare for the MIPS, APM, and other information as required by the MACRA.

Physician Compare draws its operating authority from section 10331(a)(1) of the Affordable Care Act. As required, by January 1, 2011, we developed a Physician Compare Internet Web site with information on physicians enrolled in the Medicare program under section 1866(j) of the Act, as well as information on other EPs who participate in the PQRS under section 1848 of the Act. More information on Physician Compare can be accessed on the Physician Compare Initiative Web site at https://www.cms.gov/medicare/quality-initiatives-patient-assessment-instruments/physician-compare-initiative/ .

The first phase of Physician Compare was launched on December 30, 2010 ( http://www.medicare.gov/physiciancompare ). Since the initial launch, Physician Compare has been continually improved and more information has been added. Currently, Web site users can view information about approved Medicare professionals, such as name, Medicare primary and secondary specialties, practice locations, group affiliations, hospital affiliations that link to the hospital's profile on Hospital Compare as available, Medicare Assignment status, education, residency, and American Board of Medical Specialties (ABMS) board certification information. For group practices, users can view group practice names, specialties, practice locations, Medicare assignment status, and affiliated professionals. In addition, Medicare professionals and group practices that satisfactorily or successfully participated in a CMS quality program have a green check mark on their profile page to indicate their commitment to quality.

Consistent with section 10331(a)(2) of the Affordable Care Act, Physician Compare also phased in public reporting of information on physician performance that provides comparable information on quality and patient experience measures for reporting periods beginning January 1, 2012. To the extent that scientifically sound measures are developed and are available, Physician Compare is required to include, to the extent practicable, the following types of measures for public reporting: Measures collected under PQRS and an assessment of efficiency, patient health outcomes, and patient experience, as specified. The first set of quality measures were publicly reported on Physician Compare in February 2014. Currently, Physician Compare publicly reports 14 group practice level measures collected through the Web Interface for groups of 25 or more EPs participating in 2014 under the PQRS and for ACOs participating in the Shared Savings Program or Pioneer ACO program, and six individual level measures collected through claims for individual EPs participating in 2014 under the PQRS. A complete history of public reporting on Physician Compare is detailed in the CY 2016 PFS final rule (80 FR 71117-22).

As finalized in the CY 2015 and CY 2016 PFS final rules (79 FR 67547 and 80 FR 70885) Physician Compare will expand public reporting over the next several years. This expansion includes publicly reporting both individual EP and group practice level QCDR measures starting with 2015 individual EP measures to be publicly reported on Physician Compare in late 2016, and expanding to group practice QCDR measures in late 2017 (80 FR 71125), which is consistent with section 101(d)(1)(B) of the MACRA.

Section 1848(q)(9)(A) and (D) of the Act facilitates the continuation of the phased approach to public reporting by requiring the Secretary to make available on the Physician Compare Web site, in an easily understandable format, individual MIPS eligible clinician and groups performance information, including:

• The MIPS eligible clinician's CPS;
• The MIPS eligible clinician's performance under each MIPS performance category (quality, resource use, CPIA and advancing care information);
• Names of eligible clinician's in Advanced APMs and, to the extent feasible, the names of such Advanced APMs and the performance of such models; and
• Periodically post aggregate information on the MIPS, including the range of composite scores for all MIPS eligible clinician's and the range of the performance of all MIPS eligible clinician's with respect to each performance category.

Section 1848(q)(9)(B) of the Act also requires that this information indicate, where appropriate, that publicized information may not be representative of the eligible clinician's entire patient population, the variety of services furnished by the eligible clinician, or the health conditions of individuals treated. In order to ensure the information mandated under section 1848(q)(9) of the Act are publicly reported, the information must be in compliance with the existing mandate and regulations previously established under section 10331(a)(2) and 10331(b) of the Affordable Care Act. As required under section 10331(a)(2) of the Affordable Care Act, all measure data included on Physician Compare must be comparable. In addition, section 10331(b) of the Affordable Care Act requires that we include, to the extent practicable, processes to ensure that data made public are statistically valid, reliable, and accurate, including risk adjustment mechanisms used by the Secretary. In addition to the Affordable Care Act informed public reporting standards—statistically valid and reliable data, that are accurate and comparable—existing regulation notes that all the data must also prove through consumer testing to resonate with and be accurately interpreted by consumers in order to be included on Physician Compare profile pages. Together, we refer to these conditions as the Physician Compare public reporting standards (80 FR 71118-20). Section 10331(d) of the Affordable Care Act also requires us to consider input from multi-stakeholder groups, consistent with sections 1890(b)(7) and 1890A of the Act. We also continue to receive general input from stakeholders on Physician Compare through a variety of means, including rulemaking and different forms of stakeholder outreach (for example, Town Hall meetings, Open Door Forums, webinars, education and outreach, Technical Expert Panels, etc.).

In addition, section 1848(q)(9)(C) of the Act requires the Secretary to provide an opportunity for MIPS eligible clinicians to review the information that will be publicly reported prior to such information being made public. This is generally consistent with section 10331(a)(2) of the Affordable Care Act and current regulations that established a 30-day preview period for all measurement performance data that will allow physicians and other EPs to view their data as it will appear on the Web site in advance of publication on Physician Compare (80 FR 71120). Section 1848(q)(9)(C) of the Act also requires that MIPS eligible clinicians be able to submit corrections for the information to be made public. We propose that this extension of the current Physician Compare 30-day preview period will be implemented starting with data from the 2017 MIPS performance period. We propose a 30-day preview period in advance of the publication of any data on Physician Compare. We will coordinate efforts between Physician Compare and the four components of MIPS in terms of data review and appeal and any relevant data resubmission or correction. All data available for public reporting—measure rates, scores, and/or attestations—will be available for review and correction during the targeted review process (see section II.E.8.c. of this proposed rule). The process will begin at least 30 days in advance of the publication of new data. Data under appeal and review will not be publicly reported until the review is complete. All corrected measure rates, scores, and/or attestations submitted will be available for public reporting. The technical details of the process will be communicated directly to affected MIPS eligible clinicians and groups and detailed outside of rulemaking.

As with the current process, the details will be made public on the Physician Compare Initiative page on cms.gov and communicated through Physician Compare and other CMS listservs.

In addition, section 1848(q)(9)(D) of the Act requires that aggregate information on the MIPS be periodically posted on the Physician Compare Web site; including the range of composite scores for all MIPS eligible clinicians and the range of performance for all MIPS eligible clinicians with respect to each performance category.

Lastly, section 104 of the MACRA requires the Secretary to make publicly available, on an annual basis (beginning with 2015), in an easily understandable format, information with respect to physicians and other eligible clinician's on items and services furnished to Medicare beneficiaries, and to include, at a minimum:

• Information on the number of services furnished under Part B, which may include information on the most frequent services furnished or groupings of services;
• Information on submitted charges and payments for Part B services; and
• A unique identifier for the physician or other eligible clinician that is available to the public, such as an NPI.

The information would further be required to be made searchable by at least specialty or type of physician or other eligible clinician; characteristics of the services furnished (such as, volume or groupings of services); and the location of the physician or other eligible clinician.

Therefore, at § 414.1395(a) we propose public reporting of an eligible clinician's MIPS data; in that for each program year, we would post on a public Web site, in an easily understandable format, information regarding the performance of MIPS eligible clinicians or groups under the MIPS.

Furthermore, in accordance with section 104(e) of the MACRA, we finalized in the CY 2016 PFS final rule (80 FR 71130) to add utilization data to the Physician Compare downloadable database. Utilization data is currently available at http://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/Medicare-Provider-Charge-Data/Physician-and-Other-Supplier.html . As finalized (80 FR 71130), this information will be integrated on the Physician Compare Web site via the downloadable database targeted for late 2016. Not all available data will be included. The specific HCPCS codes included will be determined based on analysis of the available data, focusing on the most used codes. Additional details about the specific HCPCS codes that will be included in the downloadable database will be provided to stakeholders in advance of data publication. And, all data available for public reporting—on the consumer-facing Web site pages or in the downloadable database—will be available for preview during the 30-day preview period.

We believe section 10331 of the Affordable Care Act supports our overarching goals of the MACRA by providing consumers with quality information that will help them make informed decisions about their health care, while encouraging clinicians to improve the quality of care they provide to their patients. In accordance with section 10331 of the Affordable Care Act, section 1848(q)(9) of the Act, and section 104(e) of the MACRA, we plan to continue to publicly report performance information on Physician Compare. As a result, we propose inclusion of the following information on Physician Compare.

a. Composite Score, Performance Categories, and Aggregate Information

As noted, section 1848(q)(9)(A) and (D) of the Act requires that we publicly report on Physician Compare the composite score for each MIPS eligible clinician, performance of each MIPS eligible clinician for each performance category, and periodically post aggregate information on the MIPS, including the range of composite scores for all MIPS eligible clinicians and the range of performance of all the MIPS eligible clinicians for each performance category. We propose that these data, to the extent that they meet the previously established public reporting standards, will be added to Physician Compare for each MIPS eligible clinician or group, either on the profile pages or in the downloadable database, as technically feasible. Statistical testing and consumer testing, as well as consultation of the Physician Compare Technical Expert Panel (TEP), will determine how and where these data are reported on Physician Compare. We request comments on these proposals.

In addition, we seek comment on the advisability and technical feasibility of including data voluntarily reported by EPs and groups that are not subject to MIPS payment adjustments, such as those practicing through RHC, FQHCs, etc., on Physician Compare. Any regulatory changes would be made through separate notice-and-comment rulemaking.

b. Quality

The quality performance category is discussed in detail in section II.E.5.b. of this proposed rule. Consistent with the current policy that makes all current PQRS measures available for public reporting, we now propose to make all measures under the MIPS quality performance category (see section II.E.5.b. of this proposed rule) available for public reporting on Physician Compare. This includes all available measures reported via all available submission methods, and applies to both MIPS eligible clinicians and groups.

Also consistent with current policy, although all measures will be available for public reporting not all measures will be made available on the consumer-facing Web site profile pages. As explained in the CY 2016 PFS final rule (80 FR 71120), providing too much information can overwhelm consumers and lead to poor decision making. Therefore, we propose that all measures in the quality performance category that meet the public reporting standards would be included in the downloadable database, as technically feasible. We also propose that a subset of these measures would be publicly reported on the Web site's profile pages, as technically feasible. Statistical testing and consumer testing will determine how and where measures are reported on Physician Compare. In addition, we do not publicly report first year measures, meaning new measures that have been in use for less than 1 year, regardless of submission methods. After a measure's first year in use, we will evaluate the measure to see if and when the measure is suitable for pubic reporting (80 FR 71118).

Currently, there is a minimum sample size requirement of 20 patients for performance data to be included on the Web site. As part of the MIPS and APMs RFI we asked for comment on moving away from this requirement and moving to a reliability threshold for public reporting. In general, commenters supported a minimum reliability threshold. As a result, we are now proposing to institute a minimum reliability threshold for public reporting on Physician Compare.

The reliability of a measure refers to the extent to which the variation in measure is due to variation in quality of care as opposed to random variation due to sampling. Statistically, reliability depends on performance variation for a measure across entities, the random variation in performance for a measure within an entity's panel of attributed beneficiaries, and the number of beneficiaries attributed to the entity. High reliability for a measure suggests that comparisons of relative performance across entities, in this case groups or eligible clinicians, are likely to be stable and consistent, and that the performance of one entity on the quality measure can confidently be distinguished from another. Conducting analysis to determine reliability of the data collected will allow us to calculate the minimum reliability threshold for those data. Once an appropriate minimum reliability threshold is determined, the reporting of reporters' performance rates for a given measure can be restricted to only those meeting the minimum reliability threshold.

We propose to also include the total number of patients reported on per measure in the downloadable database to facilitate transparency and more accurate understanding and use of the data. We request comments on these proposals.

We also are seeking comment on the types of data that should be reported on Physician Compare as the MIPS program evolves, specifically in regard to the quality performance category. Any regulatory changes would be made in separate notice-and-comment rulemaking.

c. Resource Use

The resource use performance category is detailed in section II.E.5.e. of this proposed rule. We propose to make all measures under the MIPS resource use performance category (see section II.E.5.e. of this proposed rule) available for public reporting on Physician Compare. This includes all available measures reported via all available submission methods, and applies to both MIPS eligible clinicians and groups.

We have found that resource use data do not resonate with consumers and can instead lead to significant misinterpretation and misunderstanding. Therefore, we propose to include a sub-set of resource use measures, that meet the aforementioned public reporting standards, on Physician Compare, either on profile pages or in the downloadable database, if technically feasible. Statistical testing and consumer testing will determine how and where measures are reported on Physician Compare. In addition, we do not publicly report first year measures, meaning new measures that have been in use for less than 1 year, regardless of submission methods. After a measure's first year in use, we will evaluate the measure to see if and when the measure is suitable for pubic reporting (80 FR 71118). We request comments on these proposals.

We also are seeking comment on the types of data that should be reported on Physician Compare as the MIPS program evolves, specifically in regard to the resource use performance category. Any regulatory changes would be made in separate notice-and-comment rulemaking.

d. CPIA

The CPIA performance category is detailed in section II.E.5.f. of this proposed rule. We propose to make all activities under the MIPS CPIA performance category (see section II.E.5.f. of this proposed rule) available for public reporting on Physician Compare. This includes all available CPIAs reported via all available submission methods, and applies to both MIPS eligible clinicians and groups.

We propose to include a subset of CPIA data that meet the aforementioned public reporting standards, on Physician Compare, either on the profile pages or in the downloadable database, if technically feasible. For those eligible clinicians that successfully meet the CPIA performance category requirements this may be posted on Physician Compare as an indicator. The CPIA performance category is a new field of data for Physician Compare so concept and consumer testing will be needed to ensure these data are understood by consumers. Therefore, statistical testing and consumer testing will determine how and where CPIAs are reported on Physician Compare. In addition, since we do not publicly report first year measures, we are also applying this policy to CPIA, meaning new CPIAs that have been in use for less than 1 year, regardless of submission methods. After a CPIA's first year in use, we will evaluate the activity to see if and when the activity is suitable for pubic reporting (80 FR 71118). We request comments on these proposals.

We also are seeking comment on the types of data that should be reported on Physician Compare as the MIPS program evolves, specifically in regard to the CPIA performance category. Any regulatory changes would be made in separate notice-and-comment rulemaking.

e. Advancing Care Information

Since the beginning of the EHR Incentive Programs in 2011, participant performance data has been publically available in the form of public use files on the CMS Web site. In the 2015 EHR Incentive Programs final rule, we addressed comments requesting that CMS not only continue this practice but also include a wider range of information on participation and performance. In that rule, we stated our intent to publish the performance and participation data on Stage 3 objectives and measures of meaningful use in alignment with quality programs which utilize publicly available performance data such as Physician Compare (80 FR 62901). At this time there is only a green check mark on Physician Compare profile pages to indicate that an EP successfully participated in the current Medicare EHR Incentive Program for EPs.

As MIPS will now include advancing care information as one of the four MIPS performance categories, we are proposing to include more information on eligible clinician's performance on the objectives and measures of meaningful use on Physician Compare. An important consideration is that to meet the aforementioned public reporting standards, the data added to Physician Compare must resonate with the average Medicare consumer and their caregivers. Consumer testing to date has shown that people with Medicare value the use of certified EHR technology and see EHR use as something that if used well can improve the quality of their care. In addition, we believe the inclusion of indicators for providers who achieve high performance in key care coordination and patient engagement activities provide significant value for consumers.

We are therefore proposing to include an indicator for any eligible clinician or group who successfully meets the advancing care information performance category, as detailed in section II.E.5.g. of this proposed rule, as technically feasible on Physician Compare. Also as technically feasible, we are proposing to include additional indicators, including but not limited to, identifying if the eligible clinician or group scores high performance in patient access, care coordination and patient engagement, or health information exchange; as further specified in section II.E.5.g. of this proposed rule. To reiterate, any advancing care information objectives or measures must meet the public reporting standards to be posted on Physician Compare, either on the profile pages or in the downloadable database. This includes all available objectives or measures reported via all available submission methods, and applies to both MIPS eligible clinicians and groups. Statistical testing and consumer testing will determine how and where objectives and measures are reported on Physician Compare. In addition, we do not publicly report first year measures, meaning new measures that have been in use for reporting for less than 1 year, regardless of submission methods. After a measure's first year in use, we will evaluate the measure to see if and when the measure is suitable for pubic reporting (80 FR 71118). We request comment on these proposals.

We also are seeking comment on potentially including an indicator to show low performance in the advancing care information performance category, as well as, the types of data that should be reported on Physician Compare as the MIPS program evolves, specifically in regard to the advancing care information performance category. Additionally, we would need to perform consumer testing and evaluate the feasibility of potentially including an indicator to show low performance in the advancing care information performance category to ensure this is understood by consumers. Any regulatory changes would be made in separate notice-and-comment rulemaking.

f. Utilization Data

As discussed above, we previously finalized to begin to include utilization data in the Physician Compare downloadable database in late 2016 using the most currently available data (80 FR 71130) to meet section 104(e) of the MACRA. As there are thousands of Healthcare Common Procedure Coding System (HCPCS) codes in use, not all available data will be included. The specific HCPCS codes included will be determined based on analysis of the available data, focusing on the most used codes. The goal will be to include counts that can facilitate a greater understanding and more in-depth analysis of the other measure and performance data being made available. We propose to continue to include utilization data in the Physician Compare downloadable database. We request comment on this.

g. APM Data

As discussed above, section 1848(q)(9)(A)(ii) of the Act requires us to publicly report names of eligible clinicians in Advanced APMs and, to the extent feasible, the names and performance of Advanced APMs. We see this as an opportunity to continue and build on reporting we are now doing of ACO data on Physician Compare. At this time, if an EP or group submitted quality data as part of an ACO, there is an indicator on the EP's or group's profile page indicating this. In this way, it is known which EPs and groups took part in an ACO. Also, currently, all ACOs have a dedicated page on the Web site to showcase their data. If technically feasible, we propose to use this model as a guide as we add APM data to Physician Compare. We propose to indicate on eligible clinician and group profile pages when the eligible clinician or group is participating in an APM. We also propose to link eligible clinicians and groups to their APMs data, as relevant and possible, through Physician Compare. Data posting would be considered for both Advanced and non-eligible APMs.

At the outset, APMs will be very new concepts for consumers. Testing shows that at this time, ACOs are not a familiar concept to the average Medicare consumer. It is very easy for consumers to misunderstand an ACO as just a type of group. We expect at least the same lack of familiarity when introducing the broader concept of APM, of which ACOs comprise only one type. In these early years, indicating who participated in APMs and testing language to accurately explain that to consumers provides useful and valuable information as we continue to evolve Physician Compare. As we come to understand how to best explain this concept to consumers, we can continue to assess how to most fully integrate these data on the Web site. We request comment on these proposals.

F. Overview of Incentives for Participation in Advanced Alternative Payment Models

Section 1833(z) of the Act, as added by section 101(e)(2) of the MACRA, requires that an incentive payment be made to Qualifying APM Participants (QPs) for participation in eligible alternative payment models (referred to as Advanced APMs). Key statutory elements of the incentives for participation in Advanced APMs under the Quality Payment Program addressed in this proposed rule include:

• Beginning in 2019, if an eligible clinician participates in a certain type of APM (an Advanced APM), they may become a QP. Eligible clinicians who are QPs are excluded from the MIPS.
• For years from 2019 through 2024, QPs receive a lump sum incentive payment equal to 5 percent of their prior year's payments for Part B covered professional services, and beginning in 2026, QPs receive a higher update under the PFS than non-QPs.
• For 2019 and 2020, eligible clinicians may become QPs only through participation in Advanced APMs.
• For 2021 and later, eligible clinicians may become QPs through a combination of participation in Advanced APMs and APMs with other payers (Other Payer Advanced APMs).
• This section of the rule proposes the definitions, requirements, procedures, and thresholds of participation that will govern this program.

1. Policy Principles

Several core policy principles are derived from both the MACRA law and the Department's broad vision for better care, smarter spending, and healthier people. These principles drive many of our decisions in developing the overall framework for making APM Incentive Payments to QPs and for approaching interactions between MIPS and APMs found in this proposed rule. In addition to increasing the quality and efficiency of care delivered in the Medicare program and across the health system, these principles include the following seven goals:

• To the greatest extent possible, continue to build a portfolio of APMs that collectively allows participation for a broad range of physicians and other practitioners. We believe finding better ways to deliver care across settings and specialties can lead to improved health outcomes and more efficient health care spending. Doing this requires active CMS engagement with stakeholders, as well as input from those stakeholders to refine ideas in ways that meet statutory and delivery system reform goals.
• Design the program such that the APM Incentive Payment is attainable by increasing numbers of practitioners over time, yet remains reserved for those eligible clinicians participating in organizations that are truly engaged in care transformation. We believe the structure of the law is clear in that the APM Incentive Payments are earned through participation in APMs that are designed to be challenging and involve rigorous care improvement activities. In general, we believe eligible clinicians that receive incentives should be those who: Take on financial risk for potential losses under an APM; are accountable for performance based on meaningful quality metrics; and use certified EHR technology.
• Maximize participation in both Advanced APMs and other APMs. Although we want to maintain high standards for eligible clinicians to earn the APM Incentive Payment, we also want to enable and encourage high levels of participation in a broad range of APMs, including those that are not Advanced APMs. We believe participation in any APM offers eligible clinicians and beneficiaries significant benefits.
• Create policies that allow for flexibility in future innovative Advanced APMs. We do not want to constrain the robust development of new Advanced APMs by framing standards only in terms of today's APMs but rather in ways that allow many avenues for meeting the Advanced APM criteria.
• Support multi-payer models and participation in innovative models in Medicaid and commercial markets in order to promote high quality and efficient care across the health care market.
• Recognize that the APM Incentive Payment added by the MACRA primarily incentivizes participation in Advanced APMs that involve covered professional services under Medicare Part B. We believe the new provisions of section 1833(z) of the Act distinguish between participation in Advanced APMs that involve Medicare Part B covered professional services and participation in Other Payer Advanced APMs, which could include those sponsored by Medicare Advantage organizations. The Quality Payment Program has the potential to influence a wide range of payment arrangements, such as those under Medicare Advantage, but there is a clear distinction between Medicare Part B and all other payers in how calculations are performed for QP determinations and the APM Incentive Payment. Through the all-payer route to the APM Incentive Payment, we hope to encourage cooperation across payers and create demand for arrangements that, like Advanced APMs, meaningfully incorporate financial risk, quality measure performance, and use of certified EHR technology as strategies for improving care outcomes.
• Minimize burden on organizations and professionals. Between APM participation and MIPS reporting, we hope to coordinate administrative processes, minimize overall reporting burden, and make transitioning between being a QP and being subject to MIPS as seamless as possible.
• We do not intend to create additional performance assessments or audits beyond those specified under an APM. Rather, we believe the process for determining whether an eligible clinician receives the APM Incentive Payment should focus on the relative degree of participation by eligible clinicians in Advanced APMs, not on their performance within the APM. The Quality Payment Program does not alter how each particular APM measures and rewards success within its design. Rather, it rewards a substantial degree of participation in certain APMs.

2. Overview of Proposed APM Policies

The incentives for Advanced APM participation established by MACRA includes several sets of related requirements that must be met. Three distinct roles play important parts in the program structure: (1) The Advanced Alternative Payment Model (Advanced APM), which is a health care payment and/or delivery model that includes payment arrangements and other design elements as part of a particular approach to care improvement; (2) the Advanced APM Entity, which is the entity participating in the Advanced APM and which meets criteria established under section 1833(z) of the Act; and (3) the eligible clinician, who is the individual physician or practitioner, or group of physicians or practitioners, who is a participant of the Advanced APM Entity and may be determined to be a QP.

In this rule we are proposing a series of steps that result in the determination of certain eligible clinicians as QPs for a particular year (the payment year). QPs would receive the APM Incentive Payment as specified in section 1833(z) of the Act for each of the years they qualify from 2019 through 2024, and the differential update incentive in section 1848(d)(20) of the Act for each of the years they qualify beginning in 2026. Per section 1833(z)(1)(A) of the Act, the APM Incentive Payment that an eligible clinician receives as a QP for a year between 2019 and 2024 is a lump sum payment equal to 5 percent of the QP's estimated aggregate payments for Medicare Part B covered professional services (services paid under or based on the Medicare PFS) for the prior year. Eligible clinicians who are QPs for a year are also excluded from MIPS for that year. In addition, beginning in 2026, QPs receive a higher Medicare PFS update (the “qualifying APM conversion factor”) than non-QPs. This QP determination is made for one calendar year at a time.

The proposed steps that would result in a QP determination can be summarized as follows: (1) We determine whether the design of an APM meets three specified criteria for it to be deemed an Advanced APM; (2) an entity (the Advanced APM Entity) with a group of individual eligible clinicians participates in the Advanced APM; (3) we determine whether, during a performance period (the QP Performance Period), the eligible clinicians in the Advanced APM Entity collectively have at least a specified percentage of their aggregate Medicare Part B payments for covered professional services, or patients who received covered professional services, through the Advanced APM; (4) all of the eligible clinicians in the Advanced APM Entity are designated QPs for the payment year associated with that QP Performance Period. Those QPs would receive the 5 percent lump-sum APM Incentive Payments mentioned above for the payment year. This QP determination process would occur each year following the QP Performance Period, with the first payment year being 2019. In section II.F.5.a, we propose that the QP Performance Period will be the calendar year 2 years prior to the payment year.

Under the MACRA, for payment years 2019 and 2020, QP determinations must be based only on payments or patients under Medicare Part B (the Medicare payment threshold option, which we refer to as the “Medicare Option”). Beginning in payment year 2021—which according to our proposal would be based on 2019 calendar year data—there would be an additional option for eligible clinicians to become QPs through a combination of their participation in Advanced APMs and similar payment arrangements with other payers (Other Payer Advanced APMs). This option is the combination all-payer and Medicare payment threshold option, which we refer to as the “All-Payer Combination Option.” An eligible clinician need only meet the threshold for one of the options to be a QP for a year. Thus, an Advanced APM Entity may be able to compensate for a relatively low level of Advanced APM participation with participation in Other Payer Advanced APMs such as those with State Medicaid programs and commercial payers. Figure B illustrates the stages of determinations that result in QP determinations.

3. Terms and Definitions

The proposed Quality Payment Program relies on a set of interrelated defined terms. The bases for some core terms are set forth at sections 1833(z)(3) and 1848(q)(1)(C)(iii) of the Act, and others we will propose to define in this proposed rule.

We use the statutory text as a foundation to develop definitions for other key terms used in this proposed rule. The terms cover three primary topics: (1) The different types of APMs and their participating individuals and entities; (2) the timing, process and thresholds for determining QPs and partial qualifying APM participants (Partial QPs); and (3) the payment of the 5 percent lump sum incentive to QPs.

As discussed in sections II.D and II.F.3 of this proposed rule, we are proposing definitions for the following APM-specific terms at § 414.1302 of new subpart O:

• Affiliated Practitioner.
• APM Entity.
• APM Incentive Payment.
• Attributed beneficiary.
• Attribution-eligible beneficiary.
• Alternative Payment Model (APM).
• Advanced Alternative Payment Model (Advanced APM).
• Advanced APM Entity.
• Episode payment model.
• Incentive Payment Base Period.
• Medicaid APM.
• Medicaid Medical Home Model.
• Medical Home Model.
• Other Payer APM.
• Other Payer Advanced APM.
• Partial Qualifying APM Participant (Partial QP).
• Partial QP Patient Count Threshold.
• Partial QP Payment Amount Threshold.
• Qualifying APM Participant (QP).
• QP Patient Count Threshold.
• QP Payment Amount Threshold.
• QP Performance Period.
• Threshold Score.

To organize the terms, we have proposed the term “Advanced APM” for those APMs defined by section 1833(z)(3)(C) of the Act that meet the criteria under section 1833(z)(3)(D) of the Act. The MACRA uses the term “Eligible APM” in the heading for section 1833(z) of the Act, in section 1848(q)(9)(A)(ii) of the Act, and indirectly defines it at section 1833(z)(3)(D) of the Act as the APMs in which “eligible alternative payment entities” participate. We have decided to use the term “Advanced” in lieu of “Eligible,” and rather than referring indirectly, as is done in section 1833(z)(3)(D)(i) of the Act, to the APM in which an eligible alternative payment entity participates, we believe it is essential to the understanding of this proposed rule to be able to identify and propose requirements directly for an Advanced APM.

Similarly, we propose to use the term “Advanced APM Entity” instead of “alternative payment entity” because it highlights the connected but different roles of the Advanced APM (for example, a CMS Innovation Center ACO model meeting specified criteria) and the Advanced APM Entity (for example, a specific ACO participating in that ACO model). We also believe that it is important to the clarity of this proposed rule to define “APM Entity” in addition to “Advanced APM Entity” so that we can easily distinguish between the two under both MIPS and the APM incentives. We propose that an APM Entity would be any participating entity in an APM, whereas we propose that an Advanced APM Entity would be one that participates in an APM that CMS has in fact determined to be an Advanced APM.

We also propose to define the terms “Medical Home Model” and “Medicaid Medical Home Model” as subsets of APMs and Other Payer APMs, respectively. The MACRA provides no definition for the term “medical homes” but makes it an instrumental piece of the law under sections 1848(q)(5)(C)(i), 1833(z)(2)(B)(iii)(II)(cc)(BB), 1833(z)(2)(C)(iii)(II)(cc)(BB), and 1833(z)(3)(D)(ii)(II) of the Act.

We note that medical homes would be the APM Entities in an APM, not the APM itself. The requirements in the MACRA and in this proposed rule actually relate to the disposition of the APM, not the participating medical homes. For instance, as described in section II.F.4.b.(6) of this preamble, section 1115A(c) of the Act relates to the expansion of models (APMs), not the participants (APM Entities) of such models. APM participants are not expanded under section 1115A(c) of the Act. Therefore, we discuss medical homes in terms of the Medical Home Model, which is the concept to which the MACRA and this proposed rule actually refer. Although the definitions are identical but for their payer context, we distinguish Medicaid Medical Home Models because there are specific requirements for them under the determination of Other Payer Advanced APMs as described in section II.F.7.b.(3) of this preamble.

We propose that a Medical Home Model must have the following elements at a minimum:

• Model participants include primary care practices or multispecialty practices that include primary care physician and practitioners and offer primary care services.
• Empanelment of each patient to a primary clinician.

In addition to these elements, we propose that a Medical Home Model must have at least four of the following elements:

• Planned coordination of chronic and preventive care.
• Patient access and continuity of care.
• Risk-stratified care management.
• Coordination of care across the medical neighborhood.
• Patient and caregiver engagement.
• Shared decision-making.
• Payment arrangements in addition to, or substituting for, fee-for-service payments (for example, shared savings, population-based payments).

The two required elements are consistent with the fundamental characteristics of medical homes in the various incarnations and accreditation standards across the health care market. Therefore, we believe that an APM cannot be a Medical Home Model unless it has a primary care focus with an explicit relationship between patients and their practitioners. To determine that an APM has a primary care focus, we propose that the Medical Home Model would have to have involve specific design elements related to Eligible clinicians practicing under one or more of the following Physician Specialty Codes: 01 General Practice; 08 Family Medicine; 11 Internal Medicine; 37 Pediatric Medicine; 38 Geriatric Medicine; 50 Nurse Practitioner; 89 Clinical Nurse Specialist; and 97 Physician Assistant. We solicit comments on whether this proposal for determining that an APM has a primary care focus is sufficiently specified.

We believe the optional elements should be present in Medical Home Models, but individually, each is less definitive of a characteristic than the two required elements. We also want to adhere to our principle of enabling future flexibility of APM design. Extensive rigid Medical Home Model criteria would not serve the purpose of promoting the development of new and potentially better ways of managing patient care through primary care.

We seek comment on these elements and which of the elements should be required as opposed to optional. Our proposed definition of Medicaid Medical Home Model is identical to Medical Home Model, except that it specifically describes a payment arrangement operated by a State under title XIX. It is important to separate the terms because Medicaid Medical Home Models have distinct implications in the Other Payer Advanced APM determination and the QP determination under the All-Payer Combination Option.

We believe that these proposed terms and definitions are sufficient to clearly implement the Quality Payment Program. For example, these terms cover all steps of the incentive payment process, from participation in Advanced APMs to QP determinations and payment of incentives. We are aware that this is a complex program and that we are proposing a significant number of terms. We believe that using more distinctive terms is preferable to using fewer terms that could overlap and convey different meanings in different contexts. For instance, Partial QP Patient Count Threshold is a highly specific term, but we believe that it is necessary in context because there are differences between QPs and Partial QPs, and there are differences between the payment amount and patient count thresholds used to determine whether an Eligible clinician becomes a QP or a Partial QP.

We seek comment on these terms, including how we have defined the term, the relationship between terms, any additional terms that we should formally define to clarify the explanation and implementation of this program, and potential conflicts with other terms used by CMS in similar contexts. We also seek comment on the naming of the terms and whether there are ways to name or describe their relationships to one another that make the definitions more distinct and easier to understand. For instance, we would like to know if commenters believe there are more intuitive or efficient terms than those proposed that would still adhere to the statutory language and the intended purposes of the terms. In particular, we would consider options for a framework of definitions that might more intuitively distinguish between APMs and Other Payer APMs and between APMs and Advanced APMs.

We also seek comment on alternative terms or definitions that are both useful in the calculations described in § 414.1430, § 414.1435, § 414.1440, and § 414.1445 of the proposed rule and easily understood by stakeholders.

4. Advanced APMs

The purpose of this section is to define and outline the proposed criteria for Advanced APMs, APMs through which eligible clinicians would have the opportunity to become QPs as specified in section 1833(z)(3)(C) and (D) of the Act. Other Payer Advanced APMs, types of alternative payment arrangements related to the All-Payer Combination Option, are addressed below in section II.F.7 of this preamble.

First, an Advanced APM must, by statute, meet certain requirements, and we propose details for these requirements within this section. First, the broad category of APMs is defined at section 1833(z)(3)(C) of the Act, which states that an APM is any of the following: (i) A model under section 1115A (other than a health care innovation award); (ii) the Shared Savings Program under section 1899; (iii) a demonstration under section 1866C; or (iv) a demonstration required by Federal law.

We believe it necessary to propose additional clarification around the requirements as defined in section 1833(z)(3)(C)(iv) of the Act given the broad scope of programs and demonstrations required by federal legislation that are administered by the Department. We propose that in order to be an APM as a “demonstration required by Federal law,” the demonstration must meet the following 3 criteria: (1) The demonstration must be compulsory under the statute, not just a provision of statute that gives the agency authority, but one that requires the agency to undertake a demonstration; (2) there must be some “demonstration” thesis that is being evaluated; and (3) the demonstration must require that there are entities participating in the demonstration under an agreement with CMS or under a statute or regulation. We seek comment on our proposal for these criteria defining a demonstration required under Federal law.

Second, to be considered an Advanced APM, an APM must meet all three of the following criteria, as required under section 1833(z)(3)(D) of the Act. The criteria are:

• The APM must require participants to use certified EHR technology;
• The APM must provide for payment for covered professional services based on quality measures comparable to those in the quality performance category under MIPS;
• The APM must either require that participating APM Entities bear risk for monetary losses of a more than nominal amount under the APM, or be a Medical Home Model expanded under section 1115A(c) of the Act. For a discussion of our proposals for Medical Home Models under this criterion, see section II.F.4.b.(6) of this preamble.

We propose that an APM Entity is the participating entity in an APM that is primarily responsible for the cost and quality of care provided to beneficiaries under the terms of a direct agreement with CMS. The term “eligible alternative payment entity” (which we refer to as an “Advanced APM Entity”) is defined under section 1833(z)(3)(D) of the Act. An Advanced APM Entity is an APM Entity that participates in an Advanced APM that, through terms of a Participation Agreement with CMS or through Federal law or regulation, meets the criteria proposed in this rule. In section II.E.2 of this proposed rule, we propose that each unit—APM, APM Entity, and eligible clinician—would be clearly identified in CMS systems by a unique combination of APM identifier/APM Entity identifier/TIN/NPI to be considered for possible determination as an Advanced APM, Advanced APM Entity, or QP, respectively.

In some cases, APMs offer multiple options or tracks with variations in the level of financial risk, or multiple tracks designed for different types of organizations, and we propose to assess the eligibility of each such track or option within the APM independently. For instance, the Medicare Shared Savings Program (Shared Savings Program) has three distinct tracks, the Comprehensive ESRD Care Initiative (CEC) consists of one track for large dialysis organizations and another track for non-large dialysis organizations, and the Next Generation ACO Model has two risk arrangement options that feature different levels of financial risk.

Significant distinctions between the design of different tracks or options may mean that some tracks or options within an APM would meet the proposed Advanced APM criteria while other tracks or options would not. For example, APM Entities may have the option to assume two-sided risk (meaning that they bear a portion of the losses when spending exceeds expectations and share in the savings when spending is below expectations) or one-sided risk (meaning that they share in the savings when spending is below expectations, but do not bear a portion of the losses when spending exceeds expectations) under an APM. If the one-sided risk track does not meet the standard for financial risk as discussed in section II.F.4.b.(3) of this preamble, APM Entities in this track would not be Advanced APM Entities, whereas those in the two-sided risk track could be Advanced APM Entities. In these instances, we propose that we would distinguish that the APM is only an Advanced APM for specific options or tracks.

All entities participating in Advanced APMs are Advanced APM Entities, and distinguishing between the model and the participating entities allows us to directly identify and discuss the requirements unique to each. This approach to identifying Advanced APMs and Advanced APM Entities is also consistent with our proposal for determining QPs, described in section II.F.5 of this preamble, at the Advanced APM Entity level. We believe that because the Advanced APM Entity is the main participant in an Advanced APM, it should therefore be the operative unit by which QP determinations are made.

We propose that an eligible clinician's QP status for a given payment year would be based on a collective evaluation of a group consisting of all eligible clinicians participating in an Advanced APM Entity. All eligible clinicians in an Advanced APM Entity would be identified as participants according to their APM participant identifiers in CMS systems as described in section II.E.2 of this preamble. To attain QP status, we propose that an eligible clinician would have to be listed on December 31 of the QP Performance Period as part of an Advanced APM Entity that, through the collective calculation of all its eligible clinicians, meets the QP Payment Amount Threshold or the QP Patient Count Threshold, both of which are described in section II.F.5 of this preamble. The form and collection of this list is part of the APM's design. For example, an ACO in the Shared Savings Program is comprised of a list of participating Medicare-enrolled TINs (ACO participants) that includes all eligible clinicians, as identified by their NPIs, who bill through those TINs. The group of eligible clinician TIN/NPI combinations determined as of December 31 at the end of each performance year, consistent with the proposals above, would be used to make a QP determination that would apply to all eligible clinicians on the list.

Only eligible clinicians in Advanced APM Entities during the QP Performance Period would have the potential to become QPs and to qualify for the APM Incentive Payment. If the eligible clinicians in the Advanced APM Entity collectively meet the QP Payment Amount Threshold, QP Patient Count Threshold, Partial QP Payment Amount Threshold, or Partial QP Patient Count Threshold criteria as described in section II.F.5 of this preamble, we propose that all of those eligible clinicians in the group defined by the Advanced APM Entity would receive the QP status for the relevant payment year. For example, in the event that a track in the Shared Savings Program is determined to be an Advanced APM and the eligible clinicians in an ACO participating in that track (the Advanced APM Entity) collectively meet the QP threshold criteria, all of the eligible clinicians (as identified by their TIN/NPI combinations) in the ACO would become QPs.

In sections II.F.5 and II.F.8 of the proposed rule, we propose that such QP status would apply to the individual eligible clinician's NPI across all of the TINs to which he or she reassigned the right to receive Medicare payment, not solely to the billing TIN affiliated with the Advanced APM Entity. We believe that this approach is consistent with the statute and prevents situations in which an eligible clinician may be excluded from MIPS for part of his or her practice but still subject to MIPS with respect to another part of his or her practice.

Table 27 illustrates how hypothetical APM designs could intersect with proposed MACRA definitions.

a. Advanced APM Determination

In order to determine Advanced APMs and achieve transparency for the Quality Payment Program, we propose to establish a process by which we identify and notify the public of the APMs (including specific APM tracks or options) that would be considered Advanced APMs for a QP Performance Period. We would post this notification to the CMS Web site prior to the beginning of the first QP Performance Period and update the information on a rolling basis according to the proposals below. We believe that making this information available in an accessible format is important for stakeholders to understand how CMS applies the Advanced APM criteria to existing APMs, and to be informed as early as possible about whether an APM they are considering joining is an Advanced APM. Similar to our stated principles earlier in this preamble, we believe that participation in APMs that are not Advanced APMs would continue to offer significant opportunities to eligible clinicians who are not immediately able or prepared to take on the additional risk and requirements of Advanced APMs.

To determine Advanced APMs, we propose two phases of determination and notice. First, we propose to release an initial set of Advanced APM determinations no later than January 1, 2017, for APMs that will be operating during the first QP Performance Period. Second, for new APMs that are announced after January 1, 2017, CMS would include its Advanced APM determination in conjunction with the first public notice of the model, such as the Request for Applications (RFA) or proposed rule. We propose that determinations of Advanced APMs would be posted on the CMS Web site and updated on an ad hoc basis to the extent feasible, but no less frequently than annually, as new APMs become available and others end or change. Both the initial and ad hoc notifications would contain descriptions of whether each track or option within an APM would result in different Advanced APM statuses. We believe that this proposal incorporates both the interest in immediate dissemination of Advanced APM determinations for the existing APM portfolio following finalization of this rule and the structure for making the Advanced APM status a regular part of the development and release of new APMs in the future.

We seek comment on the proposals for both the initial and ad hoc notices of Advanced APM determinations. In particular, we seek comments on optimal times, locations, formats, and other methods of notice of Advanced APM determinations to promote clarity and consistency around which APMs are considered Advanced APMs for a particular QP Performance Period.

In addition to identifying Advanced APMs, we propose that we would identify Other Payer Advanced APMs. The Other Payer Advanced APM identification process would go into effect starting in the third QP Performance Period (applicable for payment year 2021) and would align with the availability of the All-Payer Combination Option for QP determinations. We propose that Other Payer Advanced APM determinations and associated notice would rely on information submitted by APM Entities and eligible clinicians as described in section II.F.7.d of this preamble and would operate in conjunction with the QP determination process under the All-Payer Combination Option as described in section II.F.7 of this preamble. If the information needed by CMS to make a determination for the Other Payer Advanced APM is not submitted in the manner and by the deadlines set by CMS through subregulatory guidance, we would not assess that Other Payer APM as explained under section II.F.7 of this preamble.

b. Advanced APM Criteria

Under MACRA, for an APM to be an Advanced APM it must meet the criteria set forth in sections 1833(z)(3)(C) and (D) of the Act and discussed below. An Advanced APM must be an APM that:

• Requires its participants to use certified EHR technology (CEHRT), as described in section II.F.4.b.(1) of this preamble;
• Provides for payment for covered professional services based on quality measures comparable to measures under the quality performance category under MIPS, as described in II.F.4.b(2); and
• EITHER (a) requires its participating Advanced APM Entities to bear financial risk for monetary losses that are in excess of a nominal amount, as described in section II.F.4.b(3) of this preamble, or (b) is a Medical Home Model expanded under section 1115A(c) of the Act, as described in section II.F.4.b(4) of this preamble.

These requirements as set forth in the statute and proposed in this section must be met through the design of the APM. Whether an APM is an Advanced APM depends solely upon how the APM is designed, rather than on assessments of participant performance within the APM. Some stakeholders have suggested that actual performance (for example, on clinical quality measures or on whether the Advanced APM Entity generates savings) be considered in the determination of QPs. However, the incentives for Advanced APM participation, as required under section 1833(z) of the Act, does not provide for consideration of actual performance in making such determinations. Performance assessments are already part of APMs, and we believe it is important and consistent with the statutory framework to continue to foster flexibility in structuring the specific rewards and consequences of performance within each APM.

For example, an APM that ties payments to performance on quality measures comparable to those under MIPS may be an Advanced APM regardless of an Advanced APM Entity's actual performance on those quality measures. If an Advanced APM Entity fails to meet quality performance standards under the Advanced APM, it would face consequences within the Advanced APM, such as financial penalties, loss of access to data or certain waivers, or termination of its participation agreement. The termination scenario would have the downstream effect of terminating Advanced APM Entity status and the eligible clinicians' potential eligibility for the APM Incentive Payment because the entity would no longer be participating in the Advanced APM. As another example, an Advanced APM Entity that bears more than nominal financial risk for monetary losses in accordance with the standards set forth in section II.F.4.b.(3) of this preamble would be an Advanced APM Entity regardless of whether it actually earns shared savings or generates shared losses under the Advanced APM. This would work similarly for an Other Payer Advanced APM.

We do not intend to add additional performance assessments on top of existing Advanced APM standards. As stated in the discussion of policy principles at the beginning of section II.F.1 of this preamble, the proposed QP determination process assesses the relative degree of participation of the Advanced APM Entity and eligible clinician in Advanced APMs, not their performance success as assessed under the APM. The Quality Payment Program would not alter how each particular APM measures and rewards success within its design. Rather, the Quality Payment Program rewards a substantial degree of participation in certain APMs.

(1) Use of Certified EHR Technology

The first criterion an APM must meet to be considered an Advanced APM is that it requires participants in such model to use certified EHR technology (as defined in section 1848(o)(4) of the Act), as specified in section 1833(z)(3)(D)(i)(I) of the Act. Furthermore, to be considered an Other Payer Advanced APM, as described under sections 1833(z)(2)(B)(iii)(II)(bb) and 1833(z)(2)(C)(iii)(II)(bb) of the Act, payments must be made under arrangements in which certified EHR technology is used. Although the statutory requirement is phrased slightly differently for Advanced APMs and Other Payer Advanced APMs, we believe that there is value in keeping the two standards as similar as possible. We received a number of comments on the MIPS and APMs RFI regarding the definition and use of CEHRT by APMs. A number of commenters recommended that CMS use the same CEHRT definition for APMs that is used for the MIPS program to reduce confusion among participants in these programs and to align the program requirements. Some commenters suggested we should not require additional CEHRT requirements for APMs, while others indicated that current health IT is not adequate to support practice transformation efforts to perform as a patient centered medical home. Other commenters indicated the focus should not be on the technology used, but rather the design and purpose of the APM. A few commenters indicated there was a need to develop certified health IT for specialty eligible clinicians. Additionally, psychologists, plastic surgeons, radiologists, and other specialists commented that they did not want to be left out of APMs because they did not have certified health IT meeting the CEHRT definition now or may not use CEHRT for the same functions as other eligible clinicians.

After consideration of these comments, we propose to adopt for Advanced APMs and Other Payer Advanced APMs, the definition of CEHRT that is proposed for MIPS and the APM incentive under § 414.1305. In the 2015 EHR Incentive Programs final rule (80 FR 62872 through 62873), we established the definition of CEHRT for EHR technology that must be used by Eligible Professionals to meet the meaningful use objectives and measures in specific years. In this proposed rule, we are proposing to adopt the specifications from within the current definition of CEHRT in this regulation at § 414.1305 for eligible clinicians participating in MIPS or in APMs. This definition is similar to the definition that applies to eligible hospitals, CAHs, and eligible professionals (EPs) in the EHR Incentive Programs. The definition includes the certification criteria for a wide range of standards for use in capturing patient health information like vital signs, medications and medication allergies, problem list, and lab results among other data elements including the common clinical data set (CCDS). It also includes the certification criteria and standards for functions related to information exchange, patient engagement, quality reporting, and protecting the privacy of electronic protected health information. For further information on the certification criteria see the 2015 Edition Certification Criteria final rule (80 FR 62602 through 62759) and for example Table 8: “Common Clinical Data Set” (80 FR 62696).

This approach aligns the APM health IT certification requirements for Advanced APMs with those used by MIPS eligible clinicians. We understand this proposed CEHRT definition may include some EHR functionality used by MIPS eligible clinicians which may be less relevant for an APM participant, and likewise APM participants may use additional functions that are not required for MIPS participation. However, we observe that APM participants often work in the same office space, group, entity, or organization with eligible clinicians that are not APM participants. At times they might share common resources, such as the same EHR system. Using the same CEHRT definition for both MIPS and Advanced APMs would allow Eligible clinicians to continue to use shared EHR systems and give eligible clinicians flexibility of participation as a MIPS eligible clinician or an eligible clinician in an Advanced APM without needing to change or upgrade EHR systems. Although updates to the certified health IT for APM participants, MIPS participants, or both may be necessary in future years, we believe that aligning the APM and MIPS definition for CEHRT is appropriate at this time.

We seek comment on the proposed definition of CEHRT for Advanced APMs and Other Payer Advanced APMs and whether the definition should be the same for both.

The statute does not specify the number of eligible clinicians who must use CEHRT or how CEHRT must be used in an Advanced APM. We believe CMS has discretion to define the ways in which an Advanced APM uses CEHRT. In accordance with section 1833(z)(3)(D)(i)(I) of the Act, we propose that an Advanced APM must require at least 50 percent of eligible clinicians who are enrolled in Medicare (or each hospital if hospitals are the APM participants) to use the certified health IT functions outlined in the proposed definition of CEHRT to document and communicate clinical care with patients and other health care professionals. Communicating clinical care means that other eligible clinicians and/or the patient can view the clinical care information. Later in this section, we also propose an alternative set of criteria applicable to the Shared Savings Program to demonstrate the use of CEHRT by their eligible clinicians in order to be an Advanced APM. We propose the 50 percent threshold be confined to the first QP Performance Period (proposed later in this rule to be 2017). That is, only in 2017 could APMs use the 50 percent threshold for eligible clinicians in each participating entity to meet the use of CEHRT requirement. We propose that the threshold requirement for use of CEHRT would increase to 75 percent beginning for the second QP Performance Period (proposed to be 2018). The requirement for hospitals participating in Advanced APMs would remain the same over time because it is an all-or-nothing requirement of the hospital as a single entity.

We believe there are a few reasons why having a lower threshold requirement for the use of CEHRT by the eligible clinicians participating in an APM Entity in the first year is appropriate. First, we want to ensure that APMs have sufficient time to alter their terms and conditions to meet this standard. We also acknowledge that eligible clinicians will be expected to upgrade from technology certified to the 2014 Edition to technology certified to the 2015 Edition for use in 2018, and some eligible clinicians who have not yet adopted CEHRT may wish to delay acquiring CEHRT products until a 2015 Edition certified product is available.

Although these are important considerations for the first year of the program, we believe that APMs should expect their APM Entities to meet a higher standard for the use of certified EHR technology in future years. We note that several APMs that are likely to meet the other criteria to be an Advanced APM have already demonstrated higher rates of achievement of meaningful use under the EHR Incentive Program that exceed the requirements under the APM. For instance, an analysis of 2014 performance year quality reporting data under the SSP showed that an average of 86 percent of primary care physicians met meaningful use requirements in 2014 (See https://www.cms.gov/Newsroom/MediaReleaseDatabase/Fact-sheets/2015-Fact-sheets-items/2015-08-25.html ). Other APMs require all eligible clinicians to use CEHRT as a requirement for participation in the APM. We believe that, based on the focus of an Advanced APM, this criterion should challenge APMs and their participants to adopt CEHRT at high rates and use its capabilities to deliver high value care. The adoption of CEHRT is critical to supporting increased care coordination, electronic clinical quality measure reporting, electronic clinical decision support, and many other capabilities supportive of success in APMs, and we believe these capabilities should be widely available to eligible clinicians participating in APMs. Therefore, we believe that raising the threshold for use of CEHRT required to be an Advanced APM would be appropriate for future years beginning in QP Performance Period 2018.

Stakeholders should keep in mind that this CEHRT requirement would be based on the requirements that an APM places on its participating APM Entities. In determining whether an APM meets this criterion, CMS does not propose to assess the level of use of each APM Entity or individual eligible clinician participating in the APM but rather whether the APM requirements meet the standard set forth in this proposed rule.

We invite comment on whether the proposed thresholds for use of CEHRT for APM Entities that are not hospitals (50 percent for the first QP Performance Period (proposed 2017) and 75 percent for the second QP Performance Period (proposed 2018) and later are appropriate, or if we should consider additional options such as a higher or lower percentage in 2018, or an additional incremental increase for 2019. We also invite comment on whether we should consider higher thresholds for APMs that target eligible clinician populations with higher-than-average adoption of certified health IT, such as eligible clinicians in patient-centered medical homes. Finally, we invite comment on whether we should explore ways to set lower thresholds for those APMs targeting eligible clinician populations that may have lower average adoption of certified health IT, such as specialty-focused APMs.

We also propose an alternative criterion for determining whether an APM meets the CEHRT requirement, exclusively applicable for the Shared Savings Program. We believe this method is appropriate for the Shared Savings Program because although the Shared Savings Program requires ACOs to encourage and promote the use of enabling technologies (such as EHRs) to coordinate care for assigned beneficiaries, the Shared Savings Program does not require a specific level of CEHRT use for participation in the program. Instead, the Shared Savings Program includes an assessment of EHR use as part of the quality performance standard which directly impacts the amount of shared savings/losses generated by the Shared Savings Program ACO. We believe it is important to incentivize ever-increasing level of CEHRT use. However, in contrast to CMS APMs under section 1115A of the Act, CMS would have to undertake significant rulemaking to adopt an eligibility standard for the Shared Savings Program that is consistent with the proposed criterion for other CMS APMs. Following such rulemaking, we would have to collect additional information from each existing and applying ACO outside the routine application process in the weeks prior to the start of the 2017 performance year which we believe could introduce uncertainty and burden for CMS, ACOs, and participating EPs. Moreover, we believe that the proposed alternative criterion builds on established Shared Savings Program rules and incentives that directly tie the level of CEHRT use to the ACO's financial reward which in turn has the effect of directly incentivizing ever-increasing levels of CEHRT use among EPs. We believe that the proposed alternative criterion for the Shared Savings Program is consistent with the goals of the APM incentive and reduces burden and uncertainty for the Shared Savings Program participants. Therefore, because most other APMs can accommodate a new CEHRT use requirement for eligible clinicians without modifying our regulations, we are restricting this method to the Shared Savings Program. We propose that this alternative would allow the Shared Savings Program to meet the criterion if it holds APM Entities accountable for their eligible clinicians' use of CEHRT by applying a financial penalty or reward based on the degree of CEHRT use (such as the percentage of eligible clinicians that use CEHRT or the engagement in care coordination or other activities using CEHRT). One of the quality measures used in the Shared Savings Program's quality performance standard assesses the degree to which certain eligible clinicians in the ACO successfully meet the requirements of the EHR Incentive program, which requires the use of CEHRT by certain eligible clinicians in the ACO. Successful reporting of the measure for a performance year gives the ACO points toward its overall quality score, which in turn affects the amount of shared savings or shared losses an ACO could earn or be liable for, respectively. Because of this, ACOs in the Shared Savings Program actively promote and seek to improve upon the EHR measures annually, leading to greater use of CEHRT among eligible clinicians participating in Shared Savings Program ACOs. We believe our proposed criteria for APMs, generally, and our alternative for the Share Savings Program, would meet the statutory requirement of section 1833(z)(3)(D)(i)(I) of the Act, as both hinge upon the Advanced APM requiring its participants use CEHRT with consequences for failure to meet the APM's standards. We solicit comment on our proposed methods for meeting the criterion for an Advanced APM to require its participants to use CEHRT as specified in section 1833(z)(3)(D)(i)(I) of the Act.

In addition to these proposals, we are interested in what other health IT functionalities APM participants might need to effectively provide care to their patients and how the use of interoperable health IT can strengthen and encourage higher quality patient care and more effective care coordination across all APMs. Recent research and input from experts, practitioners, and the public (See https://www.healthit.gov/facas/sites/faca/files/HITPC_AHMWG_Meeting_Slides_Final_Version_9_2015-11-10.pdf ) has identified priority health IT capabilities that will be important for participants in APMs but are not yet widely available in current health IT systems, such as the ability to manage and track status of referrals and create and maintain electronic shared care plans for team-based care management.

We look forward to receiving comments as to whether new health IT standards and certification criteria may be needed to ensure that participants in APMs have access to interoperable health IT products and services necessary for effective care coordination, population health management, and patient engagement. We will work with the Office of the National Coordinator (ONC) to explore opportunities for certified health IT capabilities reflected in the CEHRT definition to evolve in ways that meet the needs of participants in APMs while supporting eligible clinicians in MIPS to fulfill the EHR performance category under MIPS.

We believe that all patients, families, and healthcare professionals should have consistent and timely access to health information in a standardized format that can be securely exchanged between these parties (See HHS August 2013 Statement, “Principles and Strategies for Accelerating Health Information Exchange”). The secure, appropriate exchange of health information can help health care professionals improve quality of care through more robust care coordination, and improve the efficiency of care through access to patient information across settings. Interoperability is a key priority for the healthcare industry. HHS recently received pledges from companies that provide 90 percent of the electronic health records used by hospitals nationwide, as well as the top five largest health care systems in the country, to: help consumers easily and securely access their electronic health information; help clinicians share individuals' health information for care with other clinicians and their patients whenever permitted by law and not block electronic health information; and implement federally recognized, national interoperability standards, policies, guidance, and practices for electronic health information.

A growing number of organizations across the country are now focused on facilitating health information exchanges (HIEs) among healthcare professionals at the national, state, and community levels. According to one figure, there were 267 organizations providing HIE services operating in the U.S. in 2014 (see https://ehi-rails-app.s3.amazonaws.com/uploads/article/file/476/2014_eHI_Data_Exchange_Survey_Results_Webinar_Slides.pdf ), including community-based organizations, statewide efforts, and other healthcare delivery entities supporting exchange. While representing a wide variety of stakeholders, services and structures, these organizations play an important role in facilitating care coordination and data sharing for many health care professionals across the country. We encourage the growth of these services and encourage healthcare professionals to explore partnering with organizations offering HIE services.

We seek comment on how requirements for the use of CEHRT within APMs could evolve to support expanded participation in organizations supporting HIEs. For instance, should CMS consider expanding in future rulemaking the CEHRT criterion for Advanced APMs to include recognition of participation with an organization providing HIE services? Would this option be likely to spur further interest among entities in partnering with organizations that provide HIE services? Should these organizations be required to adhere to specific standards that promote interoperability across health information systems? How could a potential future governance mechanism for HIE (that is, establishing a common set of standards, services, policies, and practices) be incorporated into requirements for APMs? We seek comment on these and any other issues related to advancing participation in HIEs though the use of CEHRT in APMs.

(2) Comparable Quality Measures

The second criterion for a APM to be an Advanced APM is that it provides for payment for covered professional services based on quality measures comparable to measures under the performance category described in section 1848(q)(2)(B)(i) of the Act, which is the MIPS quality performance category. We interpret this criterion to require the APM to incorporate quality measure results as a factor when determining payment to participants under the terms of the APM.

Our proposed policy for this criterion is informed by our proposed policy for the MIPS quality performance category. For more information on quality measures under the MIPS quality performance category, please see section II.E.3.b of this preamble of this preamble, in which CMS proposes eligible clinicians will select quality measures from the MIPS measures list in section II.E.3 of this preamble for the first performance year of MIPS. We will publish a list of quality measures annually, through notice and comment rulemaking, from which MIPS eligible clinicians may choose measures for assessment under the MIPS quality performance category. The measures included in the annual list of MIPS measures must adhere to specific criteria that include the following: (1) Measures must have an evidence-based focus if the measures are not endorsed by a consensus-based entity as described in section 1848(q)(2)(D)(v) of the Act; and (2) new measures and the method for developing and selecting such measures, including clinical and other data supporting such measures, must be submitted to a specialty-appropriate, peer-reviewed journal prior to inclusion of the measure in MIPS as described in section 1848(q)(2)(D)(iv) of the Act.

The statute also establishes priorities for both the quality domains of measures to be developed and the types of measures to be prioritized in the measure development plan, which are located, respectively at sections 1848(s)(1)(B) and (D) of the Act. The priority measure types include outcome, patient experience, care coordination, and measures of appropriate use of services such as measures of overuse.

We are considering a number of ways to implement the Advanced APM requirement to base payment on measures comparable to those in MIPS, as well as how to define which measures would reflect the statutory requirement to be “comparable” to MIPS quality measures. Some of the options we explored for defining measures comparable to those in MIPS included: (1) Limiting comparable measures to those from the annual MIPS list of measures; and (2) including quality measures from the annual MIPS list of measures and/or measures that have an evidence-based focus and are found to be reliable and valid through measure testing. We also explored whether we should require a minimum number of measures for all Advanced APMs, and whether the number of measures would need to be the same as those required under the MIPS quality performance category.

In exploring these options we decided that while they all have merit, we are concerned they may be overly restrictive for the variety of APMs, many of which are designed to have the flexibility to test new ways of paying for and delivering care. We want to ensure that APMs have the latitude to base payment on quality measures that meet the goals of the model and assess the quality of care provided to the population of patients that the APM participants are serving. It is important to note that many APMs include some common measures that are proposed for inclusion in MIPS. For example, many of the quality measures used in the Shared Savings Program and the Next Generation ACO Model that are submitted to CMS through the CMS web interface, are also proposed for inclusion in MIPS.

However, APMs that focus on patients with specific clinical conditions, such as end-stage renal disease or patients undergoing specific surgical procedures, would have valid reasons for including different quality measures than those that target more general populations. Similarly, some models may focus on specialist eligible clinicians for whom there may be only a small number of valid and relevant quality measures. Lastly, we cannot predict the specific care goals and payment designs of future physician-focused payment models and other APMs. Consequently, we do not want to impose measure requirements that may prevent CMS from including quality measures that may be better suited to the specific aims of new innovative APMs.

We received a number of comments on the MIPS and APMs RFI on the use of MIPS-comparable quality measures by an Advanced APM. A commenter suggested CMS include high-value performance measures to assess and improve the quality of care that are clinically important, evidence-based, transparent, feasible, valid and reliable, actionable, and rigorously audited to ensure accuracy. Other commenters indicated APMs should not be required to have the same reporting requirements as is required under the quality reporting performance category for MIPS because each APM is designed differently and may be developed with a specific specialty or condition in mind, so broad reporting requirements would not be relevant. Commenters also indicated the need for measures that could be used across APMs and MIPs to reduce the eligible clinician's reporting burden when switching from one program to the other.

After consideration of the comments and the options above, we recognize the need to propose a measure framework for comparable measures that reflects a few key principles. For the Advanced APM measures to be comparable to MIPS measures, the measures should have an evidence-based focus and as appropriate, target the same priorities, (for example, clinical outcomes, use and overuse). However, as each APM Entity is different, there needs to be the flexibility to determine which measures are most appropriate for use in their respective APM for the purpose of linking those measures to payment under the model. We agree that measures that could be used in both MIPS and APMs is beneficial to eligible clinicians who may switch from one program to the other, but we also do not want to restrict APMs from including new innovative measures that may not be included in MIPS initially, or until later years of the program.

We also note that under the MACRA and in this proposal, not all quality measures under which an APM is assessed are required to be “comparable” and not all payments under the APM must be based on comparable measures. However, at least some payments must be tied to measures comparable to MIPS, regardless of whether those comparable measures are the only ones the APM uses. Under this proposal, APMs retain sufficient freedom to innovate in paying for services and measuring quality. For instance, an APM may have incentive payments related to quality, total cost of care, participation in learning activities, and adoption of health IT. The existence of all of the payments associated with non-quality aspects does not preclude the APM from meeting this Advanced APM criterion. In other words, this criterion only sets standards for payments tied to quality measurement, not other methods of payment. Conversely, an APM may, as current models at the CMS Innovation Center currently do, test new quality measures that do not fall into the MIPS-comparable standard. So long as the APM meets the requirements set forth in this criterion, there is no additional prescription for how the APM tests additional measures that may or may not meet the standards under this criterion. Therefore, we propose that the quality measures on which the Advanced APM bases payment must include at least one of the following types of measures provided that they have an evidence-based focus and are reliable and valid:

(1) Any of the quality measures included on the proposed annual list of MIPS quality measures;

(2) Quality measures that are endorsed by a consensus-based entity;

(3) Quality measures developed under section 1848(s) of the Act;

(4) Quality measures submitted in response to the MIPS Call for Quality Measures under section 1848(q)(2)(D)(ii) of the Act; or

(5) Any other quality measures that CMS determines to have an evidence-based focus and be reliable and valid.

We believe that quality measures that are endorsed by the National Quality Forum would meet these criteria. We also propose to establish an Innovation Center quality measure review process for those measures that are not NQF-endorsed or included on the final MIPS measure list to assess if the quality measures have an evidence-based focus, and are reliable and valid. For example, the Comprehensive ESRD Care Model includes NQF #0226 Influenza Immunization for the ESRD Population which is not a measure included for reporting in MIPS but meets the proposed criteria for MIPS-comparable quality measures. We believe under the proposed categories above MIPS-comparable quality measures may include measures that are fully developed after being tested in an APM and found to be reliable and valid. Similarly, we believe that MIPS-comparable quality measures may include QCDR measures provided that the QCDR measures used by the Advanced APM for payment have an evidence-based focus and are reliable and valid.

The statute identifies outcome measures as a priority measure type and we want to encourage the use of outcome measures for quality performance assessment in APMs. Therefore, we propose that in addition to the general comparable quality measure requirements proposed in this section, an Advanced APM must include at least one outcome measure if an appropriate measure (that is, the measure addresses the specific patient population and is specified for the APM participant setting) is available on the MIPS list of measures for that specific QP Performance Period, determined at the time when the APM is first established. If there is no such measure available on the MIPS list at the time the APM is established, then CMS would not require an outcome measure be included after APM implementation.

We believe that this framework would provide the flexibility needed to ensure APM quality performance metrics meet the APM's goals. We invite comments on whether measures to be considered comparable to MIPS should all be reliable and valid and have an evidenced-based focus.

(3) Financial Risk for Monetary Losses

(a) Overview

The third criterion that a APM must meet to be an Advanced APM is that it must either be a Medical Home Model expanded under section 1115A(c) of the Act as described below, or the APM Entities under the APM must bear financial risk for monetary losses under such APM that are in excess of a nominal amount. We will refer to the latter criterion as the “financial risk criterion.” The proposed correlating financial risk criterion for Other Payer Advanced APMs is described in section II.F.7 of this preamble with the requirements for consideration under the All-Payer Combination Option that is applicable in payment years 2021 and later.

The proposed financial risk criterion for Advanced APMs would apply to the design of the APM financial risk arrangement between CMS and the participating APM Entity. If the structure of the arrangement meets the proposed financial risk requirements, then this criterion would be met. This proposal would not impose any additional performance criteria related to bearing financial risk. For example, eligible clinicians under the Advanced APM Entity would not need to bear financial risk under the APM so long as the APM Entity bears that risk. Furthermore, an APM Entity would not need to actually achieve savings or other metrics for success under the APM in order for the APM to meet this criterion.

This discussion is broken into two main topics: (1) What it means for an APM Entity to bear financial risk for monetary losses under a APM; and (2) what levels of risk CMS would consider to be in excess of a nominal amount. In developing our proposed policies we prioritized keeping these standards consistent across different types of APMs, including Other Payer Advanced APMs as described in section II.F.7.b.(6) of this preamble. We believe that keeping these standards consistent to the extent possible would make it easier for stakeholders, APM Entities, and eligible clinicians to understand the type of financial risk required in order for an APM to be an Advanced APM. However, we do propose to specify small variations in the requirements in order to accord with the differing characteristics of certain types of APMs.

In particular, we propose specific standards that would apply for Medical Home Models. We believe that, given the unique financial risk and nominal amount standards we are proposing for Medical Home Models in this section below, it would be appropriate to impose size and composition limits for the Medical Home Models to which the unique standards would apply in order to ensure that the focus is on organizations with a limited capacity for bearing the same magnitude of financial risk as larger APM Entities do. We propose that beginning in the second QP Performance Period (proposed to be 2018), the Medical Home Model financial risk standard and nominal amount standard, described in section II.F.4.b.(4) of this preamble, would only apply to APM Entities that participate in Medical Home Models and that have 50 or fewer eligible clinicians in the organization through which the APM Entity is owned and operated. Thus, in a Medical Home Model that is an Advanced APM, the proposed Medical Home Model financial risk and nominal amount standards would only apply to those APM Entities owned and operated by organizations with 50 or fewer eligible clinicians. We believe it is appropriate to use eligible clinicians, rather than physicians, when setting this threshold as the number of eligible clinicians both reflects organizational resources and capacity and also may fluctuate widely around a specific number of physicians. We also believe that this size threshold of 50 eligible clinicians is appropriate because organizations of that size have demonstrated the capacity and interest in taking on higher levels of two-sided risk either by themselves or by joining with other organizations. In the event that a Medical Home Model happens to meet the generally applicable financial risk and nominal amount standards, this organizational size limitation would be moot.

Measuring organizational size based on the size of the “parent organization” differs from measuring it based on the size of the APM Entity. Collecting accurate information on the number of eligible clinicians affiliated with a parent organization will require additional, but we believe achievable, reporting by APM Entities. We believe that size of the organization is generally a better indication of risk-bearing capacity than APM Entity size. For instance, an APM Entity may be very small if it represents one practice site, but that practice site may be one of many affiliated with a health system or independent physician association of substantial size. We believe that the proposed limits on the types and sizes of entities that can be Advanced APM Entities under Medical Home Models will encourage larger organizations to move into Advanced APMs with greater levels of risk than the smaller levels that could enable Medical Home Models to become Advanced APMs. This is consistent with our goals that the incentives for Advanced APM participation should reward commitment to challenging models. However, we do not intend to imply that participation in Medical Home Models is necessarily inappropriate for larger organizations. We recognize that Medical Home Models differ from other APMs, such as ACO initiatives, because Medical Home Models focus on improving primary care through much more targeted and intensive interventions than those commonly found in other APMs. We hope to encourage participation in Medical Home Models for all organizations that can derive value from their designs, not just those that are too small to join ACO initiatives and other higher risk APMs.

We propose implementing this size limitation for Advanced APMs that are Medical Home Models beginning in the second year of the Quality Payment Program (proposed QP Performance Period 2018) because we understand that applications for many APMs will be due to CMS before this rule will be finalized, precluding APM Entities from having time to substantially adjust their APM participation strategies for the 2017 QP Performance Period. We propose that CMS would make a determination of whether an APM Entity meets the size limitation prospectively before a QP Performance Period, and that the determinations would not subsequently change based on changes in organizational size during or after the QP Performance Period (although changes in organizational size would, as applicable, affect determinations for subsequent QP Performance Periods). We want all organizations to have the greatest amount of knowledge possible about their APM participation options prior to making the important decision of which APM or APMs to pursue.

We seek comment on this proposal, particularly with regard to the use of the count of eligible clinicians in the parent organization of the APM Entity as the metric of organizational size for Medical Home Models, and whether setting the limit at 50 for the number of eligible clinicians in the organization would constitute a reasonable threshold to distinguish between organizations that we could expect to have the financial capability to join APMs, such as ACO initiatives, that have two-sided risk. We also seek comment on an alternative option to establish the size limitation based on the number of eligible clinicians in the Medical Home Model, rather than on number of eligible clinicians in the APM Entity's organization. Under this alternative option, we would modify the Medical Home Model definition so that an APM could only be considered a Medical Home Model if no more than 10 percent of eligible clinicians (or, alternatively, 10 percent of APM Entities) in the APM are part of parent organizations with more than 50 eligible clinicians. If this element of the Medical Home Model definition were met (along with all other Medical Home Model elements), all APM Entities participating in the APM would be considered medical homes regardless of their size. Conversely, if more than 10 percent of eligible clinicians (or alternatively, 10 percent APM Entities) participating in the APM are part of parent organizations with more than 50 eligible clinicians, the entire APM would not be a Medical Home Model, and, in the event that the APM does not meet the generally applicable Advanced APM financial risk criterion, none of the participating APM Entities would be Advanced APM Entities.

(b) Bearing Financial Risk for Monetary Losses

In this section, we propose a generally applicable financial risk standard for Advanced APMs and a unique standard that would apply only for Advanced APMs that are identified as Medical Home Models.

(i) Generally Applicable Advanced APM Standard

First, we propose that the generally applicable financial risk standard for Advanced APMs would be that an APM must include provisions that, if actual expenditures for which the APM Entity is responsible under the APM exceed expected expenditures during a specified performance period, CMS can:

• Withhold payment for services to the APM Entity and/or the APM Entity's eligible clinicians;
• Reduce payment rates to the APM Entity and/or the APM Entity's eligible clinicians; or
• Require the APM Entity to owe payment(s) to CMS.

The proposed financial risk standard for Advanced APMs reflects our interpretation of the statutory requirement that Advanced APM Entities must bear financial risk for monetary losses to encompass “losses” that could be incurred through either direct repayments to CMS or reductions in payments for services. The former would cover two-sided risk arrangements such as shared savings initiatives in which an Advanced APM Entity may receive shared savings or be liable for shared losses. The latter would cover a range of alternative methods for linking performance to payment, such as payment withholds subject to successful performance, or discounts in payment rates retrospectively applied at reconciliation similar to those in many episode-based bundled payment models. We note that the proposed generally applicable financial risk standard would not include reductions in bonus payments—such as shared savings payment incentives that vary based on quality performance—whereas, as described below, the Medical Home Model financial risk standard could be satisfied by such reductions in bonus payments if appropriate conditions are met. As such, except when the Medical Home Model standard applies, one-sided risk arrangements would not meet this financial risk criterion.

We believe that statute supports a financial risk criterion that should be met only by those APMs that are most focused on challenging organizations, physicians, and practitioners to assume financial risk and provide high-value care. Our proposal reflects our belief that more and more APMs will meet this high bar over time. In response to the MIPS and APMs RFI, many stakeholders commented that business risk should be sufficient to meet this financial risk criterion to be an Advanced APM. We also considered whether the substantial time and money commitments required by participation in certain APMs would be sufficient to meet this financial risk criterion. However, we believe that financial risk for monetary losses under an APM must be tied to performance under the model as opposed to indirect losses related to financial investments APM Entities may make. The amount of financial investment made by APM Entities may vary widely and may also be difficult to quantify, resulting in uncertainty regarding whether an APM Entity had exceeded the nominal amount required by statute. In addition to the difficulty in creating an objective and enforceable standard for determining whether an entity's business risk associated with the Advanced APM exceeds a nominal amount, we strongly believe that the statutory scheme under section 1833(z) of the Act recognizes that not all APMs will meet this criterion. We do not intend for our proposal to diminish the substantial time and money commitments in which APM Entities invest in order to become successful participants. We welcome comments on how we could potentially create an objective and meaningful financial risk criterion that would define financial risk for monetary losses based on performance under the APM differently.

(ii) Medical Home Model Standard

Second, we propose to adopt a slightly different financial risk standard for Medical Home Models. For a Medical Home Model to be an Advanced APM, it must include provisions that potentially:

• Withhold payment for services to the APM Entity and/or the APM Entity's eligible clinicians;
• Reduce payment rates to the APM Entity and/or the APM Entity's eligible clinicians;
• Require the APM Entity to owe payment(s) to CMS; or
• Lose the right to all or part of an otherwise guaranteed payment or payments, if either:
• Actual expenditures for which the APM Entity is responsible under the APM exceed expected expenditures during a specified performance period; or
• APM Entity performance on specified performance measures does not meet or exceed expected performance on such measures for a specified performance period.

With regard to the proposed financial risk standard for Medical Home Models, we believe that the Medical Home Model is a unique type of APM that is treated differently under both the MIPS and APM programs. For example, under the MIPS clinical practice improvement activity performance category, as described in section II.E.3.f of this preamble of this proposed rule, eligible clinicians participating in medical homes receive an automatic 100 percent score, whereas eligible clinicians participating in other APM Entities receive a minimum of a 50 percent score. Additionally, both Medical Home Models and Medicaid Medical Homes Models are distinct from other APMs in that, if they are models tested under section 1115A of the Act, there is the possibility of having an alternate pathway through expansion under section 1115A(c) of the Act to meet the financial risk criterion, and Medicaid Medical Home Models play a role in whether Medicaid payments or patients are excluded in the All-Payer Combination Option for QP determinations (see sections 1833(z)(2)(B)(ii)(I)(bb) and (iii)(II)(cc)(BB), 1833(z)(2)(C)(ii)(I)(bb) and (iii)(II)(cc)(BB), 1833(z)(3)(C)(ii)(II), and 1848(q)(5)(C)(i) of the Act). Medical Home Models and their APM Entities (medical homes) are different from other APMs in that: (1) Medical homes tend to be smaller in size and have lower Medicare revenues relative to total Medicare spending than other APM Entities, which affects their ability to bear substantial risk, especially in relation to total cost of care; and (2) to date, neither publicly nor commercially-sponsored medical homes have been required to bear the risk of financial loss, which means the assumption of any financial risk presents a new challenge for medical homes. For example, a common group practice in the Comprehensive Primary Care (CPC) initiative may consist of less than twenty individuals, including physicians, non-physician practitioners, and administrative staff. Making large lump sum loss payments or going without regular payment for a substantial period of time could put such practices out of business, whereas large ACOs may comprise an entire integrated delivery system with sufficient financial reserves to weather direct short-term losses.

We therefore believe that the unique characteristics of Medical Home Models warrant the application of a financial risk standard that reflects these differences in order to provide incentives for participation in the most advanced financial risk arrangements available to medical homes practitioners.

The proposed financial risk standard for Medical Home Models is similar to the generally applicable Advanced APM standard in its first three conditions. The difference is in the inclusion of the fourth condition for the proposed financial risk standard for Medical Home Models, which would allow a performance-based forfeiture of part of all of a payment under an APM to be considered a monetary loss. For example, a Medical Home Model would meet this standard if it conditions the payment of some or all of a regular care management fee to APM Entities upon meeting specified performance standards. Because the APM does not require any direct payment or repayment to CMS, a medical home penalized in such a manner would not necessarily be in a weaker financial position than it had been prior to the decreased payment; however, it would be in a comparatively worse position in the future than it otherwise would have been had it met performance standards. We believe that this financial risk standard respects the unique challenges of medical homes in bearing risk for losses while maintaining a more rigorous standard than business risk.

We seek comment on the proposed standards set forth for both Advanced APM Medical Home Models and for all other APMs. We would consider any comments on alternative standards suggested by the public that could achieve our stated goals and the statutory requirements. We also seek comment on types of financial risk arrangements that may not be clearly captured in this proposal.

(4) Nominal Amount of Risk

If the APM risk arrangement meets the proposed financial risk standard, we would then consider whether the amount of the risk is in excess of a nominal amount in order for this Advanced APM criterion to be met. We believe the statutory requirement that an APM Entity bear risk under an APM in excess of a nominal amount (which we will term the “nominal amount standard”) relates to a particular quantitative risk value at which CMS would consider the risk arrangement to involve potential losses of more than a nominal amount. Similar to the financial risk portion of this assessment, we propose to adopt a generally applicable nominal amount standard for Advanced APMs and a unique nominal amount standard for Medical Home Models. Under the generally applicable nominal amount standard, the total risk percentages are of the APM Entity benchmark or, in the case of episode payment models, the target price, which is the amount of Medicare expenditures (which can vary as to the involvement of Parts A and B depending on the APM) above which an APM Entity owes losses and below which an APM Entity earns savings. In the case of Medical Home Models, the risk percentages for Medical Home Models are based on Medicare Parts A and B revenue. As an alternative, we considered assessing total risk under the generally applicable nominal amount standard (for APM other than episode payment models) in relation to the APM Entity's Parts A and B revenue instead of in relation to the APM benchmark. We note that the ratio between entity revenue and the expenditures reflected in an APM's benchmark may vary across different types of entities, such as when the APM benchmark is based on total cost of care. However, we are not proposing the alternative of basing the generally applicable standard on Parts A and B revenue because that policy would prevent a general determination that an APM meets such standards. Instead, it would require case-by-case determinations at the APM Entity level that could change from year to year. We are also concerned that assessing total risk based on an APM Entity's revenue instead of the APM benchmark would set meaningfully different standards for different types of entities regarding the extent to which they must be held financially responsible if expenditures exceed the benchmark. In general, we believe we should apply a common standard to all types of entities. That being said, we understand that setting the total risk standard too high could create challenges for smaller organizations for which a total cost of care benchmark represents more risk in relation to revenue than it does for larger organizations.

(a) Advanced APM Nominal Amount Standard

In general, we believe that the meaning of “nominal” is, as plain language implies, minimal in magnitude. However, in the context of financial risk arrangements, we do not believe it to be a mere formality. For instance, we do not believe the law was intended to consider one dollar of risk to be more than nominal. That would create an arbitrary distinction between an APM that has only upside reward potential and one that has the same upside reward potential with a fractional and relatively meaningless downside risk. Therefore, in arriving at the proposed values, we sought amounts that would be meaningful for the entity but not excessive. As reference points to anchor the proposed values, we used the percentage amounts of MIPS adjustments in the MACRA and surveyed current APM risk arrangements, including those in Tracks 2 and 3 of the Shared Savings Program, the Pioneer ACO Model, and the Bundled Payments for Care Improvement (BPCI) Initiative. We consider the potential losses and marginal risk rates of those initiatives to be optimal in that they have been vetted through the APM development process and determined to be the appropriate amount of risk for each initiative such that, in the context of the APM, it is anticipated that the amount of risk would motivate the desired changes in care patterns in order to reduce costs and improve quality. As stated above, we believe that the term “nominal” is clearly an amount that is lower than optimal but substantial enough to drive performance. In other words, we are confident that risk levels in current APMs with downside risk are sufficient for a wide variety of providers and suppliers, but in certain circumstances, we would want to encourage participation in APMs with slightly lower levels of risk, though not levels of risk that are so low that an APM becomes no more effective at motivating desired changes than APMs with no downside risk.

Except for risk arrangements described under section II.F.4.b.(4) of this preamble, we propose to measure three dimensions of risk described in this section to determine whether an APM meets the nominal amount standard: (a) Marginal risk, which is a common component of risk arrangements—particularly those that involve shared savings—that refers to the percentage of the amount by which actual expenditures exceed expected expenditures for which an APM Entity would be liable under the APM; (b) minimum loss rate (MLR), which is a percentage by which actual expenditures may exceed expected expenditures without triggering financial risk; and (c) total potential risk, which refers to the maximum potential payment for which an APM Entity could be liable under the APM. Except for risk arrangements described under section II.F.4.b.(3) of this preamble, we propose that for a APM to meet the nominal amount standard the specific level of marginal risk must be at least 30 percent of losses in excess of expected expenditures, and a minimum loss rate, to the extent applicable, must be no greater than 4 percent of expected expenditures, and total potential risk must be at least 4 percent of expected expenditures. As described in greater detail in section II.F.7 of this preamble, the proposed Other Payer Advanced APM nominal risk standard parallels the standard described here for Advanced APMs. In general, we define expected expenditures to be the level of expenditures reflected in the APM benchmark. However, for episode payment models, we defined expected expenditures to be the level of expenditures reflected in the target price.

To determine whether an APM satisfies the marginal risk portion of the nominal risk standard, we would examine the payment required under the APM as a percentage of the amount by which actual expenditures exceeded expected expenditures. We propose that we would require that this percentage exceed the required marginal risk percentage regardless of the amount by which actual expenditures exceeded expected expenditures. APM arrangements with less than 30 percent marginal risk would not meet the nominal risk standard. We believe that meaningful risk arrangements can be designed with marginal risk rates of greater than 30 percent. Any marginal risk below 30 percent creates scenarios in which the total risk could be very high, but the average or likely risk for an APM Entity would actually be very low. We also propose that the payment required by the APM could be smaller when actual expenditures exceed expected expenditures by enough to trigger a payment greater than or equal to the total risk amount required under the nominal risk standard (as specified in Table 28). This is essentially an exception to the marginal risk requirement so that the standard does not effectively require APMs to incorporate total risk greater than the amount required by the total risk portion of the standard.

An example of marginal risk is the sharing rate in the Shared Savings Program. For instance, an ACO in Track 2 or Track 3 of the Shared Savings Program that has a sharing rate, or marginal risk, of 50 percent and exceeds its benchmark (expected expenditures) by $1 million would be liable for $500,000 of those losses. The inclusion of a marginal risk standard is intended to focus on maintaining a more than nominal level of average or likely risk under an Advanced APM. For instance, a APM with a large (for example, 20 percent of benchmark) total potential risk could have a very small (for example, 10 percent) sharing rate as its marginal risk, which substantially mitigates the amount of loss the APM Entity would reasonably expect to incur. We believe that including marginal risk in the Advanced APM financial risk criterion clarifies for APM Entities and eligible clinicians the type of risk they must bear should they pursue becoming QPs. Focusing on marginal risk in the proposed criterion for Other Payer Advanced APMs in section II.F.7.b.(6) of this preamble additionally acts as a guard against gaming through strategic development of risk arrangements with very low marginal risk.

We propose a maximum allowable “minimum loss rate” (MLR) of 4 percent in which the payment required by the APM could be smaller than the nominal amount standard would otherwise require when actual expenditures exceed expected expenditures by less than 4 percent; this exception accommodates APMs that include zero risk with respect to small losses but otherwise satisfy the marginal risk standard. If actual expenditures exceed expected expenditures by an amount exceeding the MLR, then all excess expenditures (including excess expenditures within the MLR) would be subject to the marginal risk requirements. For example, ACOs participating in performance-based risk arrangements under Tracks 2 and 3 of the Shared Savings Program are permitted to choose their own minimum savings rate (MSR) and MLR as long as they are symmetrical. If losses do not exceed the chosen MLR, the ACO is not held responsible for losses. If the ACO has a very large MLR, there may be little to no risk with respect to losses below a certain percentage of the benchmark. Therefore, we believe that proposing a maximum allowable MLR is appropriate. We recognize that there may be instances where an APM can satisfy the marginal risk portion of the nominal risk standard even with a high MLR. Therefore, we also propose a process through which CMS could determine that a risk arrangement with an MLR higher than 4 percent could meet the nominal amount standard, provided that the other portions of the nominal risk standard are met. In determining whether such an exception would be appropriate, CMS would consider: (1) Whether the size of the attributed patient population is small; (2) whether the relative magnitude of expenditures assessed under the APM is particularly small; and (3) in the case of a test of limited size and scope, whether the difference between actual expenditures and expected expenditures would not be statistically significant even when actual expenditures are 4 percent above expected expenditures. We note that CMS would grant such exceptions rarely, and CMS would expect APMs considered for such exceptions to demonstrate that a sufficient number of APM Entities are likely to incur losses in excess of the higher MLR. In other words, the potential for financial losses based on statistically significant expenditures in excess of the benchmark must remain meaningful for participants.

To determine whether an APM satisfies the total risk portion of the nominal risk standard, we would identify the maximum potential payment an APM Entity could be required to make as a percentage of expected expenditures under the APM. If that percentage exceeded the required total risk percentage, then the model would satisfy the total risk portion of the nominal amount standard.

In evaluating both the total and marginal risk portions of the nominal amount standard, we would not include any payments the APM Entity or its eligible clinicians would make to CMS under the APM if actual expenditures exactly matched expected expenditures. In other words, payments made to CMS outside the risk arrangement related to expenditures would not count toward the nominal risk standard. This requirement ensures that perfunctory or pre-determined payments do not supersede incentives for improving efficiency. For example, an APM that simply requires an APM Entity to make a payment equal to 5 percent of the APM benchmark at the end of the year, regardless of actual expenditure performance, would not satisfy the nominal amount standard.

We believe that this approach to measuring the amount of risk flexibly accommodates a wide variety of risk structures, including APMs in which marginal risk varies with the amount of losses. For example, an APM could have a sharing rate of 75 percent for expenditure amounts that exceed the benchmark by up to 2 percent and a sharing rate of 50 percent for expenditure amounts that exceed the benchmark by 2 percent or more. Because the smallest sharing rate is 50 percent, the marginal risk rate exceeds 30 percent at all levels of expenditures, so the model satisfies the marginal risk portion of the nominal amount standard. Because this hypothetical APM does not have MLR or stop loss provisions, it satisfies the total risk and MLR portions of the nominal amount standard.

In particular, the financial risk an Advanced APM Entity would bear under an Advanced APM need not take a shared savings structure in which the financial risk increases smoothly based on the amount by which an Advanced APM Entity's actual expenditures exceed expected expenditures. An example of a risk arrangement being based on shared savings is Tracks 2 and 3 of the Shared Savings Program, where the greater the losses in relation to the expenditure benchmark, the greater the potential amount of shared losses an ACO would be required to repay CMS. On the other hand, an Advanced APM could require APM Entities to pay a penalty based on expenditure targets, regardless of the degree to which the APM Entity actually exceeded those expenditure targets, provided that the payments are otherwise structured in a way that satisfies both the marginal and total risk requirements under the nominal amount standard.

We seek comment on appropriate levels for the allowable minimum loss rate and the parameters we should consider when determining whether a risk arrangement should warrant an exception from the minimum loss rate portion of the nominal amount standard.

Table 28 summarizes the generally applicable nominal amount standard. Tables 29 and 30 provide examples of types of risk arrangements that would and would not meet the financial risk criterion. The examples are divided between shared savings-style arrangements in which marginal risk is a component and non-shared savings arrangements.

Figures C and D illustrate types of payment arrangements would meet the nominal amount standard. Figure C represents the minimum nominal amount standard, so any APM in which the risk for required payments would be on or above the line would satisfy the nominal amount standard. Figure D represents an example of a risk arrangement that would exceed the nominal amount standard.

We seek comment on the Advanced APM nominal amount standard. In particular, we seek comment on whether the Advanced APM benchmark or the Advanced APM Entity revenue is a more appropriate basis for assessing total risk and on the proposed amounts of total potential risk, marginal risk, and maximum allowable minimum loss rate. In particular, we seek comment on whether 30 percent is a sufficient level of marginal risk to be considered “more than nominal.” We also seek comment on whether CMS could adopt a meaningful standard that only includes total and marginal risk without the minimum loss rate component. Finally, we seek comment on a tiered nominal risk structure in which different levels of marginal risk could be paired with different levels of total risk.

In commenting on possible alternatives, we encourage commenters to refer to the policy principles articulated in section II.F.1 and to consider the extent to which their proposed alternatives would be more or less consistent with those principles.

(b) Medical Home Model Standard

We propose that for Medical Home Models, the total annual amount that an Advanced APM Entity potentially owes CMS or foregoes under the Medical Home Model must be at least the following amounts in a given performance year:

• In 2017, 2.5 percent of the APM Entity's total Medicare Parts A and B revenue;
• In 2018, 3 percent of the APM Entity's total Medicare Parts A and B revenue.
• In 2019, 4 percent of the APM Entity's total Medicare Parts A and B revenue.
• In 2020 and later, 5 percent of the APM Entity's total Medicare Parts A and B revenue.

We believe the statute's explicit discussion of medical homes gives us unique latitude to separately set financial risk and nominal amount standards for Medical Home Models that fall below an amount we consider sufficient to be “more than nominal” in the context of other types of APMs. We also believe that the meaning of the term “nominal” depends on the situation in which it is applied, so we believe it is appropriate to consider the characteristics of the medical home class of APM Entities in setting the nominal amount standard for Medical Home Models. As we noted in discussing the financial risk standard, few medical homes have had experience with financial risk, and many would be financially unable to provide sufficient care or even remain a viable business in the event of substantial disruptions in revenue. As such, we believe we should base the nominal amount standard on the APM Entity's total Medicare Parts A and B revenues and also not include a potentially excessive level of risk for such entities in the first year of the program. Thus, our proposal sets forth a gradually increasing but achievable long-term amount of risk that would apply in subsequent years. In general, we believe that this scheme allows Medical Home Models to craft incentive designs that allow medical homes to succeed through care transformation and the provision of high-value care while not threatening the ability of small practices to function.

Some benchmarks are based on total cost of care, and, as discussed with respect to the generally applicable nominal amount standard, we generally believe that the APM benchmark or target price is the appropriate basis for evaluating the nominal amount standard. However, we note that, for a small practice, the benchmark can be an amount that is significantly greater than the practice's revenue from all payment sources. Thus, basing the Medical Home Model nominal amount standard on percentage of risk in relation to a total cost of care benchmark would mean that certain types of entities would be required to bear greater total risk in relation to their revenues than other entities, which we believe would be undesirable in light of the special characteristics of Medical Home Models. On the other hand, most APMs base risk on the benchmark instead of revenue, and using revenue as the basis for determining the nominal risk standard could cause the APM Entity's eligibility to vary from year to year based on changes in an APM entity's revenue despite the core risk arrangement remaining unchanged.

For the Medical Home Model nominal amount standard, we seek additional comment on the length of the proposed multi-year “ramp up period” and the magnitude of the total risk amounts during such a period. We also seek comment on the potential addition of a marginal risk amount to the extent applicable and on whether the Advanced APM benchmark or Advanced APM Entity revenue is the most appropriate standard for measuring total risk.

In commenting on possible alternatives, we encourage commenters to refer to the policy principles articulated in section II.F.1 and to consider the extent to which their proposed alternatives would be more or less consistent with those principles.

(5) Capitation

We propose that full capitation risk arrangements would meet the Advanced APM financial risk criterion. We propose that, for purposes of this rulemaking, a capitation risk arrangement means a payment arrangement in which a per capita or otherwise predetermined payment is made to an APM Entity for all items and services furnished to a population of beneficiaries, and no settlement is performed for the purpose of reconciling or sharing losses incurred or savings earned by the APM Entity. We also would like to reiterate that—in line with statute—Medicare Advantage and other private plans paid to act as insurers on the Medicare program's behalf are not Advanced APMs.

We believe that capitation risk arrangements, as defined here, involve full risk for the population of beneficiaries covered by the arrangement, recognizing that it might require no services whatsoever or could require exponentially more services than were expected in calculating the capitation rate. The APM Entity bears the full downside and upside risk in this regard. Thus, we believe capitation arrangements inherently require an APM Entity to bear financial risk for monetary losses in excess of a nominal amount. We propose that, where payment is made to participating entities in a APM using a capitation risk arrangement, the APM and participating entities would meet the criterion under section 1833(z)(3)(D)(ii)(I) of the Act.

In implementing this proposed policy, it is important to distinguish capitation as a risk arrangement from capitation as only a cash flow mechanism. A capitation risk arrangement adheres to the idea of a global budget for all items and services to a population of beneficiaries during a fixed period of time. Cash flow mechanisms that make payments in predetermined amounts that are later reconciled or adjusted based on actual services are not necessarily a full risk arrangement. For example, an APM Entity has a capitation arrangement under an APM that pays $1,000 per beneficiary per month for a population of 100 beneficiaries, totaling $1.2 million per year. If expenditures for services actually furnished to these beneficiaries would have totaled $1.3 million if paid on a fee-for-service basis, a payment mechanism without risk might make a reconciliation payment of $100,000 to the entity. In that case, the APM Entity is not bearing any financial risk for monetary losses under the APM. If there is partial reconciliation, the arrangement would not meet the proposed capitation risk arrangement definition but still may meet the financial risk and nominal amount standards through the assessments described in this section above. In contrast, if this arrangement is a capitation risk arrangement, there would be zero reconciliation for those losses. Under our proposal, we would categorically accept that a capitation risk arrangement under an APM would meet the Advanced APM financial risk criterion.

We seek comment on our proposal for acceptance of capitation risk arrangements and on our proposed definition of a capitation risk arrangement. We also seek comment on other types of arrangements that may be suitable for such treatment for purposes of this financial risk criterion. Finally, we seek comment on potential limits or qualifications to the capitation standard in order to prevent potential abuse or incentives that are not consistent with the provision of high value care.

(6) Medical Home Expanded Under Section 1115A(c) of the Act

Section 1833(z)(3)(D)(ii)(II) of the Act states that an Advanced APM must either meet the financial risk criterion or be a Medical Home Model expanded under section 1115A(c) of the Act. We will refer to the latter criterion as the expanded Medical Home Model criterion. We propose that a Medical Home Model that has been expanded under section 1115A(c) of the Act would meet the expanded Medical Home Model criterion and thus would not need to meet the Advanced APM financial risk criterion as described above. Under this this proposal, an APM would have to both be determined to be a Medical Home Model as defined in this rulemaking and in fact be expanded using the authority under section 1115A(c) of the Act. Such expansion is contingent upon whether, for an APM tested under section 1115A(b) of the Act:

• The Secretary determines that such expansions is expected to reduce spending under the applicable title without reducing the quality of care; or improve the quality of patient care without increasing spending;
• CMS' Chief Actuary certifies that such expansion would reduce (or would not result in any increase in) net program spending under the applicable titles; and
• The Secretary determines that such expansion would not deny or limit the coverage or provision of benefits under the applicable title for applicable individuals. In determining which models or demonstration projects to expand under the preceding sentence, the Secretary shall focus on models and demonstration projects that improve the quality of patient care and reduce spending.

We note that the expanded Medical Home Model criterion cannot met unless a Medical Home Model has been expanded under section 1115A(c). Merely satisfying expansion criteria would not be sufficient to meet this Advanced APM criterion. This expanded Medical Home Model criterion is directly related to a similar criterion addressed in this proposed rule for Medicaid Medical Home Models, which addresses how such APMs can meet the Other Payer Advanced APM financial risk criterion by having criteria comparable to an expanded Medical Home Model. We request comments on the proposed requirements for this and all proposed Advanced APM criteria.

(7) Application of Criteria to Current and Recently Announced APMs

Using the Advanced APM criteria proposed in sections II.F.4.b.1-6 of this preamble, we have identified the current APMs that we anticipate would be Advanced APMs for the first QP Performance Period. We note that since no CMS Medical Home APMs have been expanded under section 1115A(c) of the Act, we have not included this criterion in the table.

The information presented in Table 32 is based on the preliminary application of proposed Advanced APM criteria in this preamble and does not preclude any changes to the list based on: (1) Any changes made to the proposed criteria in the publication of the final rule in response to public comments; (2) any modifications to the design of current APMs; or (3) any new APMs announced between publication of this proposed rule and the beginning of the first QP Performance Period. Consistent with our proposal in section II.F.4.a, we propose to post an official determination of which APMs would meet the final Advanced APM criteria prior to the beginning of the first QP Performance Period and update that list in accordingly.

We note that the Comprehensive Care for Joint Replacement (CJR) model does not meet the Advanced APM criteria proposed in sections II.F.4.b.1-6 of this preamble. We seek comment on how we might change the design of CJR through future rulemaking to make it an Advanced APM, and we seek comment on how to include eligible clinicians in CJR for purposes of the QP determination as described in section II.F.5.

5. Qualifying APM Participant (QP) and Partial QP Determination

The QP determination process is specified under section 1833(z)(2) of the Act, in which QPs are defined as those eligible clinicians who meet the specified threshold(s).

In this section, we propose a process for determining which eligible clinicians would be QPs or Partial QPs for a given payment year through their participation in Advanced APMs during a corresponding QP Performance Period. Per sections 1833(z)(2) and 1848(q)(1)(C)(ii)(I) and (II) of the Act, an eligible clinician would become a QP or Partial QP for a payment year if they are determined at the end of the performance period to be eligible clinicians in an Advanced APM Entity that collectively meets the threshold values for participation in an Advanced APM during the corresponding QP Performance Period, and starting in 2021, the threshold values for participation in an Other Payer Advanced APMs as proposed here. Each year, CMS would determine whether an eligible clinician achieved the threshold to become a QP or Partial QP during the corresponding QP Performance Period. CMS would make this assessment independent of QP or Partial QP determinations made in previous years and accounting for Advanced APMs that begin or end on timeframes that do not align precisely with the QP Performance Period. The following would apply to an eligible clinician whom CMS determines to be a QP for a particular year:

• For payment years 2019-2024, the QP will receive a lump sum payment equal to 5 percent of the estimated aggregate payment amounts for Medicare Part B covered professional services for the prior year, as described in section II.F.8 of this preamble;
• The QP will be excluded from MIPS payment adjustments, as described in section II.E.3 of this preamble; and
• For payment years 2026 and later, payment rates under the Medicare physician fee schedule for services furnished by the eligible clinician will be updated by the 0.75 percent qualifying APM conversion factor as specified in sections 1848(d)(1)(A) and (d)(20) of the Act.

Through the APM Entity group determination described in section II.F.5.b of this preamble, CMS would identify eligible clinicians who do not meet the QP threshold but reach the Partial QP threshold for a year to be Partial QPs. Partial QPs would not be eligible for the 5 percent APM Incentive Payment for years from 2019 through 2024 or, beginning for 2026, the qualifying APM conversion factor. However, as described below, Partial QPs would have an opportunity to decide whether they wish to be subject to a MIPS payment adjustment, which could be positive or negative.

The statute requires that we use two options to determine whether an eligible clinician is a QP or Partial QPs for a payment year—one is the Medicare Option and, beginning in 2021, the other is the All-Payer Combination Option. While these are the terms based on statutory language that we have chosen to use for the purposes of describing the process by which we can calculate an eligible clinician's Threshold Score, we note that the use of the word “option” does not imply that an eligible clinician will have the ability to choose between the two. We further outline in this section our proposed process by which we will assess eligible clinicians under both options (beginning in 2021) to the extent that sufficient data is submitted to CMS.

The Medicare Option, described in this section, focuses on participation in Advanced APMs, and CMS would make determinations under this option based on Medicare Part B covered professional services attributable to services furnished through an Advanced APM Entity. The Medicare Option is the only option available for QP determinations during the first two years of this program (payment years 2019-2020). The All-Payer Combination Option, described in section II.F.7 of this preamble, is applicable beginning in the third payment year (2021) and would allow CMS to make determinations based on participation in both Advanced APMs and Other Payer Advanced APMs. The All-Payer Combination Option would not replace or supersede the Medicare Option; instead it would allow eligible clinicians to become QPs by meeting a relatively lower threshold based on Medicare Part B covered professional services through Advanced APMs and an overall threshold based on services through both Advanced APMs and Other Payer Advanced APMs. With our proposals for the QP Threshold Score methodologies described in this section, we generally interpret payments “through” an Advanced APM Entity to mean payments made by CMS for services furnished to attributed beneficiaries, who are the beneficiaries for whose costs and quality of care an Advanced APM Entity is responsible under the Advanced APM. Under section 1848(q)(1)(C)(iii) of the Act, the calculations used for Partial QP determinations are the same, but the threshold percentages to be a Partial QP for each year are lower than those required to be a QP.

The QP and Partial QP Thresholds under the Medicare Option are shown in Tables 33 and 35. The QP and Partial QP Threshold values under the All-Payer Combination Option are shown in Tables 34 and 36. CMS will determine an eligible clinician's QP status for a payment year by calculating an eligible clinician's Threshold Score, and comparing the eligible clinician's Threshold Score (either based on payment amounts or patient counts) to the relevant QP Threshold or Partial QP Threshold. In addition, we discuss our proposal to make QP determinations at a group level based on an entire Advanced APM Entity in section II.F.5.b of this preamble.

According to section 1833(z)(2)(D) of the Act, the Secretary may base the determination of whether an eligible clinician is a QP or a Partial QP by using counts of patients in lieu of using payment amounts and using the same or similar percentage criteria as those used for the payment amount method, as the Secretary determines is appropriate. For QP and Partial QP determinations using patient count calculations, we propose to use the percentage values displayed in Tables 35 and 36. The purpose of the proposed design of the Medicare patient count method is to make QP status determinations accessible to entities and individuals who are clearly and significantly engaged in delivering value-based care through participation in Advanced APMs. We also propose that when determining whether to use the payment amounts or patient counts method to calculate the QP threshold status, CMS will use both methods in tandem for each Advanced APM Entity group of eligible clinicians. We further propose that after QP and Partial QP threshold calculations have been completed, we will use the QP threshold method that is more favorable to the Advanced APM Entity group of eligible clinicians.

By performing preliminary analyses using our proposed QP determination methodologies with historical APM data, we found that the proposed QP and Partial QP Patient Count Thresholds are similar in magnitude and trajectory to those specified in the statute for the payment-based calculations. Due to varying attribution and organizational characteristics, we anticipate that using our proposed thresholds, the method—payment amount or patient count—that results in the most favorable QP status will likely vary across different Advanced APMs and Advanced APM Entities. We believe that each eligible clinician should have every opportunity to reach the QP threshold for each year, and do not intend to limit this opportunity by preemptively selecting one method over another.

We seek comment on the proposed QP Patient Count Threshold and Partial QP Patient Count Threshold percentage values for both the Medicare Option and the All-Payer Combination Option, on our proposal to calculate the Threshold Score under the payment amount and patient count methods simultaneously, and on our proposal to use the method that is most favorable to the Advanced APM Entity group of eligible clinicians.

We propose that, beginning with payment year 2021, CMS will conduct the QP determination sequentially so that the Medicare Option is applied before the All-Payer Combination Option. We propose to apply the All-Payer Combination Option only to an Advanced APM Entity group of eligible clinicians or eligible clinicians who do not meet either the QP Payment Amount or Patient Count Threshold under the Medicare Option but who do meet the lower Medicare threshold for the All-Payer Combination Option. This process is illustrated in Figures E and F, which show that the first assessment is whether the Medicare QP Threshold has been met under either the Medicare Option or the All-Payer Combination Option.

Because the Medicare Option (either based on payment amounts or patient counts) is also part of the All-Payer Combination Option, and because all eligible clinicians must reach at least a minimum Medicare Threshold Score through Advanced APMs to be QPs, we believe that this sequential approach streamlines the analytic and operational requirements to make QP determinations under the All-Payer Combination Option. Figure E illustrates the proposed process for making QP determinations under the Medicare Option for 2019 and 2020. Figure F illustrates the process proposed for making QP determinations under both the Medicare and All-Payer Combination Options for payment years 2021-2024. Figure G provides an example of the proposed process for making QP determinations in payment years 2023-2024. Figures E, F, and G only discuss the payment amount method, but a similar process would apply for the patient count method.

a. QP Performance Period

According to section 1833(z)(2) of the Act, we are required to determine QP and Partial QP status based on payment amounts (or patient counts) during the most recent period for which data are available (which may be less than a year). We propose that the QP Performance Period is the full calendar year that aligns with the MIPS performance period (for instance, 2017 would be the QP Performance Period for the 2019 payment year). We believe that having a QP Performance Period parallel with the proposed MIPS performance period offers will reduce operational complexity and gives CMS the opportunity to clearly communicate an eligible clinician's status in this program throughout the process. We also believe that having a QP Performance Period that concludes one year and one day before the payment year enables CMS to provide all eligible clinicians participating in Advanced APMs the best opportunity to monitor their performance through the Advanced APM and make the most informed decisions regarding their decision whether to not to be subject to MIPS in the event that they become a Partial QP. We seek comment on this proposal and any alternative QP Performance Period timeframes that would both enable meaningful QP assessment and ensure operational alignment with MIPS.

b. Group Determination and Lists

(1) Group Determination

The statute consistently refers to an eligible clinician throughout section 1833(z) of the Act and clearly identifies that the QP determinations are to be made for an eligible clinician. In section 1833(z)(3)(B) of the Act, the definition of an eligible clinician includes a group of such professionals. We received several comments to our MIPS and APMs RFI recommending that CMS make QP determinations at a group level and indicating a preference for entity cohesion over a highly precise analysis for individual eligible clinicians. Commenters stated a number of reasons why they recommended that QP determinations should be made at the group level. These reasons included promoting administrative simplicity, the need to foster collaboration among group members (instead of promoting barriers), and the fact that while many beneficiaries are attributed to an APM Entity based on the services rendered by one eligible clinician, many of the eligible clinicians participating in the APM Entity may play a role in the actual diagnosis, treatment, and management of many beneficiaries in the APM Entity population. Each of these individual eligible clinicians could potentially view themselves as being instrumental in providing quality care to the beneficiary that is in line with the objectives of the APM, regardless of whether their individual services are counted towards APM-specific attribution methods. A few commenters indicated that the Advanced APM Entities themselves should determine whether individual eligible clinicians meet the annual threshold to become a QP.

An Advanced APM Entity faces the risks and rewards of participation in an Advanced APM as a single unit, and is responsible for performance metrics that are aggregated to the level of that entity. This policy is also based on the premise that positive change occurs when entire organizations commit to participating in an Advanced APM and focusing on its cost and quality goals as a whole. It also mitigates situations in which individual eligible clinicians who practice together in an Advanced APM Entity receive different QP determinations and thus are treated differently for purposes of APM Incentive Payments, MIPS payment adjustments, and eventually, differential fee schedule updates under the PFS. We believe that such discrepancies could potentially lead to confusion and lack of cohesion among eligible clinicians and Advanced APM Entities and place additional burdens on eligible clinicians and organizations to track these differences. Additionally, we wish to avoid any additional burden, confusion, and operational difficulties for both eligible clinicians and CMS that would result from allowing eligible clinicians or Advanced APM Entities to elect whether to be assessed at the Advanced APM Entity level. We believe that a simple, overarching rule is preferable to adding extra variables to the already complex processes under this program.

We understand that, as with any group assessment, there will be some situations in which individual Threshold Scores would differ from group Threshold Scores if assessed separately. This could lead to some eligible clinicians becoming QPs when they would not have met the QP Threshold individually (a “free-rider” scenario) or, conversely, some eligible clinicians not becoming QPs within an Advanced APM Entity when they might have qualified individually (a dilution scenario). We believe that through the methodology we propose for QP determination in this proposed rule, the magnitude of such discrepancies will be relatively small compared to the value of maintaining Advanced APM Entity cohesion.

We propose, except in the specific situations discussed below in this section, to make the QP determination at a group level. As a result, the QP determination for the group would apply to all the individual eligible Clinicians who are identified as part of an Advanced APM Entity. If that eligible Clinician group's collective Threshold Score meets the relevant QP threshold, all eligible Clinicians in that group would receive the same QP determination for the relevant year. The QP determination calculations described in this proposed rule would be aggregated using data for all eligible clinicians participating in the Advanced APM Entity during the QP Performance Period.

In some cases, the list of eligible clinicians who will be grouped together for purposes of the QP determination may include eligible clinicians who have relationships with the Advanced APM Entity but no relationship with each other. We believe this is appropriate for purposes of the QP determination because it support the Advanced APM Entity as the coordinator of its participating eligible clinicians to contribute to its success and promotes eligible clinician coordination when appropriate to further the success of the Advanced APM Entity.

(2) Groups Used for QP Determination

We propose that the group of eligible clinicians would consist of all the eligible clinicians identified as participants in an Advanced APM Entity during the QP Performance Period on a Participation List provided to CMS, with one exception for Advanced APMs whose participants are not eligible clinicians. We propose to define participant for the purposes of participation in an APM as an entity participating in an APM under an agreement with CMS or statute or regulation that may either include eligible clinicians or be an eligible clinician and that is directly tied to beneficiary attribution, quality measurement or cost measurement under the APM. This definition encapsulates those entities and eligible clinicians under an APM who have roles of central importance to performance under the APM. We propose that the Participation List for each Advanced APM Entity would be compiled from CMS-maintained lists that will be used to identify each eligible clinician by a unique TIN/NPI combination attached to the identifier of the Advanced APM Entity. Therefore, an eligible clinician must be officially identified using an Advanced APM Entity's Participation List to be part of the QP determination for that group.

In APMs, the APM Entity that has an agreement with CMS or is identified as such under statute or regulation is considered a participant in the APM. Some APMs have eligible clinicians under the APM Entity who are also under our definition considered participants in the Advanced APM Entity. For example, in an APM like the Comprehensive Primary Care Initiative with physician group practices as participants, the APM Entity, the Practice, may have a Participation List it provides to CMS that can be used to identify each eligible clinician participant participating in the APM through that APM Entity by a unique TIN/NPI combination attached to the identifier of the APM Entity. As stated above, we propose to include of all the eligible clinicians identified using a Participation List as participants in an Advanced APM Entity during the QP Performance Period for purposes of the QP determination.

In certain APMs, a Participation List may not include any eligible clinicians. For example, in an APM where all APM Entities are hospitals, the APM Entity will not have eligible clinicians identified by a unique TIN/NPI combination attached to the identifier of the Advanced APM Entity on a Participation List because there will not be eligible clinicians who are participants under the APM Entity. An Advanced APM Entity may have a list of entities, including eligible clinicians, who are affiliated with and support the Advanced APM Entity in its participation in the Advanced APM, but are not participants and are therefore not on a Participation List. For example, a list of gainsharers under an APM might include eligible clinicians where the Participation List does not.

Where there is a Participation List that can be used to identify eligible clinicians, we propose that it automatically be the list that is considered for the QP Determination. Where there is no Participation List that can be used to identify eligible clinicians, but there is another list of eligible clinicians who have a contractual relationship with the Advanced APM Entity based at least in part on supporting the Advanced APM Entity's quality or cost goals under the Advanced APM (Affiliated Practitioners), we propose to use the list of those eligible clinicians, the Affiliated Practitioner List, for purposes of the QP determination. Where there is both a Participation List and an Affiliated Practitioner List that can be used to identify eligible clinicians under an Advanced APM, we propose only to use the Participation List for purposes of the QP determination. We seek comment on whether to limit the proposed policy to use an Affiliated Practitioner List for the QP Determination to the Medicare payment threshold option, as it may be less likely that Affiliated Practitioners support the Advanced APM Entity as a group in Other Payer Advanced APMs than eligible clinicians on a Participation List.

This proposed policy was developed to capture the group or groups of eligible clinicians who are the most closely associated with the performance of the Advanced APM Entity under an Advanced APM and to recognize their role in supporting the Advanced APM Entity. We believe this policy appropriately considers those eligible clinicians who have the most central role or roles in supporting the Advanced APM Entity's performance under an Advanced APM to be the eligible clinician group for purposes of the QP determination. We believe this policy provides for flexibility in the design of Advanced APMs while providing the APM Incentive Payment to those eligible clinicians who are the most engaged in the Advanced APM. We believe this will promote more robust engagement by eligible clinicians in Advanced APMs, and appropriately incentivize participation in Advanced APMs where eligible clinicians have a less direct relationship with the Advanced APM Entity than eligible clinicians who are on a Participation List. We also believe that although the relationship an Affiliated Practitioner has with an Advanced APM Entity is less direct than an eligible clinician on a Practitioner List, the contractual relationship the Affiliated Practitioner has with the Advanced APM Entity is sufficient for an Affiliated Practitioner can become a QP based on their support of the Advanced APM Entity.

We seek comment on our proposals for defining the eligible clinician group for QP determinations, particularly our proposals to define the eligible clinician group for QP determination as the Participation List, and the exception for Advanced APMs in which there are no eligible clinicians on the Participation List but there are eligible clinicians on an Affiliated Practitioner List. Because there may be Advanced APMs in the future that have multiple lists of Affiliated Practitioners, we plan to propose a policy for such situations in future rulemaking, and we seek comment on approaches for grouping those separate lists for purposes of the QP determination.

(3) Timing of Group Identification for Eligible Clinicians

We propose that we will identify the eligible clinician group for each Advanced APM Entity at a specified point in time for each QP Performance Period. We propose that this point in time assessment will occur on December 31st of each QP Performance Period. We believe that taking a “snapshot” of the participant list on the last day of the proposed QP Performance Period provides the best opportunity to comprehensively assess the eligible clinicians' active participation in an Advanced APM throughout an entire QP Performance Period. Under this proposal, we would use the eligible clinicians identified using the Participant List as the group of eligible clinicians who would be assessed together for the purposes of QP determination. We considered taking the “snapshot” at an earlier point in the QP Performance Period, but we felt that because certain APMs allow for changes in participation (either adding or dropping participants from the APM Entity) during the calendar year, an earlier “snapshot” date would not be the most accurate reflection of active eligible clinician participation in a APM throughout the QP Performance Period. We believe that these proposals maintain cohesiveness for eligible clinicians and Advanced APM Entities and maintain consistency with the participation structure of Advanced APMs.

We seek comment on our proposal to assess each Participation List for each Advanced APM Entity at a specified point in time during the QP Performance Period. We also seek comment on the proposed date of the Participant List assessment, and whether this date should be earlier in the QP Performance Period or should instead be a range of time.

(3) Exception

We propose one exception to making QP determinations at the group level. Some eligible clinicians may participate in multiple Advanced APMs. For instance, an eligible clinician could participate in an ACO under the Shared Saving Program and an episode payment model with another entity, both of which have been determined to be Advanced APM Entities. In such a case, we propose the following:

• Consistent with the general policy proposed above, if one or more of the Advanced APM Entities in which the eligible clinician participates meets the QP threshold, the eligible clinician becomes a QP.
• If none of the Advanced APM Entities in which the eligible clinician participates meet the QP threshold, CMS proposes to assess the eligible clinician individually, using combined information for services associated with that individual's NPI and furnished through all such eligible clinician's Advanced APM Entities during the QP Performance Period. CMS will adjust to assure that services are not double-counted (for example, a surgeon participating in a bundled payments model, in which some of the procedures are performed on patients affiliated with an ACO that the surgeon is also a part of, would only have payments or patients from those procedures count once towards the QP determination).

We believe that this proposal maintains the general simplicity of the Advanced APM Entity-level QP determination while acknowledging individual eligible clinicians who are participating in multiple advanced initiatives that support CMS goals. This also complements the policy described under the All-Payer Combination Option for QP determinations in which an eligible clinician may submit information on participation in Other Payer Advanced APMs in order to be assessed as an individual under that option in the event that the APM Entity or Entities in which the eligible clinician participates do not submit sufficient information.

We seek comment on the proposal to make most QP determinations at the Advanced APM Entity level and our proposals for exceptions to that policy. In particular, we seek comment on the merits of making all determinations at the individual eligible clinician level versus through some alternative grouping methodology. We also seek comment on our proposal to assess an eligible clinician who participates in multiple Advanced APM Entities, and any other potential exceptions to the proposed general policy to make QP determinations at the Advanced APM level.

c. Partial QP Election To Report to MIPS

Section 1848(q)(1)(C)(ii)(II) of the Act excludes from the definition of MIPS eligible clinician an eligible clinician who is a Partial QP for a year. However, under section 1848(q)(1)(C)(vii) of the Act, an eligible clinician who is a Partial QP for a year and reports on applicable measures and activities as required under the MIPS is considered to be a MIPS eligible clinician for the year. To carry out these provisions, we propose to require that each Advanced APM Entity must make an election each year on behalf of all of its identified participating eligible clinicians on whether to report under MIPS in the event that the eligible clinicians participating in the Advanced APM Entity are determined as a group to be Partial QPs for a year. We propose that the Advanced APM Entity could change its election for a year at any time during the QP Performance Period, but the election would become permanent at the close of the QP Performance Period. We believe that this is consistent with our proposed general policy to make QP determinations at the Advanced APM Entity level; and with related MIPS policies described in section II.E.3.h of this preamble, under which we propose that each APM Entity would be considered a group for purposes of MIPS reporting. Therefore, we believe that the decision of whether to report and subsequently be subject to MIPS adjustments should also be made at the group level. We seek comment on whether the Advanced APM Entity or each individual eligible clinician should make the Partial QP MIPS reporting election.

As discussed in section II.E.3.h. of this preamble, we recognize that the Shared Savings Program eligible clinicians participate as a complete TIN such that all of the eligible clinician participants in the participant billing TIN participate in the Shared Savings Program. Therefore, we also seek comment on an alternative approach for Shared Savings Program APM Entities in which each individual billing TIN participating in the APM Entity would make the Partial QP election on behalf of its individual eligible clinicians and that election would be applied to all eligible clinicians in that individual billing TIN, as opposed to having the APM Entity (ACO) make the Partial QP election. We would only undertake this alternative paired with determining MIPS CPS for each TIN within an APM Entity (ACO) at the TIN level, an alternative discussed under the APM scoring standard elsewhere in this proposed rule.

Our proposal that Partial QPs may choose whether to report to MIPS has two additional interactions with other proposed policies. First, because we have proposed unique MIPS scoring policies for MIPS eligible clinicians participating in certain APMs, the election by the APM Entity not to report under MIPS is in effect a decision to tell CMS not to score the information submitted by the APM Entity under MIPS. Under our proposal, that decision would be made at the APM Entity level. APM Entities and eligible clinicians would continue to report to their respective APMs as required under the terms of their participation agreements with CMS.

Second, given the proposed timeframe for QP determinations under section II.F.5.a, our proposed treatment of claims run-out, claims adjustments, supplemental service payments, and alternative payment methods for purposes of QP determination (further detailed in section II.F.8 of this preamble), and the and subsequent notification of QP determinations proposed under section II.F.5.d of this preamble, eligible clinicians who become Partial QPs would not receive notification of this status until after the proposed timeframe for the MIPS reporting period will have closed. We do not believe that it would be in the best interest of APM Entities and eligible clinicians, nor would it be operationally feasible, to have APM Entities wait to make a Partial QP election to be included in MIPS until after the close of the MIPS reporting period. Although the information necessary for MIPS reporting would already be prepared in the CMS systems by the time the Partial QP determination is made, a prospective election by the Advanced APM Entity to not be scored under MIPS and receive a MIPS payment adjustment would signal us to not transfer information from our reporting system to the MIPS scoring system in the event of a Partial QP determination, and that any submitted information is not to be used for purposes of a MIPS assessment or payment adjustment. Thus, by choosing not to report under MIPS, those Advanced APM Entities and eligible clinicians determined to be Partial QPs would be exempted from the MIPS payment adjustment for that year. We seek comment on the timing and process for Advanced APM entities to elect whether to be subject to MIPS in the event of a Partial QP determination.

d. Notification of QP Determination

We propose to notify both Advanced APM Entities and their participating eligible clinicians of their QP and Partial QP status as soon as CMS has made the determination and performed all necessary validation of the results. Given the proposed timeframe for QP determinations under section II.F.5.a of this preamble and our proposed treatment of claims run-out (further detailed in section II.F.8 of this preamble), we do not anticipate that this notification could be made before the summer of the subsequent year. We propose that this notification would be made directly to the Advanced APM Entity and eligible clinician, and made in combination with a general public notice on the CMS Web site that such determinations have been completed for the applicable QP Performance Period. We propose that this notification would also contain other necessary and useful information, such as what actions, if any, an Advanced APM Entity or eligible clinician may or should take with respect to MIPS. We believe that this is the most efficient method for dissemination of this information to all QPs, Partial QPs, and MIPS eligible clinicians.

We seek comment on our proposals to make the QP and Partial QP status notifications. We also seek comment on an alternative approach for Shared Savings Program ACOs in which we would separately notify each billing TIN participating in the ACO. We seek comment on other methods and media for the notification of QP and Partial QP status. We also seek comment on the content of such notifications so that they may be as clear and useful as possible.

6. Qualifying APM Participant Determination: Medicare Option

a. In General

Under the Medicare Option, we propose to calculate a Threshold Score for an Advanced APM Entity—or eligible clinician in the cases of an exception described in section II.F.5.b of this preamble—based on participation in an Advanced APM by analyzing claims for Medicare Part B covered professional services. Under the alternative calculation using patient counts in lieu of payments (patient count method), we propose to similarly calculate a Threshold Score for the Advanced APM Entity based on patient attribution as described below. Under either the payment amount or patient count method, only Medicare Part B covered professional services under the physician fee schedule will count toward the numerator and denominator of the Threshold Score calculation.

Section 1833(z)(2)(A), (B)(i) and (C)(i) of the Act describes the QP determination using the Medicare payment method as follows: A QP is an eligible clinician whose payments under this part for covered professional services furnished by such professional during the most recent period for which data are available (which may be less than a year) were attributable to such services furnished under this part through an Advanced APM Entity. Section 1833(z)(2)(D) of the Act describes the basis for the patient count method.

(1) Definitions

In section II.F.3 of this preamble, we propose two definitions that would apply specifically for the purposes of QP determination: Attributed beneficiary and attribution-eligible beneficiary. Each term describes a particular relationship between an Advanced APM Entity and the beneficiaries for whose cost and quality of care the participating eligible clinicians are held accountable. These terms are the foundation for how we propose to count services furnished through an Advanced APM Entity.

In section II.F.3 of this preamble, we propose that “attributed beneficiary” be defined as a beneficiary attributed to the Advanced APM Entity on the latest available list of attributed beneficiaries during the QP Performance Period based on each APM's respective attribution rules. There are some natural advantages to using this term for the purposes of QP determination because it is consistent with how many APMs—including the Shared Savings Program (assigned beneficiaries), Next Generation ACO Model (aligned beneficiaries ), and BPCI Model (attributed beneficiaries ) identify the beneficiaries whose outcomes and costs are included in an APM Entity's assessment. We believe that using the same construct also coordinates the incentives under the Advanced APM with the incentives under MACRA by addressing the same beneficiary population.

In most episode payment models, such as the CJR Model, attribution is defined by the beneficiaries who trigger the defined episode of care under the model, often by presenting with a specific condition at the location of a participating APM Entity. In many attribution-based APMs, such as ACO initiatives or the Comprehensive Primary Care Initiative, CMS attributes beneficiaries to APM Entities through claims-based algorithms that identify the APM Entity with the plurality of evaluation and management visits for a beneficiary. In addition, most APMs do not allow beneficiaries to be attributed to more than one APM Entity. This means that the greater the APM Entity density in a market, the lower the attributed population for a given APM Entity will be as a percent of its total beneficiaries. We seek comment on the proposed methodology for defining the attributed beneficiary population, including comment on alternative methods for capturing the most meaningful cohort of attributed beneficiaries.

Under these plurality-based approaches, typically only 30-50 percent of an Advanced APM Entity's total population of beneficiaries for whom its eligible clinicians furnish services are actually attributed to the Advanced APM Entity for a performance period. These percentages reflect a combination of CMS' design decisions, beneficiaries' underlying care patterns, and the fact that beneficiaries in Medicare FFS retain freedom of choice to select clinicians. These percentages reflect conditions that are not entirely under the control of the APM Entity or its eligible clinicians. Thus, we recognize that because Advanced APMs have different attribution methodologies, using the specific Advanced APM attributed beneficiary as the definition may create a standard that advantages or disadvantages participation in certain Advanced APMs relative to others simply based on the specific attribution policies.

The unintended consequence would be that greater APM participation in a given market could make it impossible for many highly engaged Advanced APM Entities to reach a 50 percent or 75 percent QP Payment Amount Threshold. The result could be that an ACO functioning under arrangements with significant financial risk, (for example, in the Next Generation ACO Model or Track 3 of the Shared Savings Program), would still not meet the QP threshold, particularly in later years of the program under higher thresholds. We believe this would undercut our stated CMS goal of broadly increasing participation in advanced APMs, and we have attempted to compensate for these differences with how we propose to define the terms attributed beneficiary and attribution-eligible beneficiary for the purposes of making QP determinations.

Consistent with our proposed definition of attributed beneficiary, our proposed definition for an attribution-eligible beneficiary would allow us to be more consistent across Advanced APMs in how we consider the population of beneficiaries served by an Advanced APM Entity for the purposes of QP determination. To be attributed to an Advanced APM Entity in an Advanced APM, a beneficiary is first required to first meet certain eligibility criteria. Specifically, for purposes of QP determinations, we propose that an attribution-eligible beneficiary would be one who:

(1) Is not enrolled in Medicare Advantage or a Medicare cost plan.

(2) Does not have Medicare as a secondary payer.

(3) Is enrolled in both Medicare Parts A and B.

(4) Is at least 18 years of age.

(5) Is a United States resident.

(6) Has a minimum of one claim for evaluation and management services by an eligible clinician or group of eligible clinicians within an APM Entity for any period during the QP Performance Period.

An attribution-eligible beneficiary may or may not be an attributed beneficiary. Attributed beneficiaries are a subset of attribution-eligible beneficiaries. Much like the term “attributed beneficiary,” the term attribution-eligible beneficiary is generally consistent with the attribution methodologies used in most current APMs—such as the Shared Savings Program and the Next Generation ACO Model—to identify the beneficiaries who could potentially be attributed to an APM Entity. Although the factors we are proposing for the definition of an attribution-eligible beneficiary in this context would only apply for the purposes of QP determinations, and would not change APM-specific methodologies, we believe that the factors in the proposed definition are representative of the methodologies most current APMs use to perform attribution. Therefore, we believe it would serve as a practical common set to apply in QP threshold calculations.

The purpose of using the attribution-eligible construct is to ensure that the denominator of QP determination calculations described in this section only includes payments for services furnished to patients who could potentially be attributed to an Advanced APM Entity under the Advanced APM, and thus could also appear in the numerator of the QP determination calculations. We believe that including amounts in the denominator that could not possibly be included in the numerator would be arbitrarily punitive toward certain Advanced APM Entities that furnish services to a substantial population of non-attribution-eligible beneficiaries.

We note that specialty-focused or disease-specific APMs may have attribution methodologies that are not based on evaluation and management services. Therefore, we anticipate needing targeted exceptions, especially related to the sixth factor of the definition of attribution-eligible beneficiary, for such APMs so that the attributed beneficiary population is truly a subset of the attribution-eligible population. Such exceptions would be made either through rulemaking or using available waiver authority and would be announced when the APM is announced.

For example, under the CEC Model, one criterion, among others, to be an aligned beneficiary requires that the beneficiary receive maintenance dialysis services. In the event that the CEC Model were determined to be an Advanced APM, we would consider attribution-eligible beneficiaries for the APM Entities participating in the CEC Model to be beneficiaries that meet the first five criteria outlined above and that have had at least one maintenance dialysis service billed through the Advanced APM Entity during the QP Performance Period. We would make this exception for the CEC Model to ensure that the denominator of QP determination calculations described in this section only includes payments for services furnished to patients who could potentially be attributed to an Advanced APM Entity under the Advanced APM.

Although the availability of such exceptions, as outlined above, would create multiple standards, we believe this slightly more complex approach is more appropriate and equitable because it is consistent with the design of APMs. An alternative approach could be to have a simple standard that includes in the denominator all beneficiaries who are furnished any Medicare Part B covered professional service by eligible clinicians participating the Advanced APM Entity.

We seek comment on the proposed general definition of attribution-eligible beneficiary. We further seek comment on our proposal to use of APM-specific standards as necessary to fulfill our expressed goals for specialty- or disease-focused APMs that may use alternative attribution methodologies.

(2) Attribution

We propose to use the attributed beneficiaries on Advanced APM attribution lists generated by each Advanced APM in making QP determinations. We also propose that the attributed beneficiary list would be taken from the Advanced APM's latest available list at the end of the QP Performance Period prior to making the QP determinations. For episode payment models, attributed beneficiaries would be those beneficiaries who trigger episodes of care under the terms of the APM.

We believe that this approach to attribution lists maintains consistency with the panel of beneficiaries for whom Advanced APM Entities are responsible under their respective Advanced APMs during the QP Performance Period. Therefore, we believe that such lists would be appropriate for use in QP determinations. Advanced APM Entities are already accustomed to providing care for the panel of beneficiaries represented by their APM Entity specific list. We believe that our proposal to link attribution for QP determination to Advanced APM attribution lists further strengthens the goals of the Advanced APMs in which these Advanced APM Entities participate. By using the same beneficiary population for QP determination purposes, Advanced APM Entities may continue focusing on the care they furnish to the same panel of attributed beneficiaries, instead of shifting focus and changing practice patterns to reach a QP threshold. As stated in our principles in section II.F.1 of this preamble, we intend for the QP determination process to seamlessly reward participation in the most advanced APMs, not to create a new set of performance standards distinct from the goals of APMs.

We seek comment on our proposal for determining which beneficiaries are considered attributed to an Advanced APM Entity for a QP Performance Period.

b. Payment Amount Method

This section describes our proposal for calculating a Threshold Score for the eligible clinician group in an Advanced APM Entity—or individual eligible clinician in the exception situations under section II.F. 6 of this preamble—using the payment amount method, which would then be compared to the relevant QP Payment Amount Threshold and Partial QP Payment Amount Threshold to determine if the eligible clinician meets the QP status for a payment year.

(1) Claims Methodology and Adjustments

For the payment amount method, section 1833(z)(2)(A), (B)(i) and (C)(i) of the Act requires that we use payments for Medicare Part B covered professional services to make QP determinations. Covered professional services are defined under section 1848(k)(3)(A) of the Act as services for which payment is made under, or based on, the PFS. The payment amounts discussed in this proposal only include payments for Medicare Part B services under, or based on, the Physician Fee Schedule, even if an Advanced APM bases attribution and/or financial risk on payments other than or in addition to Medicare Part B payments.

We propose to use all available Medicare Part B claims information generated during the QP Performance Period. Additionally, we propose that CMS will treat claims run-out, claims adjustments, supplemental service payments, and alternative payment methods in the same manner for purposes of calculating both the Threshold Score and for determining the APM Incentive Payment amount. We further detail our proposals to account for claims run-out, claims adjustments, non-claims-based payments, and alternative payment methods in section II.F.8 of this preamble.

We believe it is appropriate to maintain consistency across the QP determination and the incentive payment calculation in order to support internal CMS operational consistencies. It also ensures that any unique payment mechanisms within an Advanced APM do not affect the opportunity for an eligible clinician to reach the QP threshold.

We seek comment on whether the claims methodology we use under the Medicare payment method should align with the proposed claims methodology for purposes of calculating the estimated aggregate payment amount for the APM Incentive Payment.

(2) Threshold Score Calculation

In general, our proposed method for deriving a Threshold Score for an Advanced APM Entity is to divide the value described under paragraph (a) below by the value described under paragraph (b) below. This calculation would result in a percent value that CMS would compare to the QP Payment Amount Threshold and the Partial QP Payment Amount Threshold to determine the QP status for all eligible clinicians in the Advanced APM Entity for the payment year.

(a) Numerator

We propose that the numerator for this calculation would be the aggregate of all payments for Medicare Part B covered professional services furnished by the eligible clinicians in the Advanced APM Entity to attributed beneficiaries during the QP Performance Period.

We believe that this method is the most logical reading of the statute and is reflective of the population of beneficiaries for whom an Advanced APM Entity is responsible for cost and quality. Therefore, we believe that counting payments for covered professional services furnished to attributed beneficiaries is the most suitable metric for payments that are attributable to services furnished “through” an Advanced APM Entity. In episode payment models, because a beneficiary is considered attributed during the course of an episode, the payments included in the numerator for this calculation are those for Medicare Part B covered professional services furnished to an attributed beneficiary by eligible clinicians in the Advanced APM Entity during the course of an episode.

One program integrity concern is that an Advanced APM Entity might meet the higher QP Payment Amount Threshold in later years by providing substantially disproportionate amounts of care for attributed beneficiaries relative to all others. However, because of the financial risk an Advanced APM Entity bears, which is usually based on expenditures, we believe that the relatively large potential loss under the Advanced APM would outweigh the advantage of any overutilization geared toward abusing Threshold Score calculations.

We seek comment on any alternative numerators we could use for purposes of the Medicare payment method that meaningfully meet statutory requirements, are understandable, and operationally feasible.

(b) Denominator

We propose that the denominator in the Medicare payment method would be the aggregate of all payments for Medicare Part B covered professional services furnished by the eligible clinicians in the Advanced APM Entity to attribution-eligible beneficiaries during the QP Performance Period. We propose that when the QP determination is made at the eligible clinician level as described in section II.F.5 of this preamble, the denominator will be the total of all payments for Medicare Part B covered professional services furnished to attribution-eligible beneficiaries by the eligible clinician. In episode payment models, the payments included in the denominator for this calculation are those for Medicare Part B covered professional services furnished to any attribution-eligible beneficiary by eligible clinicians in the Advanced APM Entity. This includes all such services to all attribution-eligible beneficiaries whether or not such services occur during the course of an episode under the Advanced APM.

We believe that this denominator represents a meaningful alignment with the way in which current APMs perform attribution. Including payment for services furnished only to attribution-eligible beneficiaries standardizes the denominator to ensure fairness across types of eligible clinicians and geographic regions. By using the attribution-eligible population, the denominator will not penalize entities for furnishing services to beneficiaries who could not possibly be in the numerator through attribution under an Advanced APM. For example, an ACO's eligible clinicians may furnish services to a large population of beneficiaries with Medicare as a secondary payer. Those beneficiaries may not be eligible for attribution to the ACO, and could never be included in the numerator. Therefore, we believe that this methodology focuses on factors for which Advanced APM Entities have some control rather than those for which they may have no control or that disadvantage certain organizational structures or types of APMs. We seek comment on alternative methods that are consistent with the statutory language.

c. Patient Count Method

Similar to the Medicare payment method, this section describes our proposal for calculating a Threshold Score for the eligible clinicians participating in an Advanced APM Entity—or eligible clinician in situations under section II.F.6 of this preamble—using the Medicare patient count method, which would then be compared against the relevant QP Patient Count Threshold and Partial QP Patient Count Threshold to determine the QP status of an eligible clinician for the year. Given our authority under section 1833(z)(2)(D) of the Act to use patient counts in lieu of payments “as the Secretary determines appropriate,” we are interpreting the patient count method to offer a more flexible alternative to the payment method. As previously mentioned, the purpose of the proposed design of the Medicare patient count method is to make QP status determinations accessible to entities and individuals who are clearly and significantly engaged in delivering value-based care through participation in Advanced APMs.

(1) Unique Beneficiaries

We propose that when counting the number of beneficiaries under this method, CMS may count a given beneficiary in the numerator and denominator for multiple different Advanced APM Entities. For example, during a year, a beneficiary may be attributed to an ACO, Advanced APM Entity 1, be treated for an episode of care for a particular condition in a hospital participating in an episode payment model as Advanced APM Entity 2, and receive a few services from eligible clinicians in Advanced APM Entity 3. The beneficiary could be included in the numerator and denominator for Advanced APM Entity 1 and Advanced APM Entity 2 and in the denominator for Advanced APM Entity 3. However, the beneficiary could not be counted more than once under the proposed exception for determining QP status for individual eligible clinicians that do not reach QP status under a single Advanced APM; for this exception, each attributed beneficiary would only be counted once in the numerator, and the denominator would consist of all unique attribution-eligible beneficiaries for whom the eligible clinician received payment for covered Medicare professional services for the QP Performance Period.

This is a distinct issue from the question of whether CMS pays shared savings to APM Entities more than once for a given beneficiary. Such payment overlap issues are handled separately through CMS' operational rules governing APM initiative overlaps that address double payments, and are not affected by decisions regarding QP Threshold Score calculations discussed in this regulation.

We propose that CMS will not count any beneficiary more than once for any single Advanced APM Entity. In other words, for each Advanced APM Entity, CMS will count each unique beneficiary no more than one time in the numerator and one time in the denominator.

We believe that counting beneficiaries this way retains integrity of the Threshold Scores by preventing double counting of beneficiaries within an Advanced APM Entity while recognizing the reality that beneficiaries often have relationships with eligible clinicians in different organizations. We seek comment on our proposal for counting beneficiaries.

(2) Claims Methodology and Adjustments

To be consistent with the Medicare payment method, we propose that beneficiary counts would be based on any beneficiary for whom the eligible clinicians within an Advanced APM Entity receive payments for Part B covered professional services, even if an Advanced APM bases its attribution and/or financial risk on both Parts A and B. We propose that for this Threshold Score calculation, we would use any and all available Part B claims information generated during the QP Performance Period.

(3) Threshold Score Calculation

We propose that the Threshold Score would be calculated under the Medicare patient count method as a percent, by dividing the value described under paragraph (a) below by the value described under paragraph (b) below. We include the formula and examples in the summary equation below.

(a) Numerator

We propose that the numerator would be the number of unique attributed beneficiaries to whom eligible clinicians in the Advanced APM Entity furnish Medicare Part B covered professional services during the QP Performance Period. For episode payment models, this would include the number of attributed beneficiaries furnished Medicare Part B covered professional services by eligible clinicians in the Advanced APM Entity during the course of an episode under the Advanced APM.

(b) Denominator

We propose that the denominator would be the number of attribution-eligible beneficiaries to whom eligible clinicians in the Advanced APM Entity furnish covered professional services during the QP Performance Period. For episode payment models, this would include the number of attribution-eligible beneficiaries furnished Medicare Part B covered professional services by eligible clinicians in the Advanced APM Entity group at any point during the QP Performance Period, irrespective of whether such services occur during the course of an episode.

(c) Summary Equation

The proposed Medicare patient score method Threshold Score calculation can be summarized with the following equations.

Threshold Score = A/B

For episode payment models, the equation is:

Threshold Score = A/B

Where:

A = The numerator value under paragraph (a) above.

B = The denominator value under paragraph (b) above.

In general, we believe that through consistency with the payment amount method this approach balances our interests of relative simplicity and having a meaningful standard that recognizes the common aspects of attribution and accountability under Advanced APMs. Similar to the payment amount method, the patient count method represents a proportion of the patients for whom an Advanced APM Entity is accountable under the Advanced APM with respect to all patients who could potentially be attributed to the Advanced APM Entity under the Advanced APM. We believe that it important from any equity perspective to not include patients in the denominator if there is no possibility—based on Advanced APM attribution methodologies—that such individuals could be included in the numerator. We note that although we believe this method to be a fair assessment of the degree of participation in an Advanced APM, our preliminary analyses indicate that many Advanced APM Entities would still miss high thresholds set for later years of the Quality Payment Program.

We seek comment on alternative approaches to the patient count method that would achieve our goal of a simple and meaningful Threshold Score calculation.

(4) Participation in Multiple Advanced APMs

We propose that if the same Advanced APM Entity participates in multiple Advanced APMs and if at least one of those Advanced APMs is an episode payment model, that we would add the number of unique beneficiaries in the numerator of the episode payment model Advanced APM Entity to the numerator(s) for non-episode payment models in which the Advanced APM Entity participates. For example, if an Advanced APM Entity is an ACO in Track 3 of the Shared Savings Program and also in the OCM, (both of which are hypothetically considered to be Advanced APMs for purposes of this example), we would add the entity's unique attributed beneficiaries in OCM to the numerator for its Shared Savings Program Track 3 Threshold Score calculation. We propose that for purposes of this proposal, Advanced APM Entities would be considered the same if CMS determines, that the eligible clinician participant lists are the same or substantially similar, or if the Advanced APM Entity participating in one Advanced APM is the same as, or is a subset of, the other.

The purpose of this proposal is to allow the logical combination of activities under multiple Advanced APMs where appropriate. We believe that the purpose of the incentives for Advanced APM participation is to capture the degree of Advanced APM participation generally, not simply the degree of participation within a single Advanced APM. Where relevant and operationally feasible, we want this program to encourage participation in multiple Advanced APMs. The counterfactual where we would not account for a single Advanced APM Entity's participation in multiple Advanced APMs could be seen as punitive. For instance, an Advanced APM Entity could serve the vast majority of its beneficiaries through several Advanced APMs, but unless that participation is aggregated, the entity could end up with several lower Threshold Scores that are below the QP Patient Count Threshold and not indicative of its broader participation.

We understand the difficulty associated with determining whether two Advanced APM Entities are in fact the same organization. It is highly unlikely that their participant lists will be exactly the same. Therefore, we seek comment on how best to make a determination of substantial similarity, which includes, for example, matching organizational information, aligning TINs, and comparing participant lists. We also seek comment on percentages of participant list or TIN similarity that would be sufficient for APM Entities to be considered under this policy.

d. Use of Methods

CMS may apply one or both of two different methods—using payment amounts or patient counts—to arrive at an eligible clinician's Threshold Score. CMS will compare the Threshold Score against the relevant QP Threshold or Partial QP Threshold to determine an eligible clinician's QP status for the year.

We propose that CMS would calculate Threshold Scores for eligible clinicians in an Advanced APM Entity under both the payment amount and patient count methods for each QP Performance Period. We also propose that CMS would assign QP status using the more advantageous of the Advanced APM Entity's two scores.

We believe that both the payment amount and patient count methods should be considered in order to produce Threshold Scores. As the two calculations differ there may be cases in which Threshold Scores vary enough that different QP determinations could result depending on which is used. In such an event, we do not believe that prioritizing the Threshold Score using one calculation over the other would yield an appropriate, non-arbitrary result. By using the greater of the Threshold Scores achieved, we hope to promote simplicity in QP determinations and to maximize the number of eligible clinicians that attain QP status each year. We seek comment on the use of the payment and patient count methods for the Medicare Option.

e. Services Furnished Through CAHs, FQHCs, and RHCs

(1) Critical Access Hospitals (CAHs)

We propose that professional services billed by CAHs under section 1834(g)(2)(B) of the Act (Method II CAH professional services) would count towards the QP determination threshold calculations for both the Medicare payment and patient count methods in both the numerator and the denominator, as applicable. We believe these services would constitute “covered professional services” under section 1848(k)(3) of the Act because they are furnished by an eligible clinician and payment is based on the Medicare PFS. This policy is consistent with our treatment of payments for Method II CAH professional services for purposes of the EHR Incentive Program and PQRS adjustments under sections 1848(a)(7) and (8) of the Act, respectively. Under section 1848(a)(7) and (8) of the Act, the PQRS and EHR Incentive Program adjustments are applied to payments for covered professional services furnished by an eligible clinician in a Method II CAH.

CAHs were established under the Balanced Budget Act (BBA) of 1997 as a separate provider type with a distinct set of Medicare Conditions of Participation and their own payment methodology. CAHs are not subject to the Medicare Inpatient Prospective Payment System (IPPS) or the Medicare Outpatient Prospective Payment System (OPPS). Instead, CAHs are generally paid based on 101 percent of reasonable costs for inpatient services and are paid for outpatient services under one of two methods: The Standard Payment method outlined in section 1834(g)(1) of the Act (Method I), or the Optional Payment Method outlined in section 1834(g)(2) of the Act (Method II). A CAH is paid under Method I unless it elects to be paid under Method II.

Under Method I, for cost reporting periods beginning on or after January 1, 2004, payments to CAHs are made for outpatient CAH facility services at 101 percent of reasonable costs. Physicians and practitioners receive payment for professional services under the Medicare PFS. A CAH may elect Method II billing, under which the CAH bills Medicare for both facility services and professional services furnished to its outpatients by a physician or practitioner who has reassigned his or her billing rights to the CAH. Even if a CAH makes this election, each physician or practitioner who furnishes professional services to CAH outpatients can choose to either: (1) Reassign his or her billing rights to the CAH, agree to be included under the Method II billing, attest in writing that he or she will not bill Medicare for professional services furnished in the CAH outpatient department, and receive payment from the CAH for the professional services; or (2) elect to file claims for his or her professional services with Medicare for standard payment under the Medicare PFS.

As of January 1, 2004, payment for a physician's professional services provided at a CAH billing under Method II is 115 percent of the allowable amount, after applicable deductions, under the Medicare PFS. For a non-physician practitioner's professional services, the payment amount is 115 percent of the amount that otherwise would be paid for the practitioner's professional services, after applicable deductions, under the Medicare PFS.

(2) Rural Health Clinics (RHCs) and Federally Qualified Health Centers (FQHCs)

RHCs and FQHCs are facilities that furnish services that are typically furnished in an outpatient clinic setting. They are located in areas that have been designated as HPSAs, and meet other requirements.

Under section 1833(a)(3) of the Act, RHCs are paid an all-inclusive rate (AIR) based on reasonable costs, subject under section 1833(f) of the Act to a maximum payment per visit that is established by Congress and updated annually based on the percentage change in the Medicare Economic Index (MEI) and subject to annual reconciliation. The per-visit limit does not apply to RHCs determined to be an integral and subordinate part of a hospital with fewer than 50 beds. Laboratory tests (excluding venipuncture) and technical components of RHC services are paid separately. The RHC payment limit per visit for CY 2016 is $81.32, effective January 1, 2016, through December 31, 2016.

The FQHC Medicare benefit was added when section 1861(aa) of the Act was amended by section 4161 of the Omnibus Budget Reconciliation Act of 1990. FQHCs are paid according to the FQHC PPS set out under section 1834(o) of the Act, in which Medicare pays a national encounter based rate per beneficiary per day, with some adjustments based on where and by whom the services are furnished. The unadjusted 2016 PPS rate is $160.60.

We propose that professional services furnished at RHCs and FQHCs that participate in ACOs, and are reimbursed under the RHC AIR or FQHC PPS (respectively), be counted towards the QP determination calculations under the patient count method but not under the payment amount method.

In certain Medicare ACO APMs, RHC and FQHC services can be counted for purposes of attributing beneficiaries to an ACO. Therefore, we propose to include beneficiaries attributed to an Advanced APM Entity in full or in part because of services furnished by RHCs or FQHCs in the patient counts used for QP determination calculations.

As previously stated, section 1833(z)(2)(D) of the Act permits us to use patient counts in lieu of payments when determining whether an eligible clinician is a QP “as the Secretary determines appropriate.” Our proposal to include the professional services furnished by eligible clinicians at RHCs and FQHCs in the QP threshold calculations for the patient count method is essential to assure consistency with this program and existing APM attribution methodologies. An Advanced APM Entity is responsible for the cost and quality of care for all beneficiaries attributed to an APM Entity, including all professional services furnished to such beneficiaries, regardless of whether or not attribution was based on services furnished by an eligible clinician or by an RHC or FQHC. We believe such beneficiaries are clearly served through the Advanced APM Entity, and it would be potentially confusing to eligible clinicians and Advanced APM Entities to track this distinction strictly for purposes of QP determination. We also believe that it would be unduly burdensome and impractical for CMS to develop and maintain a separate list of beneficiaries aligned to each Advanced APM Entity from the full list of beneficiaries for whom an Advanced APM Entity is responsible under an Advanced APM.

Because professional services furnished by eligible clinicians at RHCs and FQHCs are not reimbursed under, or based on, the Medicare PFS, professional services furnished in these settings do not constitute “covered professional services” under section 1848(k)(3)(A) of the Act. In the Medicare Payment Amount Method, where payments for specified covered professional services are summed, only payments for covered professional services can be included.

We believe that our proposal will continue to encourage the development of APMs that span rural and/or underserved areas. We seek comment on this proposal.

7. Combination All-Payer and Medicare Payment Threshold Option

a. Overview

Beginning in 2021, in addition to the Medicare Option, eligible clinicians may also become QPs through the All-Payer Combination Option, described under section 1833(z)(2)(B)(ii) and (C)(ii) of the Act as the Combination All-Payer and Medicare Payment Threshold Option. Thus, there will be two avenues for eligible clinicians to become QPs—the Medicare Option and the All-Payer Combination Option—and an eligible clinician need only meet the QP threshold under one of them to be a QP for the payment year. The All-Payer Combination Option provides an incentive for eligible clinicians to participate in arrangements with non-Medicare payers that have payment designs similar to those in Advanced APMs. The All-Payer Combination Option uses both the methods described in the Medicare Option and methods that calculate payments for all services from all payers, with certain exceptions, that are attributable to participation in both Advanced APMs and Other Payer Advanced APMs.

Although the statutory QP threshold for an eligible clinician to be a QP (the QP Payment Amount Threshold) under the Medicare Option increases from 25 percent in 2019 and 2020 under section 1833(z)(2)(A) of the Act, to 50 percent in 2021 and 2022 under section 1833(z)(2)(B)(i) of the Act, to 75 percent beginning in 2023 under section 1833(z)(2)(C)(i) of the Act, the All-Payer Combination Option allows eligible clinicians with lower levels of participation in Advanced APMs to become QPs through sufficient participation in Other Payer Advanced APMs with payers such as State Medicaid programs and commercial payers, including Medicare Advantage plans. Similar to Medicare payment amount and patient count methods the statute also allows, under section 1833(z)(2)(D) of the Act, the QP determination to be based on payment amount or on counts of patients in lieu of payments using the same or similar percentage criteria. These QP thresholds are presented in Tables 38 and 39, and the process is shown in Figures J and K. The process shown in H and I will be similar for the patient count threshold, although only the process for the payment amount threshold is displayed. CMS may reassess the QP Patient Count Thresholds in future years based on the experience gained from eligible clinician Threshold Scores during the first years of operations. In summary, eligible clinicians may become QPs if the following steps occur as described below in the associated sections: (1) The eligible clinician submits to CMS sufficient information on all relevant payment arrangements with other payers; (2) CMS determines that an Other Payer APM is an Other Payer Advanced APM; (3) the eligible clinician meets the relevant QP thresholds by having sufficient payments or patients attributed to a combination of participation in Advanced APMs and Other Payer Advanced APMs.

Sections 1833(z)(2)(B)(ii) and (C)(ii) of the Act describe the payment amount method for making the QP determination under the All-Payer Combination Option. For purposes of making a QP determination under this option, a QP is an eligible clinician for whom we determine with respect to items and services furnished by such professional during the most recent period for which data are available (which may be less than a year) that, at least the specified percent of the sum of combined Medicare payments and all other payments regardless of payer are through Advanced APMs and Other Payer Advanced APMs that meet the criteria set forth in this section.

b. Other Payer Advanced APM Criteria

(1) In General

A payment arrangement with a non-Medicare payer (Other Payer APM) can become an Other Payer Advanced APM if the arrangement meets three criteria:

• Certified Electronic Health Record technology (CEHRT) is used;
• Quality measures comparable to measures under the MIPS quality performance category apply; and
• The APM Entity either: (1) Bears more than nominal financial risk if actual aggregate expenditures exceed expected aggregate expenditures; or (2) for beneficiaries under title XIX, is a medical home in a Medicaid Medical Home Model that meets criteria comparable to Medical Home Models expanded under section 1115A(c) of the Act.

Other Payer APMs include payment arrangements under any payer other than traditional Medicare. Medicare Advantage and other Medicare-funded private plans are categorized as a payer other than traditional Medicare for these purposes. In this section, we explain how the three criteria are applied to determine whether arrangements are Other Payer Advanced APMs.

(2) Medicaid APMs

We propose to define a Medicaid APM as a payment arrangement under title XIX that meets the criteria to be an Other Payer Advanced APM as proposed in this section. States can choose from different authorities in title XIX when implementing new payment models. We believe this proposal would provide some flexibility for States but align the core requirements for Medicaid APMs with the broader Advanced APM and Other Payer Advanced APM criteria. Otherwise, we intend to generally defer to states in their design of payment arrangements.

(3) Medicaid Medical Home Models

We propose that a Medicaid Medical Home Model is a Medical Home Model that is operated under a State title XIX program instead of under section 1115A of the Act. Section 1833(z) of the Act mentions medical homes and what we have termed Medicaid Medical Homes (those with respect to beneficiaries under title XIX) several times, but does not define the terms. In addition, Medicaid Medical Home is not defined in title XIX or in Medicaid laws or regulations. Therefore, we need to define the terms because of their importance in the Quality Payment Program.

We propose that a Medicaid Medical Home Model must have the following elements at a minimum:

• Model participants include primary care practices or multispecialty practices that include primary care physician and practitioners and offer primary care services, and
• Empanelment of each patient to a primary clinician.

In addition to these elements, we propose that a Medicaid Medical Home Model must have at least four of the following elements:

• Planned chronic and preventive care.
• Patient access and continuity.
• Risk-stratified care management.
• Coordination of care across the medical neighborhood.
• Patient and caregiver engagement.
• Shared decision-making.
• Payment arrangements in addition to, or substituting for, fee-for-service payments (for example, shared savings, population-based payments).

This definition of Medicaid Medical Home Model applies only for the purposes of the Quality Payment Program, and could be defined differently for other purposes. To define these terms, we reviewed existing and past Medical Home Models CMS developed under section 1115A of the Act, including the Comprehensive Primary Care Initiative (CPC). In addition, we reviewed a variety of other sources including several from the National Committee for Quality Assurance, the Joint Principles of the Patient-Centered Medical Home (a joint statement by the American Academy of Family Physicians, the American Academy of Pediatrics, the American College of Physicians, and the American Osteopathic Association), and the Agency for Healthcare Research and Quality. Our proposed definition of Medicaid Medical Home Model uses common elements from these sources. We believe that using a common set of elements ensures general comparability between Medical Home Models and Medicaid Medical Home Models while maintaining flexibility for the States under title XIX. In response to the MIPS and APMs RFI, some commenters suggested that we should require a specific method or accreditation process for recognizing Medicaid Medical Home Models, while others asked us not to use such an approach. We will not mandate a specific method or accreditation process. We believe that such a policy would provide limited additional benefit while unnecessarily restricting state innovation. However, we believe it likely that accredited models, such as those certified by the National Committee on Quality Assurance may also meet these proposed criteria. Medicaid Medical Home Models can be Other Payer Advanced APMs if they meet the criteria set forth in this section.

We seek comment on the definitions of Medicaid APMs and Medicaid Medical Homes Models.

(4) Use of Certified Electronic Health Record Technology

To be an Other Payer Advanced APM, as described under section 1833(z)(2)(B)(iii)(II)(bb) and (z)(2)(C)(iii)(II)(bb) of the Act, payments must be made under arrangements in which certified EHR technology is used. This is slightly different than the requirement for Advanced APMs that “requires participants in such model to use certified EHR technology (as defined in section 1848(o)(4) of the Act),” as specified in section 1833(z)(3)(D)(i)(I) of the Act. Although the statutory requirements are phrased slightly differently, we believe that there is value in keeping the two standards—for Advanced APMs and Other Payer Advanced APMs—as similar as possible.

We propose that Other Payer APMs would meet this Other Payer Advanced APM criterion under sections 1833(z)(2)(B)(iii)(II)(bb) and (z)(2)(C)(iii)(II)(bb) of the Act by requiring participants to use CEHRT as defined for MIPS and APMs under § 414.1305. This approach is consistent with the approach for Advanced APMs as described in section II.F.4.b.(1) of this preamble. In the 2015 EHR Incentive Programs final rule (80 FR 62872 through 62873), we established the definition of CEHRT for EHR technology that must be used by eligible clinicians to meet the meaningful use objectives and measures in specific years. In this proposed rule, we are proposing to adopt the specifications from within the current definition of CEHRT in our regulation at § 414.1305 for eligible clinicians participating in MIPS or in APMs. This definition is identical to the definition for use by eligible hospitals and CAHs and Medicaid eligible clinicians in the EHR Incentive Programs.

In accordance with section 1833(z)(2)(C)(iii)(II) of the Act, we propose that an Other Payer Advanced APM must require at least 75 percent of eligible clinicians in each participating APM Entity (or each hospital if hospitals are the APM participants) to use the certified health IT functions outlined in the proposed definition of CEHRT to document and communicate clinical care with patients and other health care professionals.

We seek comment on the proposed definition of CEHRT for Advanced APMs and Other Payer Advanced APMs and whether they should be the same for both. We seek comment on the proposed method for Other Payer APMs to meet the CEHRT use criterion.

(5) Application of Quality Measures Comparable to Those Under the MIPS Quality Performance Category

Another of the criteria to be considered an Other Payer Advanced APM, as described in sections 1833(z)(2)(B)(ii)(II)(aa) and (C)(iii)(II)(aa) of the Act, are quality measures comparable to those under MIPS quality performance category apply under the Other Payer APM. Under the MACRA and in this proposal, not all quality measures in an APM are required to be “comparable” and not all payments under the APM must be based on comparable measures. This approach is similar to the requirement for Advanced APMs as described in section II.F.4.b.(2) of this preamble. Under this proposal, Other Payer APMs retain sufficient freedom to innovate in paying for services and measuring quality. For instance, an Other Payer APM may have incentive payments related to quality, total cost of care, participation in learning activities, and adoption of health IT. The existence of all of the payments associated with non-quality aspects does not preclude the Other Payer APM from meeting this Other Payer Advanced APM criterion. In other words, this criterion only sets standards for payments tied to quality measurement, not other methods of payment. Conversely, an Other Payer APM may test new quality measures that do not fall into the MIPS-comparable standard. So long as the Other Payer APM meets the requirements set forth in this criterion, there is no additional prescription for how the Other Payer APM tests additional measures that may or may not meet the standards under this criterion. Therefore, we propose that the quality measures on which the Other Payer Advanced APM bases payment must include at least one of the following types of measures provided that they have an evidence-based focus and are reliable and valid as described in section II.F.4.b.(2) of this preamble:

(1) Any of the quality measures included on the proposed annual list of MIPS quality measures;

(2) Quality measures that are endorsed by a consensus-based entity;

(3) Quality measures developed under section 1848(s) of the Act;

(4) Quality measures submitted in response to the MIPS Call for Quality Measures under section 1848(q)(2)(D)(ii) of the Act; or

(5) Any other quality measures that CMS determines to have an evidence-based focus and are reliable and valid.

We want to encourage the use of outcome measures for quality performance assessment in Other Payer APMs. As we did for APMs in section II.F.4.b.(2) of this preamble, we propose that in addition to the general comparable quality measure requirement proposed in this section, an Other Payer Advanced APM must include at least one outcome measure if an appropriate measure (that is, the measure addresses the specific patient population and is specified for the APM participant setting) is available on the MIPS list of measures for that specific QP Performance Period.

We believe that this framework will provide other payers the flexibility needed to ensure that their quality performance metrics meet their unique goals. We seek comment on this proposed criterion.

(6) Financial Risk for Monetary Losses

As described in sections 1833(z)(2)(B)(iii)(II)(cc) and (C)(iii)(II)(cc) of the Act, the third criterion that an Other Payer APM must meet to be an Other Payer Advanced APM is that under the arrangement, the APM Entity must either bear more than nominal financial risk if actual aggregate expenditures exceed expected aggregate expenditures or the Other Payer APM be a Medicaid Medical Home Model that meets criteria comparable to Medical Home Models expanded under section 1115A(c) of the Act, as described in paragraph (d) below.

The financial risk standard under this criterion is similar to that proposed for the Advanced APM criterion. For purposes of determining whether the Other Payer APM is an Other Payer Advanced APM, this proposal does not impose any additional performance criteria, such as actual achievement of savings, on APM Entities in other payer arrangements. As with all the proposed Advanced APM criteria, this requirement pertains to the payment arrangement structure, not of the performance of the participants within the payment arrangement.

This section is broken into two main parts: (1) What it means for an Advanced APM Entity to bear financial risk if actual aggregate expenditures exceed expected aggregate expenditures under an Other Payer Advanced APM; and (2) what amounts of risk are considered to be more than nominal.

We prioritized keeping the standards consistent across different types of APMs, including Advanced APMs as described in section II.F.4.b.(3) of this preamble.

(a) Bearing Financial Risk for Monetary Losses

We propose a generally applicable standard for Other Payer Advanced APMs and a slightly different standard for Medicaid Medical Home Models. We want to be consistent with and comparable to the Advanced APM financial risk standard within the limits of the statutory text.

(i) Generally Applicable Other Payer Advanced APM Standard

We propose that the generally applicable financial risk standard for Other Payer Advanced APMs would be that a payment arrangement must, if APM Entity actual aggregate expenditures exceed expected aggregate expenditures during a specified performance period:

• Withhold payment for services to the APM Entity and/or the APM Entity's eligible clinicians;
• Reduce payment rates to the APM Entity and/or the APM Entity's eligible clinicians; or
• Require direct payments by the APM Entity to the payer.

We believe this financial risk criterion best distinguishes most Other Payer APMs from those that are focused on challenging physicians and practitioners to assume risk and provide high value care. We expect that an increasing proportion Other Payer APMs will meet that bar over time. This proposal is based on the statutory requirement that the APM Entity bear risk if aggregate actual expenditures exceed aggregate expected expenditures under the model, and is consistent with our proposal for the corresponding criterion proposed for Advanced APMs. Through the MIPS and APM RFI, many stakeholders commented that business risk should be sufficient to meet this Advanced APM criterion. We do not intend for our proposal to minimize the substantial time and financial commitments that APM Entities invest to become successful APM participants. We note that there is also difficulty in creating an objective and enforceable standard for determining whether an entity's business risk exceeds a nominal amount, and that the statutory framework for the APM Incentive Payment recognizes that not all alternative payment arrangements will meet the criteria to be considered for purposes of the QP determination. We seek comments regarding the proposed standard and whether there are other types of arrangements that should be incorporated into the standard.

(ii) Medicaid Medical Home Model Financial Risk Standard

We propose that for a Medicaid Medical Home Model to be an Other Payer Advanced APM if the APM Entity's actual aggregate expenditures exceed expected aggregate expenditures, the APM must:

• Withhold payment for services to the APM Entity and/or the APM Entity's eligible clinicians;
• Reduce payment rates to the APM Entity and/or the APM Entity's eligible clinicians;
• Require direct payment by the APM Entity to the payer; or
• Require the APM Entity to lose the right to all or part of an otherwise guaranteed payment or payments.

For instance, a Medicaid Medical Home Model would meet our proposed financial risk criterion if it conditions the payment of some or all of a regular care management fee to medical home APM Entities upon expenditure performance in relation to a benchmark. Because the arrangement would require no direct payment as a consequence for failure to meet expenditure standards, such a medical home would not necessarily be worse off than it had been prior to the decreased payment. However, it would be worse off in the future than it otherwise would have been had it met expenditure standards. Similarly, a Medicaid Medical Home Model that offers expenditure and quality performance payments in addition to payment withholds that can be earned back for meeting minimum requirements would also meet this criterion. Consistent with the treatment of Medical Home Models under the statute, this proposal acknowledges the unique challenges of medical homes in bearing risk for losses while maintaining a more rigorous standard than mere business risk.

We believe that because Medicaid Medical Home Models are unique types of Medicaid APMs and because they are identified and treated differently by the statute under the Quality Payment Program, it is appropriate to establish a unique standard for bearing financial risk that reflects these differences and remains consistent with the statutory scheme, which is to provide incentives for participation by eligible clinicians in advanced APMs.

Similar to Medical Home Model standards for Advanced APMs in II.F.4.b.(3), we believe that it would be appropriate to impose size and composition limits for Medicaid Medical Home Models to ensure that the focus is on organizations with a limited capacity for bearing the same magnitude of financial risk as larger APM Entities do. We propose that this limit would only apply to APM Entities that participate in Medicaid Medical Home Models and that have 50 or fewer eligible clinicians in the organization through which the APM Entity is owned and operated. Thus, in a Medicaid Medical Home Model that is an Other-Payer Advanced APM, only those APM Entities that are part of a parent organization with 50 or fewer eligible clinicians would be Advanced APM Entities. We believe it is appropriate to use eligible clinicians, rather than physicians, when setting this threshold as the number of eligible clinicians both reflects organizational resources and capacity and also may differ substantially across organizations with the same number of physicians.

We also believe that this size threshold of 50 eligible clinicians is appropriate as organizations of that size have demonstrated the capacity and interest in taking on risk, and organizations may also join together to take on risk collectively, for example, in an ACO. In the event that a Medicaid Medical Home Model happens to have criteria that meet the Advanced APM financial risk criterion that is generally applicable to all Other Payer APMs, this organizational size limitation would be moot.

There are several unique aspects of Medicaid Medical Home Models, which statute specifically singles out for unique treatment, and their participating APM Entities (medical homes) that support the need for a separate standard to assess financial risk if actual expenditures exceed expected expenditures. Medical homes are generally more limited in their ability to bear financial risk than other Entities because they tend to be smaller and predominantly include primary care practitioners, whose revenues are a smaller fraction of the beneficiaries' total cost of care than those of other eligible clinicians. Moreover, Medicaid medical homes serve low income populations and those with significant health disparities; due to the method of payment for care for these populations, Medicaid medical home practices often have relatively low revenues. Lastly, Medicaid Medical Home Models to date have not required participants to bear substantial downside risk, and including such a requirement under this program would create a significant challenge for medical homes to serve their patients.

We seek comment on the proposed financial risk standard set forth for Medicaid Medical Home Models and on alternative standards that would be consistent with the statute and could achieve our stated goals. We also seek comment on types of financial risk arrangements that may not be clearly captured in this proposal.

(b) Nominal Amount of Risk

When an Other Payer APM risk arrangement meets the proposed financial risk standard, we would then consider whether the risk is of a more than nominal amount such that it meets this nominal risk standard. Similar to the financial risk portion of this assessment, we propose to adopt a generally applicable nominal amount standard for Other Payer Advanced APMs and a unique nominal amount standard for Medicaid Medical Home Models.

We propose to measure three dimensions of risk to determine whether a model meets the nominal amount standard: (a) Marginal risk, which is a common component of risk arrangements—particularly those that involve shared savings—that refers to the percentage of the amount by which actual expenditures exceed expected expenditures for which an APM Entity would be liable under an Other Payer APM; (b) minimum loss rate (MLR), which is a percentage by which actual expenditures may exceed expected expenditures without triggering financial risk; and (c) total potential risk, which refers to the maximum potential payment for which an APM Entity could be liable under an Other Payer APM. An example of marginal risk within an Other Payer APM could be set up in a manner similar to the Shared Savings Program, where an ACO that has a sharing rate, or marginal rate, of 50 percent and exceeds its benchmark (expected expenditures) by $1 million would be liable for $500,000 of those losses. The marginal risk could also vary with the amount of losses.

When assessing whether an Other Payer APM meets the marginal and total risk portions of the nominal risk standard, we would use the same approach we proposed to use with respect to APMs. Specifically, to determine whether an Other Payer APM satisfies the total risk portion of the nominal risk standard, we would identify the maximum potential payment an APM Entity could be required to make as a percentage of the expected expenditures under the Other Payer APM. If that percentage exceeded the required total risk percentage, then the arrangement would satisfy the total risk portion of the nominal risk standard.

To determine whether an Other Payer APM satisfies the marginal risk portion of the nominal risk standard, we would examine the payment required under the Other Payer APM as a percentage of the amount by which actual expenditures exceeded expected expenditures. We propose that we would require that this percentage exceed the required marginal risk percentage regardless of the amount by which actual expenditures exceeded expected expenditures, with two exceptions.

First, we propose a maximum allowable “minimum loss rate” (MLR) of 4 percent in which the payment required by the Other Payer APM could be smaller than the nominal amount standard would otherwise require when actual expenditures exceed expected expenditures by less than 4 percent; this exception accommodates Other Payer APMs that include zero risk with respect to small losses but otherwise satisfy the marginal risk standard. We also propose a process through which CMS could determine that a risk arrangement with an MLR higher than 4 percent could meet the nominal amount standard, provided that the other portions of the nominal risk standard are met. In determining whether such an exception would be appropriate, CMS would consider: (1) Whether the size of the attributed patient population is small; (2) whether the relative magnitude of expenditures assessed under the Other Payer APM is particularly small; and (3) in the case of test of limited size and scope, whether the difference between actual expenditures and expected expenditures would not be statistically significant even when actual expenditures are 4 percent above expected expenditures. We note that CMS would grant such exceptions rarely, and CMS would expect APMs considered for such exceptions to demonstrate that a sufficient number of APM Entities are likely to incur losses in excess of the higher MLR. In other words, the potential for financial losses based on statistically significant expenditures in excess of the benchmark remains meaningful for participants.

Second, we propose that the payment required by the Other Payer APM could be smaller when actual expenditures exceed expected expenditures by enough to trigger a payment greater than or equal to the total risk amount required under the nominal amount standard (as specified in Table 40). This exception ensures that the marginal risk requirement does not effectively require Other Payer APMs to incorporate total risk greater than the amount required by the total risk portion of the standard in order to become Other Payer Advanced APMs.

In evaluating both the total and marginal risk portions of the nominal amount standard, we would not include any payments the APM Entity or its participating providers would make to the other payer if actual expenditures exactly matched expected expenditures. In other words, payments made to the other payer outside the risk arrangement related to expenditures would not count toward the nominal risk standard. This requirement ensures that perfunctory or pre-determined payments do not supersede incentives for improving efficiency. For example, an Other Payer APM that simply requires an APM Entity to make a payment equal to 5 percent of the Other Payer APM benchmark at the end of the year, regardless of actual expenditure performance, would not satisfy the nominal amount standard.

Finally, like the Advanced APM criterion described in section II.F.4.b.(4) of this preamble, the amounts described in this section need not take a shared savings structure in which financial risk increases smoothly based on the amount by which an Other Payer Advanced APM Entity's actual expenditures exceed expected expenditures. The risk arrangement must be tied to expenditures, but the amount of that risk does not have to be directly proportional to expenditures. For instance, an APM Entity could be required to pay the payer a flat amount or an amount tied to the number of attributed beneficiaries in the case of exceeding an expenditure benchmark, provided that these amounts are otherwise structured in a way that satisfies the nominal amount standard.

(i) Generally Applicable Other Payer Advanced APM Nominal Amount Standard

Except for risk arrangements described under the Medicaid Medical Home Standard in paragraph (ii) below, we propose that for an Other Payer APM to meet the nominal amount standard the specific level of marginal risk must be at least 30 percent of losses in excess of the expected expenditures and total potential risk must be at least four percent of the expected expenditures.

Other Payer APM arrangements with less than 30 percent marginal risk would not meet the nominal amount standard. We believe that meaningful risk arrangements can be designed with marginal risk rates of greater than 30 percent. Any marginal risk below 30 percent creates scenarios in which the total risk could be very high, but the average or likely risk for an Other Payer APM Entity would actually be very low.

Table 40 summarizes the generally applicable nominal amount standard.

In establishing the proposed criteria for Other Payer Advanced APMs, we are keeping the approach to nominal risk as consistent as possible with the approach for the proposed Advanced APM criteria as described in section II.F.4.b.(4) of this preamble. The statute specifies that the Other Payer Advanced APM Entity must bear more than nominal financial risk if actual aggregate expenditures exceed expected aggregate expenditures. We believe it is important, to the extent possible and consistent with the statute, to adopt consistent financial risk standards with the Advanced APM standard as described in section II.F.4.b.(3) in this preamble, so that eligible clinicians can base their decisions on participation in these Other Payer APMs on a consistent set of criteria. The Advanced APM financial risk section of this preamble, II.F.4.b.(3) describes the process by which we arrived at the proposed values.

For Medicaid APMs we propose the same standard as for Other Payer APMs. However, we recognize that Medicaid practitioners may be less able to bear substantial financial risk because they are generally reimbursed at lower payment rates, and they serve low-income populations and those with significant health disparities. Therefore, we seek comment and supporting evidence on whether the proposal offered identifies the appropriate amounts of nominal risk for Medicaid APMs.

(ii) Medicaid Medical Home Model Nominal Amount Standard

For Medicaid Medical Home Models, we propose that the minimum total annual amount that an APM Entity must potentially owe or forego to be considered an Other Payer Advanced APM must be at least:

• In 2019, 4 percent of the APM Entity's total revenue under the payer.
• In 2020 and later, 5 percent of the APM Entity's total revenue under the payer.

We believe that because few Medicaid Medical Homes have experience with financial risk, and because they tend to be smaller in size than other APM Entities, we should not include a potentially excessive nominal amount for such entities in the first year of the program. We have also taken into account that the MACRA explicitly highlights Medical Home Models, generally, for special treatment under the Quality Payment Program. We have less information on Medicaid Medical Home Models and their performance to date compared to our information on Medical Home Models. Medicaid Medical Home Models are still developing, and we believe the introduction of a nominal amount standard that is not currently widely represented in the marketplace should be approached in a measured manner. We therefore believe that the unique characteristics of Medicaid Medical Home Models warrant the application of a nominal amount standard that reflects these differences, and statute provides us with the flexibility to make such a distinction.

We seek comment on all of the proposed nominal amount standards. We also seek comment on the potential inclusion of a marginal risk amount in the standard and the extent to which it is applicable.

(c) Capitation

We propose that full capitation risk arrangements would meet this Other Payer Advanced APM financial risk criterion. We propose that for purposes of this rulemaking, a capitation risk arrangement means a payment arrangement in which a per capita or otherwise predetermined payment is made to an APM Entity for services furnished to a population of beneficiaries, and no settlement is performed for the purpose of reconciling or sharing losses incurred or savings earned by the APM Entity. Our rationale for this policy is the same as the rationale on capitation for Advanced APMs described in section II.F.4.b.(3) of this preamble. As such, we reiterate that capitation should not simply be a cash flow mechanism. We also reiterate that capitation arrangements qualifying under the financial risk standard must be structured to directly hold the provider—or the entity to which the provider has assigned their billing—accountable.

We seek comment on our proposal for categorical definition of Other Payer APM capitation risk arrangements as meeting the financial risk criterion for Other Payer Advanced APMs, and on our proposed definition of a capitation risk arrangement. We also seek comment on other types of arrangements that may be suitable for such treatment for purposes of this financial risk criterion.

(d) Criteria Comparable to Expanded Medical Home Model

In accordance with sections 1833(z)(2)(B)(iii)(II)(cc)(BB) and (C)(iii)(II)(cc)(BB) of the Act, we propose that Medicaid Medical Home Models that meet criteria comparable to a Medical Home Model expanded under section 1115A(c) of the Act would meet the Other Payer Advanced APM financial risk criterion. We propose that CMS will specify in subsequent rulemaking the criteria of any Medical Home Model that is expanded under section 1115A(c) of the Act that will be used for purposes of making this comparability assessment. We believe that the expanded Medical Home Model criteria can only be used for comparison when a Medical Home Model is, in fact, expanded as described in section II.F.4.b.(6) of this preamble, not merely by satisfying the expansion criteria under section 1115A(c) of the Act. If no such Medical Home Model has actually been expanded under section 1115A(c) of the Act, we would not have any criteria for comparison. In the absence of any expanded Medical Home Model to which we could draw comparisons, Medicaid Medical Home Models must meet the financial risk criterion through the other provisions (the financial risk and nominal amount standards) in order to be an Other Payer Advanced APM. We seek comment on how to determine the criteria of an expanded Medical Home Model that could be used for comparison, and on how similar the Medicaid Medical Home Model criteria must be to the expanded Medical Home Model criteria in order to be considered “comparable.”

(7) Medicare Advantage (MA)

We received multiple comments on the MIPS and APMs RFI requesting that participation in Medicare Advantage be credited as participation in Advanced APMs. We recognize that many eligible clinicians participating in Medicare Advantage may offer high-value care to Medicare beneficiaries enrolled in such plans.

With respect to the APM Incentive Payment, section 1833(z)(1)(A) of the Act clearly states that the APM Incentive Payment is based on payments for Part B for covered professional services (which are made under the Medicare Physician Fee Schedule) and which do not include payments for services furnished to Medicare Advantage enrollees. For QP determination calculations, we believe it is important to note that APMs may involve Medicare Advantage plans and payers other than Medicare. Under the All-Payer Combination Option for QP determinations, eligible clinicians and Advanced APM Entities can meet the QP threshold based in part on payment amounts or patients counts associated with Medicare Advantage plans and other payers, provided that such arrangements meet the criteria to be considered Other Payer Advanced APMs. However, under sections 1833(z)(2)(A), (2)(B)(i), and (3)(B)(i) of the Act, such Medicare Advantage and other payer payments cannot be included in the QP determination calculations under the Medicare Option, which requires that we only consider payment amounts or patient counts for Medicare Part B covered professional services. Regardless of which option—Medicare or All-Payer Combination—is used to determine that an eligible clinician is a QP for a year, the APM Incentive Payment calculation will only be based upon payments for Medicare Part B covered professional services, which does not include payments for services furnished to Medicare Advantage enrollees.

We recognize that Medicare Advantage contracts can include financial risk as well as quality performance standards and certified EHR and other health IT requirements that support high-value care. We propose to evaluate payment arrangements between eligible clinicians, APMs Entities and MA plans as Other Payer APMs and according to the proposed Other Payer Advanced APM criteria. In the assessment of MA plans with respect to the Other Payer Advanced APM criteria, it is important to note that the requirements refer to aspects of the payment arrangement between the MA plan and the participating APM Entity, and this includes the criterion for bearing more than a nominal amount of financial risk. To qualify as an Other Payer Advanced APM, there must be a financial risk component. We would not consider an arrangement where the MA plan meets the CEHRT and quality measures criteria outlined in this proposed rule, but pays the APM Entity on a fee-for-service basis, to be an Other Payer Advanced APM because there is no risk connected to actual cost of care exceeding projected cost of care. Because this arrangement would not be an Other Payer Advanced APM, it would not be assessed for the purposes of determining QPs. In addition, the financial relationship between CMS and the MA plan—even if the relationship is part of a APM—is not relevant to this assessment because there would not be a direct payment arrangement between CMS and the APM Entities or eligible clinicians.

We also received comments on the MIPS and APMs RFI expressing concern that the distribution of APM Incentive Payments could disadvantage Medicare Advantage plans relative to Medicare FFS by changing payment rates for health plans in a given area based on the aggregate APM incentive amounts paid to eligible clinicians in that area. APM Incentive Payments will be lump-sum payments made under Medicare Part B, but outside of the claims payment system. Medicare Advantage rates are set through a separate process, and payment policies for 2019 will be addressed in the Advance Notice and Rate Announcement for that program.

c. Calculation of All-Payer Combination Option Threshold Score

(1) Submission of Information for Other Payer Advanced APM Determination and Threshold Score Calculation

We propose that APM Entities and/or eligible clinicians must submit certain information for CMS to assess whether other payer arrangements meet the Other Payer Advanced APM criteria and to calculate Threshold Scores a QP determination under the All-Payer Combination Option. For CMS to make QP determinations at the individual eligible clinician level in the specified exception cases described in section II.F.5 and II.F.6 of this preamble, either the Advanced APM Entity or the eligible clinician may submit this information with respect to the individual eligible clinician. If we do not receive sufficient information to complete our evaluation of the other payer arrangement and perform the QP threshold calculation, we would not evaluate the eligible clinicians under the All-Payer Combination Option.

We propose that submissions by APM Entities and/or eligible clinicians must include at least sufficient information for CMS to determine whether the payment arrangement meets the Other Payer Advanced APM criteria described in this section. To make the QP determination using the All-Payer Combination Option, submissions must include specific payment and patient numbers for each payer from whom the eligible clinician has received payments during the QP Performance Period, in order to calculate the Advanced APM Entity eligible clinician group's or individual eligible clinician's Threshold Score. We propose that—by a date and in a manner specified by CMS—the following data must be submitted to CMS for consideration under the All-Payer Combination Option: (1) The payment amounts and/or number of patients furnished any service through each Other Payer Advanced APM for each payer; and (2) the sum of their total payment amounts and/or number of patients furnished any service from each payer.

CMS will ask each payer to attest to the accuracy of all submitted information including the reported payment and patient data. Contracts may be subject to audit by CMS. We propose that if a payer does not attest to the accuracy of the reported payment and patient data, these data will not be assessed under the All-Payer Combination Option. However, we recognize that such a requirement leaves eligible clinicians dependent on a payer over which they may have limited control. We therefore seek comment on alternatives to requiring payer attestation, such as addressing the scope and intensity of audits to verify the submitted data. For Advanced APM Entities and eligible clinicians participating in Medicaid, CMS will initiate a review and determine in advance of the QP determination period the existence of Medicaid Medical Home Models and Medicaid APMs based on information obtained from state Medicaid agencies and other authorities, such as professional organizations or research entities. We seek comment regarding how such a review and determination could be conducted.

Detailed guidance on implementing data collection for Calculation of the All-Payer Combination Option Threshold Score will be issued prior to 2019.

(1) Use of Methods

CMS may apply one or both of two different methods—using payment amounts or patient counts—to arrive at an eligible clinician's Threshold Score. CMS will compare the Threshold Score against the relevant QP Threshold or Partial QP Threshold to determine an eligible clinician's QP status for the year.

We propose that CMS would calculate Threshold Scores for eligible clinicians in an Advanced APM Entity under both the payment amount and patient count methods for each QP Performance Period. We also propose that CMS would assign QP status using the more advantageous of the Advanced APM Entity's two scores.

We believe that both the payment amount and patient count methods should be considered in order to produce Threshold Scores. As the two calculations differ there may be cases in which Threshold Scores vary enough that different QP determinations could result depending on which is used. In such an event, we do not believe that prioritizing the Threshold Score using one calculation over the other would yield an appropriate, non-arbitrary result. By using the greater of the Threshold Scores achieved, we hope to promote simplicity in QP determinations and to maximize the number of eligible clinicians that attain QP status each year. We seek comment on the use of the payment and patient count methods for the All-Payer Combination Option.

(2) Excluded Payments

Section 1833(z)(2)(B)(ii)(I) and (C)(ii)(I) of the Act specifies that the calculation under the All-Payer Combination Option is based on the sum of both payments for Medicare Part B covered professional services and, with certain exceptions, all other payments, regardless of payer. We propose that we will include such “all other” payments in the numerator and the denominator, and we will exclude payments as specified in the statute. We also propose to exclude patients associated with these excluded payments from the patient count method.

The statue excludes payments made:

• By the Secretary of Defense for the costs of Department of Defense health care programs;
• By the Secretary of Veterans Affairs for the costs of Department of Veterans Affairs health care programs; and
• Under Title XIX in a state in which no Medicaid Medical Home Model or APM is available under the state plan.

We propose that title XIX payments or patients would be excluded in the numerator and denominator for the QP determination unless: (1) A state has at least one Medicaid Medical Home Model or Medicaid APM in operation that is determined to be an Other Payer Advanced APM; and (2) the relevant Advanced APM Entity is eligible to participate in at least one of such Other Payer Advanced APMs during the QP Performance Period, regardless of whether the Advanced APM Entity actually participates in such Other Payer Advanced APMs. This will apply to both the payment amount and patient count methods. We believe this Medicaid exclusion avoids penalizing eligible clinicians who do not have the possibility of participation in an Other Payer Advanced APM under Medicaid. We believe that failing to exclude such payments and/or patients would unduly disadvantage potential QPs by inflating denominators based on circumstances beyond their control. For example, if a state's Medicaid Medical Home Model is determined to be an Other Payer Advanced APM and is operated on a statewide basis, Medicaid payments will be included in the denominator for all eligible clinicians in that state assessed under the All-Payer Combination Option. However, if the state operates such an Other Payer Advanced APM at a sub-state level, and eligible clinicians who do not practice in the geographic area where the Medicaid Medical Home Model is available are not eligible to participate, Medicaid payments would not be included in such eligible clinicians' QP calculations. We will more fully develop the approach to identify Medicaid Medical Home Models and Medicaid APMs, as well as eligible clinician eligibility to participate in them, through subsequent rulemaking.

We seek comment on our proposals to determine exclusions and on how we could account for eligible clinician participation in Medicaid APM or Medicaid Medical Home Models, such as pilots where participation may be intentionally limited by the state.

(3) Payment Amount Method

We propose to calculate an All-Payer Combination Option Threshold Score for eligible clinicians in an Advanced APM Entity using the payment amount method, which would then be compared to the relevant QP Payment Amount Threshold and Partial QP Payment Amount Threshold to make a QP determination.

(a) Threshold Score Calculation

(i) In General

We propose to calculate the All-Payer Threshold Score for eligible clinicians in an Advanced APM Entity (or an eligible clinician that participates in multiple APMs, as this exception is discussed above) by dividing the value described under paragraph (ii) by the value described under paragraph (iii). This calculation would result in a percent value Threshold Score that CMS would compare to the QP Payment Amount Threshold and the Partial QP Payment Amount Threshold to determine the QP status of the eligible clinicians for the payment year. The calculations occur in two steps because there is a Medicare QP Threshold and an All-Payer QP Threshold. The formula for determining the payment Threshold Score is: Threshold Score = A/B, where:

A = The numerator value under paragraph (ii) below

B = The denominator value under paragraph (iii) below

(ii) Numerator

We propose that the numerator would be the aggregate of all payments from all other payers, except those excluded under sections 1833(z)(2)(B)(ii)(I) and (C)(ii)(I) of the Act, to the Advanced APM Entity's eligible clinicians—or the eligible clinician in the event of an individual eligible clinician assessment—under the terms of all Other Payer Advanced APMs during the QP Performance Period. For example, if a beneficiary is attributed to an ACO and sees a clinician outside that ACO, payments made to the non-ACO clinician would not count towards this numerator, even if the ACO is an Other Payer Advanced APM. Medicare Part B covered professional services will be calculated under the All-Payer Combination Option in the same manner as it is for the Medicare Option.

(iii) Denominator

We propose that the denominator would be the aggregate of all payments from all other payers, except those excluded under sections 1833(z)(2)(B)(ii)(I) and (C)(ii)(I) of the Act, to the Advanced APM Entity's eligible clinicians—or the eligible clinician in the event of an individual eligible clinician assessment—during the QP Performance Period. The portion of this amount that relates to Medicare Part B covered professional services will be calculated under the All-Payer Combination Option in the same manner as it is for the Medicare Option.

(b) Examples of Payment Amount Threshold Score Calculation

In this example, an Advanced APM Entity participates in a Medicare ACO initiative, a commercial ACO arrangement, and a Medicaid APM. Each of the APMs is determined to be an Advanced APM. In the QP Performance Period for payment year 2021 (proposed in this proposed rule to be 2019), the Advanced APM Entity receives the following payments:

In Table 41, the Advanced APM Entity meets the minimum Medicare threshold (30% >25%). However, it falls short of the QP Payment Amount Threshold (43% <50%). In this case, the Advanced APM Entity would meet the Partial QP Payment Amount Threshold (43% >40%).

Another Advanced APM Entity in the same year receives the following payments:

In Table 42, the Advanced APM Entity meets the minimum Medicare threshold (40% >25%). It also exceeds the QP Payment Amount Threshold (61% >50%). In this case, the eligible clinicians in the Advanced APM Entity would become QPs.

We seek comment on the payment amount method described in this proposal and any potential alternative approaches.

(4) Patient Count Method

We propose to calculate a Threshold Score for the eligible clinician group in an Advanced APM Entity—or eligible clinician in the exception situations under sections II.F.5 and II.F.6 of this preamble—using the patient count method, which would then be compared against the relevant QP Patient Count Threshold and Partial QP Patient Count Threshold to determine the QP status of an eligible clinician for the year based on the higher of the two values.

(a) Threshold Score Calculation

(i) In General

We propose that the Threshold Score calculation for the patient count method would include patients for whom the eligible clinicians in an Advanced APM Entity furnish services and receive payment under the terms of an Other Payer Advanced APM, with certain exceptions as outlined in the previous section. This calculation would result in a percent value Threshold Score that CMS would compare to the QP Patient Count Threshold and the Partial QP Patient Count Threshold to determine the eligible clinicians' QP status for the payment year. The calculations occur in two steps as there is a Medicare Threshold requirement and an All-Payer Threshold requirement. The formula for determining the patient count Threshold Score is:

Threshold Score = A/B,

where:

A = The numerator value under paragraph (iii) below.

B = The denominator value under paragraph (iv) below.

(ii) Unique Patients

First, we propose that, like the Medicare Option, the patient count method under the All-Payer Combination Option would only count unique patients, with multiple eligible clinicians able to count the same patient. Similarly, we propose to count a single patient, where appropriate, in the numerator and denominator for multiple different Advanced APM Entities when counting the number of beneficiaries under this method section II.F.6 of this preamble. We also propose that CMS will not count any patient more than once for any single Advanced APM Entity. In other words, for each Advanced APM Entity, CMS will count each unique patient one time in the numerator, and one time in the denominator.

We believe that counting patients this way maintains integrity by preventing double counting of patients within an Advanced APM Entity while recognizing the reality that patients often have relationships with eligible clinicians in different organizations. We hope to avoid distorting patient counts for such overlap situations, especially in Advanced APM Entity-dense markets.

We seek comment on our proposal for counting patients and on alternative methods for counting beneficiary overlaps across Advanced APM Entities.

(iii) Numerator

We propose that the numerator would be the number of unique patients to whom eligible clinicians in the Advanced APM Entity furnish services that are included in the measures of aggregate expenditures used under the terms of all of their Other Payer Advanced APMs during the QP Performance Period, plus the patient count numerator for Advanced APMs. A patient would count in the non-Medicare portion of this numerator only if, as stated above, the eligible clinician furnishes services to the patient and receives payment(s) for furnishing those services under the terms of an Other Payer Advanced APM.

(iv) Denominator

We propose that the denominator would be the number of unique patients to whom eligible clinicians in the Advanced APM Entity furnish services under all non-excluded payers during the QP Performance Period.

(b) Examples of Patient Count Threshold Score Calculation

In the QP Performance Period for payment year 2021 (proposed to be 2019 under this proposed rule) the Advanced APM entity experienced the following patient counts:

In Table 43, the Advanced APM Entity meets the minimum Medicare threshold (30% >20%). However, it falls short of the QP Patient Count Threshold (30% <35%). In this case, the Advanced APM Entity would meet the Partial QP Patient Count Threshold (30% >25%).

Another Advanced APM Entity in the same year experienced the following patient counts:

In Table 44, the Advanced APM Entity meets the minimum Medicare threshold (40% > 20%). It also exceeds the minimum QP Patient Count Threshold (61% > 35%). In this case, the eligible clinicians in the Advanced APM Entity would become QPs.

We seek comment on the patient count method described above and any potential alternative approaches.

d. Submission of Information for Assessment Under the All-Payer Combination Threshold Option

Under sections 1833(z)(2)(B)(ii)(III) and (C)(ii)(III), an eligible clinician can only become a QP using the All-Payer Combination Option by providing the Secretary such information as is necessary for the Secretary to determine whether an Other Payer APM is an Other Payer Advanced APM and to determine the eligible clinician's Threshold Score under section II.F.7.c of this preamble. To be considered under the All-Payer Combination Option we propose that APM Entities or individual eligible clinicians must submit by a date and in a manner determined by CMS: (1) Payment arrangement information necessary to assess whether each Other Payer APM is an Other Payer Advanced APM, including information on financial risk arrangements, use of certified EHR technology, and payment tied to quality measures; and (2) for each Other Payer APM, the amounts of revenues for services furnished through the arrangement, the total revenues from the payer, the numbers of patients furnished any service through the arrangement (that is, patients for whom the eligible clinician is at risk if actual expenditures exceed projected expenditures), and the total numbers of patients furnished any service through the payer. CMS would then assess the characteristics of the Other Payer APMs to determine if they are Other Payer Advanced APMs and would notify the APM Entities and/or eligible clinicians of the Other Payer Advanced APM determinations based on their submissions. We propose further, that an Other Payer Advanced APM is required to have an outcome measure. If an Other Payer Advanced APM has no outcome measure, the Advanced APM Entity must attest that there is no applicable outcome measure on the MIPs list. CMS intends to establish specific requirements regarding the timing and manner of submission of such information through future rulemaking.

At this time, we seek comment from stakeholders on the specific types of payment arrangement information that would be necessary to assess whether an Other Payer APM is an Other Payer Advanced APM, and the format in which CMS could reasonably expect to receive this information. We seek comment on the level of detail which CMS should require, and whether certain pieces of information would be most easily submitted directly from individual eligible clinicians or from an APM Entity. We also seek comment on the timing of when CMS could expect to receive this information from individual eligible clinicians and Advanced APM Entities for a performance year. In addition, we seek comment on the proposed requirement that an Other Payer Advanced APM must have an outcome measure.

We seek comment on the possibility of receiving information on Other Payer APMs and their participants directly from other payers in order to minimize reporting burden for APM Entities and eligible clinicians. We seek comment on the extent to which collecting voluntary submissions of data from other payers could reduce burden and increase program integrity through more accurate determinations of QP status based on payment or patient threshold calculations for Other Payer Advanced APMs. Likewise, we seek comment on the extent to which such data collection is operationally feasible or could infringe upon other payers' interests in maintaining the confidentiality of their business practices.

In addition, we propose to make early Other Payer Advanced APM determinations on other payer arrangements if sufficient information is submitted at least 60 days before the beginning of a QP Performance Period. This would allow CMS to offer eligible clinicians advance notice of their prospects of achieving QP status in the event they are assessed under the All-Payer Combination Option. This early determination would be considered final for the QP Performance Period based on the Other Payer APM information submitted. If new information is submitted based on a change in the Other Payer APM during the QP Performance Period, the initial determination could be subject to review and revision. We also propose that, to the extent permitted by federal law, CMS would maintain confidentiality of certain information that the Advanced APM Entities and/or eligible clinicians submit regarding Other Payer Advanced APM status in order to avoid dissemination of potentially sensitive contractual information or trade secrets. We propose that, unlike our proposal for Advanced APM determinations, the Other Payer Advanced APM determinations would be made available directly to participating APM Entities and eligible clinicians rather than through public notice, and we would explain how and within what timeframes such notifications will occur in subregulatory guidance. CMS may consider publicly releasing information on Other Payer Advanced APMs on the CMS Web site with general and/or aggregate information on the payers involved and the scopes of such agreements.

We seek comment on the proposed timing and method of feedback to Advanced APM Entities and eligible clinicians regarding the status of Other Payer Advanced APMs for which they have submitted information and on the proposed early determination process and the ability of Advanced APM Entities and eligible clinicians to submit sufficient information prior to the beginning of a QP Performance Period. We also seek comment on the types of information that contain potentially sensitive information.

The information submitted to determine whether an eligible clinician is a QP under the All-Payer Combination Option may be subject to audit, and eligible clinicians and Advanced APM Entities will be required to maintain copies of any supporting documentation. If an audit reveals a material discrepancy in the information submitted to CMS, and such discrepancy affected the eligible clinician's QP status, the APM Incentive Payment may be recouped. Providing false information may reflect a false claim subject to investigation and prosecution. We may provide further details on the audit and recoupment process under the All-Payer Combination Option in future rulemaking.

8. APM Incentive Payment

The APM Incentive Payment is specified under section 1833(z)(1) of the Act.

a. Amount of the APM Incentive Payment

This section describes our proposal for calculating the amount of the APM Incentive Payment and accounts for the specific scenarios outlined under sections 1833(z)(1)(A)(i) and 1833(z)(1)(A)(ii) of the Act. This section also describes the process by which CMS proposes to disburse these APM Incentive Payments to QPs.

In accordance with section 1833(z)(1)(A) of the Act, CMS will make an APM Incentive Payment for a year to eligible clinicians that achieve QP status for the year during years 2019 through 2024. In accordance with the statute, we propose that this APM Incentive Payment shall be equal to 5 percent of the estimated aggregate amounts paid for Medicare Part B covered professional services furnished by the eligible clinician from the preceding year across all billing TINs associated with the QP's NPI.

(1) Incentive Payment Base Period

The incentive payment base period is the range of dates that will be used to calculate the estimated aggregate payment amounts for the year preceding the QP payment year that will serve as the basis for the incentive payment. Section 1833(z)(1)(A) of the Act states that in calculating the amount that is equal to 5 percent of the estimated aggregate payment amounts for Medicare Part B covered professional services under this part for the preceding year, the payment amount for the preceding year may be an estimation for the full preceding year based on a period of such preceding year that is less than the full year. We believe this provision gives CMS flexibility in determining the incentive payment base period. We propose to use the full calendar year prior to the payment year as the incentive payment base period from which to calculate the estimated aggregated payment amounts.

When determining the time period for the incentive payment base period, we considered using a partial calendar year and a completion factor to forecast and account for the remainder of claims that would be billed during the remainder of the calendar year. However, there are instances where eligible clinician practice patterns change during a given period of time. For example, an eligible clinician may begin practicing, retire, change practice locations, or switch between full-time and part-time; or there could be seasonal fluctuations in an eligible clinician's practice. Given the possible variability in billings and payments over a calendar year, we believe an incentive payment base period of less than one year would produce a less accurate estimated aggregated payment amount and could potentially disadvantage some eligible clinicians based on the circumstances of their practice in a given year.

Using a complete calendar year of claims would allow for the most accurate representation of the covered professional services delivered by each eligible clinician, which we believe outweighs a modest potential delay in making the APM Incentive Payment. We seek comment on our proposal to use the entire preceding calendar year as the incentive payment base period.

(2) Timeframe of Claims

Section 1833(z)(1)(A) of the Act directs CMS to make the APM Incentive Payment in a lump sum on an annual basis “as soon as practicable.” We believe that, in implementing this provision, it is important to balance the desire for accuracy in the data used to calculate the APM Incentive Payment with the desire to expedite the payments so that the APM Incentive Payments are made in an appropriate and timely manner.

We propose to calculate the APM Incentive Payment based on data available 3 months after the end of the incentive payment base period in order to allow time for claims to be processed. For example, for the 2019 payment year, we would capture claims submitted with dates of service from January 1, 2018 through December 31, 2018 and processing dates of January 1, 2018 through March 31, 2019. We believe that 3 months of claims run-out is sufficient to conduct the APM Incentive Payment calculations in an accurate and timely manner. This methodology is consistent with the claims run-out timeframes used for reconciliation payments in several current APMs, such as the Shared Savings Program, the Pioneer and Next Generation ACO Models, and CEC. We seek comment on the potential use of a completion factor. We note that several current APMs apply the 3 month claims run-out in conjunction with a completion factor. However, where a completion factor may be appropriate for payments based on claims submitted by groups of providers and suppliers that may be billing under multiple TINs, we believe that with payments based on individual eligible clinician claims, categorical variability in claims completion across type of eligible clinicians would cause inequitable results.

We recognize that by pulling claims 3 months after the end of the performance year to conduct reconciliation, we would not have a complete claims run-out, especially for the later months of the year. We considered instead proposing a 6 month of claims run-out. On average, 99.3 percent of Medicare claims are processed within 3 months after the end of a calendar year, and 99.8 percent of claims are processed within 6 months after the end of a calendar year. We concluded that the benefit of making the incentive payments 3 months earlier outweighed the benefit of an additional 3 months of processed claims, since the difference in claims completion is extremely small. We also believe that our proposal provides an additional incentive for timely submission of claims at the end of the year because claims for services furnished during the incentive payment base period that are not submitted and processed within this 3 month run-out would not be considered in the incentive payment amount calculations.

We also considered our regulations at § 424.44 and the Medicare Claims Processing Manual (Pub. L. 100-04), Chapter 1, Section 70, stating that Medicare claims can be submitted no later than one calendar year from the date of service. We considered waiting for the full claims run-out 12 months after the end of the performance year, but were concerned that this approach would significantly delay the timing of the incentive payments and possibly dilute their effect as a reward for eligible clinician decisions to participate in APMs. We also believe that such a significant delay would not be consistent with the statutory intent of making payments as soon as practicable.

In summary, for the incentive payment base period we propose to use a complete calendar year of claims with 3 months of claims run-out from the end of the calendar year. We believe our proposed approach balances our goals of providing incentive payments in a reasonable timeframe while being able to account for the vast majority (on average, 99.3 percent of claims for) covered professional services. Given these parameters, we estimate that incentive payments could be made approximately 6 months after the end of the incentive payment base period, or roughly mid-way through the payment year. However, we propose that the APM Incentive Payment would be made no later than one year from end of the incentive payment base period. We do not propose to set a specific deadline mid-way during the payment year because we believe doing so could pose operational risks in the event that 6 months is impracticable in a given year for reasons that CMS cannot predict. We seek comment on our proposed timing of the incentive payment base period.

(3) Treatment of Payment Adjustments in Calculating the Amount of APM Incentive Payment

Part B covered professional services under the Medicare PFS are currently subject to several statutory provisions that are geared towards improving quality and efficiency in service delivery. Eligible clinicians are subject to payment adjustments under: The Medicare EHR Incentive Program for Eligible Professionals (MU), the PQRS, and the VM. Beginning in 2019, the MIPS adjustment, as described in section II.E.5, will replace payment adjustments under the MU, PQRS, and VM for all MIPS eligible clinicians. These special payment provisions directly adjust the payment amount that eligible clinicians receive under the PFS. In contrast, we consider the APM Incentive Payment to be separate from, and, as indicated under section 1833(z)(1)(A) of the Act, in addition to the amount of payments made for covered professional services under the Medicare PFS.

We propose to exclude the MIPS, VM, MU and PQRS payment adjustments when calculating the estimated aggregate payment amount for covered professional services upon which to base the APM Incentive Payment amount. For example, a QP who receives an upward fee adjustment during 2018 in VM would not see that adjustment reflected in the estimated aggregate payment amount for covered professional services used to calculate his or her APM Incentive Payment in 2019. Similarly, a QP who receives a downward fee adjustment during 2018 in VM would not see that amount reflected in the aggregate payment amount for the APM Incentive Payment.

We believe this proposed policy is most consistent with the specification in section 1833(z)(1)(A) of the Act that the APM Incentive Payment is based on the estimated aggregate payment amounts for “such” covered professional services for the preceding year, which refers to the Part B covered professional services furnished by the particular eligible clinician.

While we considered the alternative of including these performance-related payment adjustments in calculating the APM Incentive Payment, we are concerned that such a policy would create incentives that are not aligned with the intent of the APM Incentive Payment. As previously stated in our policy principles, we believe that the APM Incentive Payment is best viewed as a complementary reward for eligible clinicians that have a substantial degree of participation in the most advanced APMs and deliver high-value care, not an evaluation of their performance within the APM or in another statutorily required performance-based payment adjustment.

For example, the incentive payment base period for the 2019 payment year will be 2018, and any QP in payment year 2019 may have quality payment adjustments from the PQRS, MU, and VM payment provisions, which affect the amount of incentive payment for that year, in the incentive payment base period. The PQRS, MU, and VM payment adjustments will sunset at the end of 2018. In addition, in 2020 and later, eligible clinicians who were subject to MIPS in the previous performance year and become newly qualified as QPs may have MIPS quality payment adjustments during the base period affecting their APM Incentive Payment amounts for that period. We do not believe the intent of the APM Incentive Payment is to further magnify the currently existing and future payment adjustments because of overlapping time periods.

We also proposed in section II.F.6.b.(1) to account for payment adjustments in the QP determination process in the same manner as when calculating the amount of the APM Incentive Payment. If we were to include statutory payment adjustments when determining QP status, there could be situations where an eligible clinician could become a QP because of a positive payment adjustment amount, or conversely, there could be situations where an eligible clinician would not meet the QP threshold because of a negative payment adjustment. We believe that our proposal to not include payment adjustments when determining QP status for a year, or when calculating the amount of the APM Incentive Payment, allows CMS to assess all eligible clinicians on the same merits throughout the entire QP determination and APM Incentive Calculation process. We do not believe the intent of the statute was to enhance or negate an eligible clinician's opportunity to become a QP in a given performance year, or to enhance or negate the amount of APM Incentive Payment a QP receives, based on factors that are extraneous to APM participation.

We seek comment on this proposed approach to coordinating the various Medicare Physician Fee Schedule payment adjustments when calculating the amount of the APM Incentive Payment.

(4) Treatment of Payments for Services Paid on a Basis Other Than Fee-For-Service

We recognize that many APMs use incentives and financial arrangements that differ from usual fee schedule payments. Section 1833(z)(1)(A)(i) of the Act requires us to establish policies for payments that are made to an Advanced APM Entity rather than directly to the QP. Section 1833(z)(1)(A)(ii) of the Act requires us to establish policies for when payment is made on a basis other than fee-for-service. For the purposes of this proposed regulation, we place such payments into three categories: Financial risk payments, supplemental service payments, and cash flow mechanisms.

Financial risk payments are non-claims-based payments, based on performance in an APM when an APM Entity assumes responsibility for the cost of a beneficiary's care, whether it be for an entire performance year, or for a shorter duration of time, such as over the course of a defined episode of care. We note that in the context of categorizing these types of payments as “financial risk payments,” we refer to payments that may be based on the cost of a beneficiary's care and do not necessarily limit these payments to financial arrangements that would require an APM Entity to accept downside risk. For instance, we would consider the shared savings payments to ACOs in all tracks of the Shared Savings Program to be financial risk payments. We would also consider net payment reconciliation amounts from CMS to an Awardee (or vice versa) under the BPCI Initiative, and reconciliation payments from CMS to a participant hospital or repayment amounts from a participant hospital to CMS under the CJR model to be examples of financial risk payments.

We propose to exclude financial risk payments such as shared savings payments or net reconciliation payments, when calculating the estimated aggregate payment amount. Financial risk payments are not for specific Medicare Part B covered professional services; rather they are for performance in an APM. Therefore, we believe their inclusion in the estimated aggregate payment amount would be inconsistent with the statutory language and our stated policy principles. In addition, the difficulty of disaggregating payments to individual QPs and the lagged timing of some financial risk payments creates significant policy and operational barriers that we do not believe are in line with our objective of making APM Incentive Payments in a timely manner.

Supplemental service payments are Medicare Part B payments for longitudinal management of a beneficiary's health, or for services that are within the scope of medical and other health services under Medicare Part B that are not separately reimbursed through the physician fee schedule. Often these are per-beneficiary per-month (PBPM) payments that are made for care management services or separately billable services that share the goal of improving quality of care overall, enabling investments in care improvement, and reducing Medicare expenditures for services that could be avoided through care coordination. For example, the OCM makes a per beneficiary Monthly Enhanced Oncology Services (MEOS) payment to practices for care management and coordination during episodes of care initiated by chemotherapy treatment.

We propose to determine on a case-by-case basis whether certain supplemental service payments are in lieu of covered services that are reimbursed under the PFS. In cases where payments are for covered services that are in lieu of services reimbursed under the PFS, those payments would be considered covered professional services and would be included in the APM Incentive Payment amounts. We propose to include a supplemental service payment in calculation of the APM Incentive Payment amount if it meets all of the following 4 criteria:

(1) Payment is for services that constitute physician services authorized under section 1832(a) of the Act and defined under section 1861(s) of the Act.

(2) Payment is made for only Part B services under the first criterion above, that is, payment is not for a mix of Part A and Part B services.

(3) Payment is directly attributable to services furnished to an individual beneficiary.

(4) Payment is directly attributable to an eligible clinician.

Table 45 provides an example of how a limited number of supplemental service payments in currently operating or recently announced APMs would be considered with respect to our proposed criteria. We further propose to establish a process by which we notify the public of the supplemental service payments in all APMs and identify the supplemental service payments that meet our proposed criteria and would be included in the APM Incentive Payment calculations. Similar to our proposal to announce Advanced APM determinations, we propose to post an initial list of supplemental service payments that would be included in our APM Incentive Payment calculations on the CMS Web site. As new APMs are announced, CMS would include its determination of whether an APM related supplemental service payment would be included in our APM Incentive Payment calculations, if applicable, in conjunction with the first public notice of the APM. We propose to update the list of supplemental service payments that would be included in our APM Incentive Payment calculations on an ad hoc basis, but no less frequently than on an annual basis.

We seek comment on this proposed approach to include certain supplemental service payments when calculating the basis for the amount of the APM Incentive Payment. Specifically, we seek comment on our proposed criteria to include supplemental service payments in the basis for the APM Incentive Payment amounts, and our proposed method for announcing which supplemental service payments would be included in the basis for the APM Incentive Payment amounts.

Cash flow mechanisms involve changes in the method of payments for services furnished by providers and suppliers participating in an APM Entity. In themselves, cash flow mechanisms do not change the overall amount of payments. Rather, they change cash flow by providing a different method of payment for services. An example of a cash flow mechanism is the population-based payment (PBP) available in the Pioneer ACO Model and the Next Generation ACO Model. PBP provides ACOs with a monthly lump sum payment in exchange for a percentage reduction in Medicare fee-for-service payments to certain ACO providers and suppliers.

For expenditures affected by cash flow mechanisms, we propose to calculate the estimated aggregate payment amount using the payment amounts that would have occurred for Part B covered professional services if the cash flow mechanism had not been in place. For example, for QPs in an ACO receiving PBP with a 50 percent reduction in fee-for-service payments, we would use the amount that would have been paid for Part B covered professional services in the absence of the 50 percent reduction. Cash flow mechanisms represent a potential reallocation of dollars between eligible clinicians and entities for specific purposes related to care improvement. We do not believe that the presence of certain cash flow mechanisms should impact the APM Incentive Payment amount, and we do not intend for the APM Incentive Payment to influence the use or attractiveness of cash flow mechanisms in current and future APMs.

We recognize that new payment methods and financial arrangements may be developed that do not fit into these categories as described. For instance, in the recently announced CPC + Model, the supplemental service payments (that is, the CMFs) would meet all of our proposed criteria to be included in the APM Incentive Payment calculations. The CMFs are for Medicare Part B covered professional services, and the CMF payments will only cover Medicare Part B covered professional services. The CMF amounts will be risk adjusted based on each individual beneficiary's HCC risk scores; therefore these payments will be attributable to individual beneficiaries. Additionally, the attribution method in the CPC + Model uses a combination of the TIN/Individual NPI/Practice Address when attributing an individual beneficiary to a CPC Practice site. However, the CMF payments for attributed beneficiaries are aggregate payments is made to each CPC Practice Site. We recognize that throughout the course of a QP Performance Period more than one NPI may furnish covered professional services to an attributed beneficiary. If that occurs, more than one NPI could potentially receive the corresponding CMF for that eligible beneficiary. We do not believe it would be appropriate to count the same CMF for more than one NPI. Therefore, (assuming that the CPC + Model is deemed an Advanced APM and the APM Entity group achieves the QP threshold for a QP Performance Period), we could split the CMF amounts equally between the multiple NPIs, or we could develop a method to “assign” the NPI for which the CMFs would be counted in their APM Incentive Payment calculation based on the plurality of visits with that beneficiary.

We seek comment on the methods that CMS could use to allocate the supplemental service payments to individual NPIs in these types of scenarios in which payment for a supplemental service payment is made in the aggregate to an APM Entity.

We also recognize that payment methods and financial arrangements may evolve over time and would need to be addressed in future rulemaking. CMS seeks comment on the proposals for accounting for risk-based payments, supplemental service payments, and cash flow mechanisms when calculating the amount of APM Incentive Payment.

(5) Treatment of Other Incentive Payments in Calculating the Amount of APM Incentive Payments

Section 1833(z)(1)(D) of the Act specifies that CMS shall not include certain existing Medicare incentive payments in the calculation of the APM Incentive Payment. This includes payments made under section 1833 of the Act (subsections (m), (x), and (y)). Section 1833(m) of the Act describes the HPSA Physician Bonus Program. The HPSA Physician Bonus Program provides bonus payments to physicians for physicians' services furnished in geographic areas that are designated as of December 31 of the prior year by the HRSA as HPSAs under section 332 (a)(1)(A) of the PHS Act. The HPSA bonus payment is 10 percent of the Medicare Part B payment amount for the service; and this bonus is paid as a quarterly lump sum payment.

Subsection (x) describes the Primary Care Incentive Payment (PCIP) program. The PCIP payment amount was 10 percent of the payment amount for Medicare Part B primary care services furnished by primary care practitioners for whom primary care services accounted for at least 60 percent of their allowed fee-for-service charges in a prior qualification period. For purposes of the PCIP program, primary care practitioners were defined as those physicians with certain Medicare specialty codes and certain types of non-physician practitioners. The PCIP payment was made on a quarterly basis. This bonus payment expired under the statute on December 31, 2015.

Subsection (y) describes the HPSA Surgical Incentive Payment (HSIP). For major surgical procedures furnished by physicians with a primary specialty designation of “general surgeon” in HPSAs (under section 332(a)(1)(A) of the PHS Act), physicians received an additional 10 percent bonus payment in addition to the amount of payment that would otherwise be made. This additional payment was combined with any other HPSA payment outlined in 1833 of the Act, subsection (m), and was paid on a quarterly basis. This bonus payment expired under the statute on December 31, 2015.

Section 1833(z)(1)(D) of the Act also directs CMS not to include APM Incentive Payments when calculating payments made under section 1833 (subsections (m), (x), and (y)) of the Act. CMS considers the APM Incentive Payment to be separate from the incentive payments as previously discussed in this section and has established procedures to ensure that the APM Incentive Payment will not be included when calculating the amount of incentive payments made under section 1833 (subsections (m), (x), and (y)) of the Act.

(6) Treatment of the APM Incentive Payment in APM Calculations

Section 1833(z)(1)(C) states that the amount of the APM Incentive Payment shall not be taken into account for purposes of determining actual expenditures under an APM and for purposes of determining or rebasing any benchmarks used under the APM. As a lump sum payment, the APM Incentive Payments will be made outside of the Medicare claims processing system. Current APMs, such as the Medicare ACO initiatives and the CJR model, have established procedures for ensuring that lump sum payments from other APMs are accounted for when they do their APM reconciliations and rebasing calculations. We anticipate that each APM will have in place a procedure to avoid counting APM Incentive Payments toward determining actual expenditures or rebasing any benchmarks under the APM.

b. Services Furnished Through CAHs, RHCs, and FQHCs

(1) Critical Access Hospitals (CAHs)

Eligible clinicians who furnish services at Critical Access Hospitals (CAHs) that have elected to be paid for outpatient services under section 1834(g)(2)(B) of the Act (Method II) will be eligible to become QPs and receive the APM Incentive Payment if they are part of an Advanced APM Entity. As stated in section II.F.6.d.(1) of this proposed rule, professional services furnished at a Method II CAH are considered “covered professional services” because they are furnished by an eligible clinician and payment is based on the Medicare Physician Fee Schedule. Therefore, the APM Incentive Payment would be based on the amounts paid for those services attributed to the eligible clinician, as identified using the attending NPI included on a submitted claim, in the same manner as all other covered professional services.

For an eligible clinician who becomes a QP based on covered professional services furnished at a Method II CAH, we propose that the APM Incentive Payment would be made to the CAH TIN that is affiliated with the Advanced APM Entity. This proposal is consistent with how CMS proposes to make the APM Incentive Payment to eligible clinicians who practice at locations other than Method II CAHs. We seek comment on this proposal.

(2) Rural Health Clinics (RHCs) and Federally Qualified Health Centers (FQHCs)

As explained in section II.F.6.d.(2) of this preamble, payment for services furnished by eligible clinicians in RHCs and FQHCs is not reimbursed under or based on the PFS. Therefore, professional services furnished in those settings would not constitute covered professional services under section 1848(k)(3)(A) of the Act and would not be considered part of the amount upon which the APM Incentive Payment is based. For eligible clinicians that practice in RHCs or FQHCs, this does not preclude the inclusion of payment amounts for covered professional services furnished by those eligible clinicians in other settings. This only excludes payments made for RHC and FQHC services furnished by the eligible clinicians. For example, an eligible clinician may practice at both an FQHC and with a separate physician group practice that receives payment under the PFS. If the eligible clinician becomes a QP under the methodologies described in II.F.6, whether based on their participation in an Advanced APM Entity that includes the FQHC as outlined in section II.F.6.d.(2) or based on their participation in an Advanced APM Entity that includes the separate physician group practice, or both, only the eligible clinician's payments for covered professional services at the separate physician group practice setting would form the basis amount for the APM Incentive Payment.

c. Payment of the APM Incentive Payment

(1) Payment to the QP

The APM Incentive Payment, as described in this section, will be made to QPs who are identified by their unique TIN/NPI combination as participants in an Advanced APM Entity on a CMS maintained list.

We received a number of comments on the MIPS and APMs RFI regarding the process by which we should make the APM Incentive Payment. One commenter suggested that we give QPs the opportunity to select where they want the APM Incentive Payment to be sent, while another suggested that we send payments directly to the individual eligible clinician. A number of commenters recommended that CMS make the APM Incentive Payments directly to the Advanced APM Entity. Additionally, some commenters noted that making payments directly to the Advanced APM Entity would allow Advanced APM Entities to fairly and accurately allocate incentive payments in accordance with the shared risk for individual eligible clinicians in the APM Entity. We thank the commenters for their feedback.

After consideration of these comments, we propose that for eligible clinicians that are QPs, CMS would make the APM Incentive Payment to the TIN that is affiliated with the Advanced APM Entity through which the eligible clinician met the threshold during the QP performance period. For both individual eligible clinicians and group practices, CMS uses the TIN as the billing unit. Earlier in this section, we proposed that the APM Incentive Payment would be calculated across all billing TINs associated with an NPI. Medicare has the ability to track all unique TIN/NPI combinations associated with an individual NPI, including which TINs are affiliated with an Advanced APM entity. We considered making separate payments for each TIN/NPI combination associated with the individual eligible clinician's APM Incentive Payment, similar to the current PQRS incentive payment program. Under the current PQRS incentive payment program, incentive payments are paid to the holder of the TIN, aggregating individual incentive payments for groups that bill under one TIN. For eligible clinicians who submit claims under multiple TINs, CMS groups claims by TIN for payment purposes and any incentive payments earned are paid to that specific TIN. As a result, an eligible clinician with multiple TINs who qualifies for the PQRS incentive payment under more than one TIN would receive a separate PQRS incentive payment associated with each TIN.

However, we believe that making the APM Incentive Payments to the TIN associated with the Advanced APM Entity during the QP Performance Period would be most consistent with section 1833(z) of the Act to incentivize participation in Advanced APMs. Rewarding TINs that are not involved in an Advanced APM for the activity of their constituent NPIs through separate entities seems to be antithetical to the objective of the APM Incentive Payment. We also believe that making the APM Incentive Payments to the TIN associated with the Advanced APM Entity during the QP Performance Period is most consistent with section 1833(z) of the Act with regards to making the APM Incentive Payments to eligible clinicians who become QPs. We also hope to promote simplicity and foster QP awareness of the recipient of the APM Incentive Payment that is based on their activity within APMs. We also believe that making multiple separate payments would increase complexity for both CMS and eligible clinicians.

Additionally, we proposed in section II.F.5 of this preamble, that in order to be a QP, an eligible clinician would need to be identified using a CMS maintained participation list of eligible clinicians for the Advanced APM entity. This proposal would allow CMS to track the APM Entity/TIN/NPI identifiers for each individual eligible clinician, and we believe that this information will allow CMS to determine which of the QP's TINs should receive the APM Incentive Payment.

We recognize that there may be scenarios in which an individual eligible clinician may change his or her affiliation between the QP Performance Period and the payment year such that the eligible clinician no longer practices at the TIN affiliated with the Advanced APM Entity. In this instance, we propose to make the APM Incentive Payment to the TIN provided on the eligible clinician's CMS-588 EFT Application. This proposal is consistent with the process that CMS has used to make incentive payments under other programs, such as the PCIP program.

We seek comment on our proposal to make the APM Incentive Payments to the TIN affiliated with the Advanced APM Entity through which an individual eligible clinician becomes a QP during the QP Performance Period and our proposal to make the APM Incentive Payment when an eligible clinician no longer practice at the TIN affiliated with the Advanced APM Entity. We also seek comment on alternative options that maintain the goals of equity and simplicity, and of using the APM Incentive Payment to encourage and reward participation in Advanced APMs.

(2) Exceptions

As discussed in the exceptions section II.F.5 of this preamble, we recognize that there may be instances where none of the Advanced APM Entities with which an individual eligible clinician participates meets the QP threshold. In this instance, we have proposed to assess the eligible clinician individually, using services furnished through all Advanced APM Entities during the QP Performance Period. When we make the QP determination at the individual eligible clinician level, we propose to split the APM Incentive Payment amount proportionally across all of the QP's TINs associated with Advanced APM Entities. For example, if an eligible clinician is a QP who participates in two APMs (APM 1 and APM 2), and has 75 percent of his or her payments (or patients) used to make the QP determination through APM 1 and 25 percent of his or her payments (or patients) used to make the QP determination APM 2, we would make 75 percent of the APM Incentive Payment to the TIN affiliated with APM 1, and 25 percent of the APM Incentive Payment to the TIN affiliated with APM 2. We believe that splitting the APM Incentive Payment in this way is most consistent with section 1833(z) of the Act, as well as our goal to encourage participation in APMs. We also believe that splitting the incentive payment in this way appropriately recognizes the several activities of the individual eligible clinician toward achieving the QP threshold.

CMS seeks comment on the proposal to split the APM Incentive Payment among the QP's TINs associated with Advanced APM Entities in instances where the QP determination is made at the individual eligible clinician level. We also welcome comments regarding to which TIN(s) payments should be made in the cases where the QP changes TIN affiliations between the QP Performance Period and the payments of the APM Incentive Payment.

(3) Notification of APM Incentive Payment Amount

We anticipate that the notification of the APM Incentive Payment amount will not occur at the same time as the notification of QP status, but will occur later in the year to allow for accurate calculation and validation of the incentive payment amount. We propose to send notification to both Advanced APM Entities and their individual participating QPs of their APM Incentive Payment amount as soon as CMS has calculated the amount of the APM Incentive Payment and performed all necessary validation of the results. Following our proposed method to notify eligible clinicians of their QP status, as discussed in section II.F.5 of this preamble, we propose that the APM Incentive Payment amount notification would be made directly to QPs in combination with a general public notice that such calculations have been completed for the year. For the direct QP notification, CMS intends to include the amount of APM Incentive Payment and the TIN to which the incentive payments will be made. In the case that a QP determination is made at the individual eligible clinician level, and the incentive payment is split across multiple TINS, CMS intends to identify which TINs we will make the incentive payment, and include the amount of APM Incentive Payment that will be made to each TIN. For the notification to Advanced APM Entities, CMS intends to include the total amount of APM Incentive Payments that will be made to each participating TIN within the Advanced APM Entity, as well as QP specific payment amounts. We believe that this is the most efficient method to disseminate of this information to all QPs.

We seek comment on other methods for the notification of APM Incentive Payment amount. We also seek comment on the content of such notifications so that they may be as clear and useful as possible.

9. Monitoring and Program Integrity

In an effort to accurately award the APM Incentive Payment and preserve the integrity of the Medicare program, we propose that CMS will monitor Advanced APM Entities and eligible clinicians on an ongoing basis for non-compliance with the conditions of participation for Medicare and the terms of the relevant Advanced APMs in which they participate during the QP Performance Period. This will include vetting of applicants to Advanced APMs to determine whether they are in compliance with the conditions of participation of Medicare and ongoing, periodic assessments of Advanced APM Entities and eligible clinicians by APMs in conjunction with the CMS Center for Program Integrity and other relevant federal government departments and agencies. These actions are currently taking place for APMs and will continue in the future as stated in the proposed rule.

We propose that if an Advanced APM terminates an Advanced APM Entity or eligible clinician during the QP Performance Period for program integrity reasons, or if the Advanced APM Entity or eligible clinician is out of compliance with program requirements, CMS may reduce or deny the APM Incentive Payment to such eligible clinicians. In addition, if the APM Incentive Payment is paid during the QP Performance Period and the Advanced APM Entity or eligible clinician is later terminated due to a program integrity matter arising during the QP Performance Period, CMS may recoup all or a portion of the amount of the payment from the entity to which CMS made the payment.

We also propose that CMS will reopen and recoup any payments that were made in error in accordance with procedures similar to those set forth at §§ 405.980 and 405.370 et seq. or established under the relevant APM.

As discussed in section II.F.7.b.(7) of the preamble, we propose that APM Entities and/or eligible clinicians must submit certain information for CMS to assess whether other payer arrangements meet the Other Payer Advanced APM criteria and to calculate the Threshold Score for a QP determination under the All-Payer Combination Option. We also propose that Advanced APM Entities and eligible clinicians must maintain copies of all records related to the All-Payer Combination Option for at least ten years and must provide the government with access to these records for auditing and inspection purposes. If an audit reveals that the information submitted is inaccurate, CMS may recoup the APM Incentive Payment. We note that nothing in this proposed rule limits or restricts the authority of the Office of the Inspector General.

We seek comment on our monitoring and program integrity proposals.

10. Physician-Focused Payment Models

a. Introduction and Overview

Section 101(e)(1) of the MACRA entitled, “Increasing the Transparency of Physician-Focused Payment Models,” adds a new section 1868(c) to the Act. In general, this section establishes an innovative process for individuals and stakeholder entities (stakeholders) to propose physician-focused payment models (PFPMs) to the Physician-Focused Payment Model Technical Advisory Committee (PTAC). A copy of the PTAC's charter, established by the Secretary on January 5, 2016, is available at https://aspe.hhs.gov/charter-physician-focused-payment-model-technical-advisory-committee .

(1) Overview of the Roles of the Secretary, the PTAC, and CMS

Section 1868(c)(2)(A) of the Act requires the Secretary to establish, through notice and comment rulemaking following an RFI, criteria for PFPMs (PFPM criteria), including models for specialist physicians, that could be used by the PTAC in making comments and recommendations on PFPMs. We issued the MIPS and APMs RFI requesting stakeholder input on PFPMs on October 1, 2015, and propose PFPM criteria in this rule, section II.F.10.c. of this proposed rule.

Section 1868(c)(2)(B) of the Act specifies that stakeholders may submit proposals to the PTAC on an ongoing basis for PFPMs that they believe meet the PFPM criteria established by the Secretary. We recognize this statutory directive, but do not propose to define “ongoing basis” because we believe that the process for submitting proposals to the PTAC should be determined by the PTAC.

Section 1868(c)(2)(C) of the Act requires the PTAC to review stakeholders' proposed PFPMs, prepare comments and recommendations regarding whether such PFPMS meet the PFPM criteria established by the Secretary, and submit those comments and recommendations to the Secretary.

The PTAC, established under section 1868(c)(1)(A) of the Act, is an independent committee comprised of 11 members. As required under section 1868(c)(1)(B) of the Act, the initial appointments to the PTAC were made by the Comptroller General of the United States on October 9, 2015. The terms of the first appointed members of the PTAC are intended to be staggered, with the first set of appointments for terms of 1, 2, or 3 years. After the initial appointments, all subsequent appointments would be for terms of 3 years. PTAC members who were among the initial appointments may be reappointed for subsequent 3-year terms. There are no limitations for how many terms a PTAC member may serve. No end date for the PTAC is specified. Section 1868(c)(1)(B)(ii) and (iii) of the Act state that no more than 5 members of the PTAC shall be providers of services or suppliers, or representatives of providers of services or suppliers, and no member of the PTAC shall be an employee of the federal government. We received responses to the MIPS and APMs RFI recommending that CMS ensure that the PTAC is made up of varying ratios of professionals from particular backgrounds. We appreciate these responses; however, section 1868(c) of the Act specifies that the Comptroller General of the United States is to appoint members of the PTAC. Therefore, CMS does not have the authority to appoint members of the PTAC.

Section 1868(c)(2)(D) of the Act requires the Secretary to review the PTAC's comments and recommendations on proposed PFPMs and to post “a detailed response” to those comments and recommendations on the CMS Web site. We received comments on the MIPS and APMs RFI requesting that we review PFPM proposals from stakeholders before they are submitted to the PTAC. We also received comments on the MIPS and APMs RFI requesting that the PTAC review PFPM proposals under development by stakeholders before they are formally submitted to the PTAC. Section 1868(c) of the Act does not require either the PTAC or the Secretary to evaluate proposed PFPMs prior to their submission to the PTAC, nor does it require the Secretary to review and respond to proposed PFPMs that are not reviewed by the PTAC. The PTAC would determine whether and how it may provide feedback on proposed PFPMs. In addition, we do not propose to evaluate PFPM proposals prior to their submission to the PTAC because it might interfere with the PTAC's independent review process.

We also received responses to the MIPS and APMs RFI recommending that all proposed PFPMs that the PTAC recommends to the Secretary should be tested by us. Without being able to predict the volume, quality, or appropriateness of the PFPMs that the PTAC would recommend, we are not in a position to propose a commitment to test all such models. Section 1868(c) of the Act does not require us to test models that are recommended by the PTAC and given a favorable response by the Secretary. However, this does not imply that we would not give serious consideration to proposed PFPMs recommended by the PTAC. The PTAC serves an important advisory role in the implementation of APMs, but there are additional considerations that must be made by the Secretary beyond what is provided by the PTAC, such as competing priorities and available resources. We believe that this flexibility is important because the Secretary, and CMS through its delegated authority to test APMs, must retain the ability to make final decisions on which models to test and when, based on multiple factors including those that the Innovation Center currently uses to determine which payment models to test as described in section II.F.10.d. of this proposed rule.

While we would consider these factors separately from the PTAC's review process, the decision to test a model recommended by the PTAC would not require a second application process to us as speculated by some commenters on the MIPS and APMs RFI; we would review the proposal submitted to the PTAC along with comments from the PTAC, and any other resources we believe would be useful. Proposed PFPMs that the PTAC recommends to the Secretary but that are not immediately tested by us may be considered for testing at a later time. We would continue to test PFPMs that are developed within CMS.

(2) Deadlines for the Duties of the Secretary, the PTAC, and CMS

We received multiple responses to the MIPS and APMs RFI recommending that we establish deadlines for the PTAC's comments and recommendations on proposed PFPMs, the Secretary's response to the PTAC's comments and recommendations, and our testing of PFPMs. We do not propose to set deadlines for these tasks through regulations. We believe that setting a deadline for the PTAC's comments and recommendations would interfere with the PTAC's freedom to govern itself and develop its own process and timeline for reviewing proposed PFPMs. We wish to preserve the PTAC's independence and to give it the freedom to determine how and when it would review proposed PFPMs without rulemaking.

We believe that setting a deadline through rulemaking for the Secretary's review of the PTAC's comments and recommendations, publication of a response to them, and our potential testing of a proposed PFPM submitted to the PTAC is inappropriate because these tasks would take varying amounts of time depending on factors that we cannot predict. Proposed PFPMs may be submitted to the PTAC on “an ongoing basis” in accordance with section 1868(c)(2)(B) of the Act, and given that there may be variation in the number and frequency of proposals, setting a deadline would be difficult. We do not believe we can effectively set deadlines because we do not know how many PFPM proposals the PTAC would receive. The Secretary would need varying lengths of time to review, comment on, and respond to PFPM proposals depending on the volume and nature of each proposal.

We do not believe it would be reasonable to require us to adhere to a deadline in deciding whether to test a particular proposed PFPM. It is important for us to retain the flexibility to test models when it believes that it is the right time to do so, taking into account the other models it is currently testing, any potential design changes to the proposed PFPM, interactions with other our policies, and resource allocation. APMs generally take 18 months for us to develop, although the period of development may vary in length significantly, making a deadline difficult to establish.

We received comments on the MIPS and APMs RFI suggesting that that any proposed PFPM approved by the PTAC should be available immediately for participant enrollment, and that participant enrollment should continue on an ongoing basis. We believe that setting deadlines for testing proposed PFPMs that we decide to test would be inappropriate. Entities need time to complete applications for voluntary models and we need time to review applications and prepare participation agreements for entities to sign. Entities need time to review these participation agreements and to begin planning for implementation of the model. To maintain rigorous evaluation of model outcomes, we also need time to build the necessary model infrastructure for such functions as quality measurement, financial calculations, and payment disbursements, and to coordinate with other payers if they are included in the model's design.

We believe that proposed PFPMs that meet all of the proposed criteria may need less time to go through the development process; however, we cannot guarantee that the development process would be shortened, or estimate by how much it would be shortened. These processes depend on the nature of the PFPM's design and any attempt to impose a deadline on them would not benefit stakeholders because it would not allow us to tailor its review and development process to the needs of the proposed PFPM.

b. Definition of PFPM

(1) Proposed Definition of PFPM

Section 1868(c) of the Act does not define the term “physician-focused payment model” (PFPM). In § 414.1465, we are proposing to add the following definition of PFPM: An Alternative Payment Model wherein Medicare is a payer, which includes physician group practices (PGPs) or individual physicians as APM Entities and targets the quality and costs of physician services. We propose to require a PFPM to target physician services. To address physician services, proposed PFPMs may address such elements as physician behavior or clinical decision-making. APM Entities may be individual eligible clinicians, physician group practices (PGPs), or other entities, depending on the payment model's design. Therefore a PFPM must focus on physician services and contain either individual physicians or PGPs as APM Entities, although it may also include facilities or other practitioner types.

We propose to require that PFPMs be designed to be tested as APMs with Medicare as a payer. Other Payer APMs would therefore not be PFPMs. We believe this is an appropriate standard for PFPMs because the Secretary is interested in reviewing comments and recommendations from the PTAC on models that may be tested with Medicare as a payer and because this provision is in section 1868 of the Act, and title XVIII of the Act governs Medicare. A PFPM may include other payers in addition to Medicare under the proposed definition. We believe this definition is appropriate because it would include APMs with arms of their design that would include other payers beyond Medicare, but would not include models that are only Other Payer APMs.

We received many responses to the MIPS and APMs RFI regarding the proposed definition of PFPMs. These recommendations ranged from broadening the definition to include any payment model that alters payment for particular programs or populations to restricting the definition to specialist physicians only. We also received responses to the MIPS and APMs RFI recommending the definition of PFPM be broadened to include other care providers in the definition, such as nurses. We did not accept these suggestions because we believe that a payment model that does not specifically include individual physicians or PGPs would not appropriately be termed physician-focused. While we agree that there is merit in allowing other practitioners and facilities to be included in proposed PFPMs, we do not agree that changing the definition to explicitly include additional care providers or broadening the definition such that physicians or PGPs might not be included would satisfy the statutory directives under section 1868(c) of the Act that promote the development of PFPMs. Defining PFPM to allow the inclusion of other entities and additional targets gives stakeholders more flexibility in their proposals and may lead to models that promote broader participation in PFPMs, greater potential for care redesign, and greater potential for cost reduction.

We do not propose to limit a PFPM to exclusively targeting physicians and physician services because we believe that stakeholders should be able to propose payment models that include additional types of entities, as well as additional services. We do not propose to define PFPM as a payment model that exclusively addresses Medicare FFS payments. A proposed PFPM may also include other payers in addition to Medicare, including Medicaid, Medicare Advantage, CHIP, and private payers, which may promote broader participation in PFPMs and greater potential for cost reduction. If tested as an APM, a PFPM that includes payers in addition to Medicare would include an Other Payer APM as part of its design in addition to an APM.

(2) Relationship Between PFPMs and Advanced APMs

Section 1868(c) of the Act does not require PFPMs to meet the criteria to be an Advanced APM for purposes of the incentives for participation in Advanced APMs under section 1833(z) of the Act, and we do not propose to define PFPMs solely as Advanced APMs. Stakeholders may therefore propose either Advanced APMs or other PFPMs that might lead to better care for patients, better health for our communities, and lower health care spending. We received responses to the MIPS and APMs RFI recommending that all proposed PFPMs selected for testing by us should be Advanced APMs without needing to meet the additional criteria for Advanced APMs. Section 1833(z)(3)(C) and (D) of the Act makes a clear distinction between APMs and Advanced APMs and we do not believe the statutory requirements for Advanced APMs can or should be waived for proposed PFPMs.

We recognize that both stakeholders and the PTAC may want to discuss in their proposals, comments, and recommendations, respectively, whether a proposed PFPM would be an Advanced APM. Therefore, we recommend that stakeholders provide information in their proposal about whether their proposed PFPM might be an Advanced APM as described in section II.F.4 of this proposed rule.

c. Proposed PFPM Criteria

Section 1868(c)(2)(A) of the Act requires the Secretary to establish PFPM criteria for PFPMs, including models for specialist physicians, not later than November 1, 2016. The PFPM criteria would be used by the PTAC to make comments and recommendations on proposed PFPMs to the Secretary. The proposed PFPM criteria are listed in section II.F.10.c.(1) of this rule, and at proposed § 414.1465(b). We have designed these criteria so that they are broad enough to encompass all physician specialties and provide stakeholders with flexibility in designing PFPMs.

We propose PFPM criteria organized into three categories that are consistent with the Administration's strategic goals for achieving better care, smarter spending and healthier people: Payment incentives; care delivery; and information availability. First, we propose a category of criteria that promote payment incentives for higher-value care, including paying for value over volume and providing resources and flexibility necessary for practitioners to deliver high-quality health care.

To address paying for value over volume, we propose a criterion that PFPMs should provide incentives to practitioners to deliver high-quality health care. We believe that the correct incentives are necessary to drive change to improve quality of care. To address this criterion, the PTAC may request or stakeholders may wish to provide information about specific incentives in the proposed PFPM and how they are expected to incentivize quality, or information about any adjustments to payments to APM Entities based on quality performance. Similarly, we believe that it is important for a PFPM to provide sufficient flexibility for practitioners to deliver high-quality care. Flexibility relates to operational feasibility, the PFPM's ability to adapt to accommodate clinical differences in patient subgroups, and the APM Entity's ability to respond to changes in healthcare. To address this criterion, the PTAC may request or stakeholders may wish to provide information about how feasible it would be for APM Entities in the PFPM to deliver high-quality care as defined by the PFPM, and how the model design facilitates and encourages delivery of high-value care with respect to the dynamic and evolving nature of healthcare. In addition, the PTAC may request or stakeholders may wish to provide information about how the proposed PFPM can adapt to accommodate clinical differences in patient subgroups, and how it can adapt to account for changing technology, including new drug therapies.

This category of criteria also aligns with the Innovation Center's statutory authority under section 1115A of the Act to test models aimed to improve care, reduce cost, or achieve both of these goals, by proposing a criterion that assesses to what extent a PFPM proposal is expected to achieve these goals. To address this criterion, the PTAC may request or stakeholders may wish to provide information about specific quality measures included in the PFPM's design, including any prior validation of those measures, and whether any of those measures are patient reported outcome measures or measurements of beneficiary experience of care. We believe estimates of any cost reduction under the PFPM to the most precise extent possible would also be useful in addressing this criterion.

We propose a criterion that the PFPM proposal must pay APM Entities under a payment methodology that furthers the PFPM Criteria. The payment methodology must address how it is different from current Medicare payment methodologies, and why the payment methodology cannot be tested under current payment methodologies. We believe it is necessary for PFPM proposals to contain such a payment methodology because the PTAC is tasked with reviewing payment models and therefore cannot evaluate a proposal without knowing the payment methodology. We believe that the more robust the description of the payment methodology is, the easier it would be for the PTAC to evaluate the impact of the proposed PFPM. In addition, including information about how the proposed PFPM differs from current methodologies and why it cannot be tested under them would allow the PTAC and the Secretary to evaluate how the proposed PFPM could improve on existing methodologies. It is important for the PFPM proposal to describe how the payment methodology is different from current Medicare payment methodologies to show how the PFPM would test differences in payment and their effect on paying for value over volume. It would also help the PTAC and the Secretary to understand why the PFPM would be a significant enough departure that an APM would be required to test it. We recommend that stakeholders include a thorough description of the payment methodology. To be robust, the description of a payment methodology should include the amount of any new payments to proposed APM Entities, such as per beneficiary per month payments, performance-based payments, or shared savings payments. It should also include a methodology for calculation of these payments. It should include information about whether the proposed PFPM could include other payers in addition to Medicare, and if so, the payment methodology proposed for those payers. The payment methodology description should also include information about the use of any payment methods such as bundled payments or capitated payments and a description of the type and degree of financial performance risk assumed by APM Entities. We received comments in response to the RFI suggesting that we accept proposed PFPMs that have different payment methodologies from current APMs such as ACOs and bundled payments. We welcome completely new and innovative ideas for payment methodologies that can improve care while reducing cost.

We also propose to include in the first category a criterion that the PFPM must either aim to solve an issue in payment policy not addressed in the CMS APM portfolio at the time it is proposed or include in its design APM Entities who have had limited opportunities to participate in APMs. We believe this criterion would promote participation in APMs by broadening and expanding our portfolio of APMs in areas such as geographic location, specialty, condition, and illness, without overly limiting proposed PFPMs. We believe that because proposed PFPMs may satisfy this criterion by either addressing a new issue or including a new specialty, the criterion is sufficiently broad to allow stakeholders to submit many proposed PFPMs that could expand the CMS APM portfolio. Physicians and practitioners whose opportunities to participate in other PFPMs with us have been limited to date include, for example, those who have not been able to apply for any other PFPM because one has not been designed that would include physicians and practitioners of their specialty. We propose that a proposed PFPM that includes multiple specialties may meet the PFPM criteria where a minimum of one of the specialties in the proposed PFPM is not currently being addressed by another APM. We believe this reflects the intent of section 1868(c)(2)(A)(i) of the Act which specifically directs the Secretary to establish PFPM criteria, including models for specialist physicians.

We also propose a criterion that a PFPM proposal must have evaluable goals for the impact of cost and quality under the PFPM. To make the decision to expand an APM under section 1115A(c) of the Act, the Secretary must evaluate the model's success. This standard informed our proposed criterion not only because it would be important for any APMs that are tested under section 1115A(c) of the Act, but also because it is necessary for measuring the success of any APM that it be evaluable. It is the evaluation of an APM that tells us whether the APM is successful in reducing cost or improving quality. We believe that the more detailed the information regarding how the impact of a proposed PFPM would be evaluated, the easier it would be for the PTAC to evaluate the impact of the proposed PFPM in terms of potential expansion, as well as in terms of incentivizing high quality care and reducing costs. To address this criterion, the PTAC may request or stakeholders may wish to provide information about potential approaches for evaluation including evaluation study design, comparison groups, key outcome measures, the level of precision the evaluation may reach, and the extent that the impact of each element of the PFPM can be evaluated.

Second, we propose a category of criteria that address care delivery improvements that promote better care. Here we propose criteria to address integration and care coordination, patient choice, and patient safety. To address these criteria, the PTAC may request or stakeholders may wish to provide information about how the payment model would affect access to care for Medicare beneficiaries, including an explanation of how the payment model would not reduce benefits for Medicare beneficiaries, limit coverage for beneficiaries, how the payment model would affect disparities among Medicare beneficiaries by race, ethnicity, gender, disability, and geography, and what measures may be used to measure the provision of necessary care and monitor for any potential stinting of care. The PTAC may also request or stakeholders may wish to provide information about how patient choice is preserved under the model by accommodating individual differences in patient characteristics, conditions, and health-related preferences while furthering population health outcomes.

Third, we propose a category of criteria that address information enhancements that improve the availability of information to guide decision-making. We believe that information enhancements, particularly through use of technology are important to improving Medicare payment policy and delivering better care. Here we propose a criterion for encouraging use of health information technology. In addition, we recommend that stakeholders include information about any information enhancements that encourage transparency concerning cost and quality of care to patients and other stakeholders. To address these criteria, the PTAC may request or stakeholders may wish to provide information about how the payment model could increase transparency, or how the payment model could incorporate certified EHR technology.

In carrying out its review of PFPM proposals, the PTAC shall assess whether the PFPM meets the following criteria for PFPMs sought by the Secretary. The Secretary seeks PFPMs that:

(1) Incentives: Pay for higher-value care.

• Value over volume: Provide incentives to practitioners to deliver high-quality health care.
• Flexibility: Provide the flexibility needed for practitioners to deliver high-quality health care.
• Quality and Cost: Are anticipated to improve health care quality at no additional cost, maintain health care quality while decreasing cost, or both improve health care quality and decrease cost.
• Payment methodology: Pays APM Entities with a payment methodology designed to achieve the goals of the PFPM Criteria. Addresses in detail through this methodology how Medicare, and other payers if applicable, pay APM Entities, how the payment methodology differs from current payment methodologies, and why the Physician-Focused Payment Model cannot be tested under current payment methodologies.
• Scope: Aim to either directly address an issue in payment policy that broadens and expands the CMS APM portfolio or include APM entities whose opportunities to participate in APMs have been limited.
• Ability to be evaluated: Have evaluable goals for quality of care, cost, and any other goals of the Physician-focused Payment Model.

(2) Care delivery improvements: Promote better care coordination, protect patient safety, and encourage patient engagement.

• Integration and Care Coordination: Encourage greater integration and care coordination among practitioners and across settings where multiple practitioners or settings are relevant to delivering care to the population treated under the Physician-Focused Payment Model.
• Patient Choice: Encourage greater attention to the health of the population served while also supporting the unique needs and preferences of individual patients.
• Patient Safety: Aim to maintain or improve standards of patient safety.

(3) Information Enhancements: Improving the availability of information to guide decision-making.

• Health Information Technology: encourage use of health information technology to inform care.

d. Facilitating CMS Consideration of Models Recommended by the PTAC

In order to facilitate and potentially expedite the consideration of models for our testing following PTAC review and recommendation, we suggest “supplemental information elements” stakeholders may include in their PFPM proposals to assist our review. We do not propose to require these elements as PFPM criteria and defer to the PTAC on how it may approach requesting any supplemental information beyond that required to meet the PFPM criteria.

(1) Background on Factors Used To Evaluate Potential Innovation Center Models

Section 1115A of the Act authorizes the Innovation Center to test innovative payment and service delivery models. We have an established process by which it routinely assesses proposals for new models. Many factors are typically used in the selection of models to be tested, and can be viewed on the Innovation Center Web site at https://innovation.cms.gov/Files/x/rfi-Websitepreamble.pdf .

(2) Why and How These Factors Informed the Supplemental Information Elements for PFPM Proposals

These factors address a variety of details in an APM's design. Examining these factors helps us to answer important questions that inform its decision whether to test a payment model. They provide necessary insight about how a potential APM would fit within our current CMS APM portfolio, including information such as the scope of its impact, likelihood of success, how many practitioners and beneficiaries it would impact, whether those potential outcomes merit the required investments and opportunity costs, and whether the impact of the payment model can be measured to determine if it should be expanded. We believe that to the extent stakeholders develop PFPM proposals that address the factors used by us in evaluating payment model designs, they would increase the probability that PTAC recommendations would be positive and might lead us to test the proposed PFPMs.

We considered each factor currently used by us when we developed the suggested supplemental information elements for PFPM submission. We balanced the burden these expectations would place on stakeholders in developing their proposed PFPMs with the value this information would provide to the PTAC in its review of the proposed PFPMs and to us in our decision whether or not to test a proposed PFPM. We acknowledge that the factors used by us may change in the future and we believe that the PFPM criteria we have proposed are sufficiently broad and relevant to our evaluation of the testing of models that they would align with any future changes in our factors. While we believe that the more detail concerning these factors the stakeholder can provide the more it would facilitate our review, we have determined that certain factors are of particular importance.

We also chose not to include certain of these factors, including the size of investment required and waiver authority, in the suggested supplemental information elements because we believe the burden to evaluate how these factors apply to potential APMs should be on us, not stakeholders. For example, we received responses to the RFI both in favor of, and opposed to requiring information about whether, if the proposed PFPM cannot be implemented under existing law, we have the authority to waive any laws or regulations for purposes of testing the payment model. We decided it would be inappropriate to require stakeholders to speculate as to the scope of our waiver authority in their proposals. We and other components of HHS are responsible for interpreting the relevant laws and regulations, and for designing and issuing any potential waivers. We also decided not to include as a supplemental information element the size of investment for proposed PFPMs because we do not believe stakeholders would have the necessary information about our operational costs to include in a PFPM proposal.

(3) Supplemental Information Elements Considered Essential to CMS Consideration of New Models

There are three pieces of information we consider fundamental to evaluating new models. First, the anticipated size and scope of a proposed PFPM is essential. For example, any proposed PFPM should describe the estimated number of Medicare beneficiaries that would be affected by the model, the number and scope of eligible clinicians expected to participate, including eligible clinician specialty(s), the potential geographic location(s) included in the model, the defined period of performance or clinical episode(s) of care in the model, and the number and quality of services that would be affected by the model. A definition of the target population and any criteria for including or excluding patients from the model would also be useful in addressing the scope of the PFPM. We believe this information is vital to evaluating a proposed PFPM. Second, we also consider an estimate of the burden in terms of morbidity and mortality on a population to be relevant in describing the scope of physician services addressed by the model. For example, stakeholders could provide estimates of morbidity and mortality from peer-reviewed publications and analyses of health care data such as Medicare or Medicaid data. Third, we believe an explanation of how a proposed model would be attractive to participate in and feasible to implement for potential APM Entities from a financial perspective is necessary for us to evaluate a proposed model.

To summarize, the following specific supplemental information elements are considered essential:

• A description of the anticipated size and scope of the model in terms of eligible clinicians, beneficiaries, and services.
• A description of the burden of disease, illness or disability on the target patient population.
• An assessment of the financial opportunity for APM Entities, including a business case for how their participation in the model could be more beneficial to them than participation in traditional fee-for-service Medicare.

In addition, we recommend that proposed PFPMs submitted to the PTAC include information about whether the stakeholder or individual submitting the proposal believes it would meet the criteria to be an Advanced APM, discussed in section II.F.4. of this proposed rule. This information would allow us to evaluate whether the proposed PFPM would provide eligible clinicians with an opportunity to become QPs for purposes of the incentives for participation in Advanced APMs. We are interested in receiving proposed PFPMs from stakeholders that would be Advanced APMs and we received comments on the RFI stating that stakeholders would like this opportunity as well. As discussed in section II.F.10.b. of this proposed rule, we do not believe that it is necessary to limit stakeholders' proposed PFPMs to only those that would be Advanced APMs, but believe that it is useful for proposed PFPMs to state whether, if tested by us, they would be Advanced APMs.

e. MIPS and APMs RFI Comments on PFPM Criteria

We received multiple responses to the MIPS and APMs RFI recommending that the Secretary include specialty-specific criteria to be used by the PTAC. We appreciate the interest from multiple specialties and encourage them to submit proposed PFPMs for review by the PTAC, but do not believe that we should limit proposed PFPMs by adding specialty-specific criteria.

We received multiple comments suggesting prioritization of certain patient groups, physician specialties, diseases, and other issues. We believe that the aim of section 1868(c) of the Act to promote development of PFPMs is best satisfied by not prioritizing certain specialties or issues over others in the PFPM criteria. However, the PTAC may decide to prioritize specific patient groups, specialties, or issues in its comments and recommendations.

We received several responses to the MIPS and APMs RFI recommending that types of physicians and practitioners that have had the opportunity to participate in previous APMs should not be excluded from future proposals for PFPMs because current or previous APMs are not exhaustive of all possible APMs for any given specialty or issue. We agree with this recommendation, so long as the proposed PFPM instead aims to solve an issue in payment policy that broadens and expands the CMS APM portfolio at the time it is tested as stated in section II.F.10.c. of this proposed rule. We believe this best serves our goal of expanding and diversifying our portfolio of APMs. We believe that concurrently implementing multiple PFPMs that attempt to solve the same clinical or payment issue may not be the most efficient use of limited resources, and may complicate the evaluations of some or all of the relevant models. However, we would consider a proposed PFPM that focuses on an issue addressed in a model that we are no longer testing, even if that prior model was unsuccessful.

We also received responses to the MIPS and APMs RFI recommending that we should consider proposals that modify or extend the testing of existing models. We do not believe that the PTAC is the proper forum for considering modifications or extensions of current models. We also note that our legal authority to modify or extend existing models is contingent on other criteria that are unrelated to the criteria for proposed PFPMs. Stakeholders who wish to discuss changes to models that we are currently testing may discuss them with us directly, outside of the PTAC review process.

We received many comments suggesting payment for high-value services that we do not currently (or separately) reimburse as examples of potential PFPMs. These types of changes are an important part of moving toward value-based delivery system reform, but adding payment for specific services without any other change does not constitute a sufficient departure from current payment methodologies to meet our proposed PFPM criteria or to be considered an APM, and could be better achieved outside of the PTAC process. We do however welcome these suggestions within the context of broader model proposals.

We received responses to the MIPS and APMs RFI recommending that in addition to criteria about how the proposed PFPM would either fit in to or replace the existing Medicare payment system, there should also be criteria to identify specific barriers to care improvement that exist in the current payment system. We believe that information about how the proposed PFPM changes or fits into existing payment systems is essential to understanding how the proposed PFPM operates. We believe information about existing barriers to improving care and reducing costs and how the proposed PFPM addresses those barriers is also important. Therefore, we encourage stakeholders to include this information in their proposals although we do not propose to require it.

We received many responses to the MIPS and APMs RFI recommending that the PFPM criteria include specific quality measures and guidelines, such as those set by the Core Quality Measures Collaborative. We do not believe it would be appropriate to limit proposed PFPMs to include specific quality measures. We encourage stakeholders to propose quality measures that are tailored to their particular proposed PFPM. We also received responses to the MIPS and APMs RFI recommending we should not include criteria requiring information on the impact that the proposed PFPM would have on quality of care. We understand that the full scope of the potential impact a proposed PFPM may have on quality of care and cost reduction might not be known at the time of submission. However, we believe proposed PFPMs should provide realistic assessments and estimates of the impacts, as well as information to justify these estimates. Commentators also voiced opinions about the utilization of Clinical Data Registries managed by specialty societies or other groups. We believe that this information, if applicable, should be included in the PFPM proposal as an aspect of CEHRT use.

Finally, we received many responses to the MIPS and APMs RFI offering proposed PFPMs that we should implement. We appreciate the interest in PFPMs and encourage these commenters to submit their proposed PFPMs to the PTAC.

III. Collection of Information Requirements

Under the Paperwork Reduction Act of 1995, we are required to provide 60-day notice in the Federal Register and solicit public comment before a collection of information requirement is submitted to the Office of Management and Budget (OMB) for review and approval. In order to fairly evaluate whether an information collection should be approved by OMB, section 3506(c)(2)(A) of the Paperwork Reduction Act of 1995 requires that we solicit comment on the following issues:

• The need for the information collection and its usefulness in carrying out the proper functions of our agency.
• The accuracy of our estimate of the information collection burden.
• The quality, utility, and clarity of the information to be collected.
• Recommendations to minimize the information collection burden on the affected public, including automated collection techniques.

We are soliciting public comment on each of these issues for the following sections of this document that contain information collection requirements (ICRs).

A. Wage Estimates

To derive wage estimates, we used data from the U.S. Bureau of Labor Statistics' May 2014 National Occupational Employment and Wage Estimates and the December 2015 Employer Costs for Employee Compensation. In this regard, Table 46 presents the mean hourly wage, the cost of fringe benefits and overhead, and the adjusted hourly wages for Billing and Posting Clerks, Computer Systems Analysts, and Physicians. We are adjusting our employee hourly wage estimates by a factor of 100 percent to reflect current HHS department-wide guidance on estimating the cost of time spent by employees of regulated entities. These are necessarily rough adjustments, both because fringe benefits and overhead costs vary significantly from employer to employer, and because methods of estimating these costs vary widely from study to study. Nonetheless, there is no practical alternative and we believe that these are reasonable estimation methods. In addition, in order to calculate the costs to beneficiaries for their time, we have used Bureau of Labor Statistics (BLS) estimates for Civilian, all occupations. We have not adjusted these costs for fringe benefits and overhead because only the direct wage costs represent the “opportunity cost” to beneficiaries themselves for time spent in health care settings.

B. A Framework for Understanding the Burden of MIPS Data Submission

Because the entities permitted to submit MIPS data on behalf of eligible clinicians will vary based on APM participation and the type of data, Table 47 presents a framework for understanding the entities facing the burden of MIPS data submission. As shown in the first row of Table 47, eligible clinicians that are not in APMs will submit data either as individuals or groups to the quality, advancing care information, and CPIA performance categories.

For APMs, the entities submitting data on behalf of model participants will vary across categories of data and APM Model. When APM Entities submit quality data to fulfill the requirements of their APMs, the burden will be ascribed to their APMs, and will not contribute to the MIPS data submission burden. Many APM participants will be scored on advancing care information and CPIA performance categories, and the submitting entity for those categories differs between the Shared Savings Program and other APMs. For the Shared Savings Program, billing TINs (or groups) will submit advancing care information and CPIA performance category data on behalf of model participants. In other APMs, eligible clinicians will submit data as individuals to the advancing care information and CPIA performance categories. For Advanced APMs, Partial Qualifying APM Participant (Partial QP) elections (which will be discussed in more detail in Section I below) will be submitted by the Advanced APM Entity on behalf of all its participating eligible clinicians.

C. ICRs Regarding Quality Performance Reporting Category (§ 414.1330 and § 414.1335 and Section II.E.5.b of This Preamble) and Previously Approved Under PQRS

This section discusses the information collection requirements for the eligible clinicians who are not APM participants because burden for APM Entities' submission of quality data to fulfill the requirements of their APMs will not be ascribed to MIPS. Based on historical data in the 2014 PQRS Experience Report, we estimate that up to 703,467 MIPS eligible clinicians will submit quality performance category data including those participating as groups. Because of the exclusion of QPs from our quality performance data burden estimates, our estimates of the number of eligible clinicians submitting MIPS quality data is lower than the estimate of 822,810 professionals that submitted quality data to the 2014 PQRS. We assume that clinicians not in APMs that reported quality data to PQRS in 2014 will continue to report quality data to MIPS. We assume that some of those clinicians will be submitting voluntarily because they are not required (but are allowed) to report quality data to MIPS because they are in specialties not required to participate in MIPS.

We assume that the number of MIPS eligible clinicians who will submit through claims mechanisms (299,169), Qualified Registry or QCDR-mechanisms (214,590), certified EHR technology mechanisms (77,241), and as groups through CMS Web Interface (112,467) will be the same as the numbers submitting data through those mechanisms under the 2014 PQRS. We also assume that the number of groups that will submit quality performance category data through the CMS Web Interface will be the same as the number submitting PQRS data through the GRPO Web Interface in 2014 (300 groups submitting on behalf of 112,467 MIPS eligible clinicians). Historically, the PQRS has never experienced 100 percent participation; the participation rate for 2014 was 63 percent.

For MIPS eligible clinicians or groups, the burden associated with the requirements of the MIPS quality performance category is the time and effort associated with MIPS eligible clinicians identifying applicable quality measures for which they can report the necessary information, collecting the necessary information, and reporting the information needed to submit the MIPS eligible clinician's measures. We believe it is difficult to quantify the burden accurately because MIPS eligible clinicians and groups may have different processes for integrating quality reporting into their practices' work flows. Moreover, the time needed for an MIPS eligible clinician to review the quality measures and other information, select measures applicable to his or her patients and the services he or she furnishes to them, and incorporate the use of quality data codes into the office work flows is expected to vary, along with the number of measures that are potentially applicable to a given professional's practice.

For MIPS eligible clinicians and groups, we estimate a total of 6 hours as the amount of time needed for an MIPS eligible clinician's billing clerk to review the quality measures list, review the various submission options, select the most appropriate submission option, identify the applicable measures or specialty measure sets for which they can report the necessary information, review the measure specifications for the selected measures or measures groups, and incorporate submission of the selected measures or specialty measure sets into the office work flows. The measures list contains the measure title and brief summary information for the MIPS eligible clinician's billing clerk to review. The 6 hour estimate for the billing clerk is comprised of reviewing the performance criteria (up to 2 hours) and reviewing measure specifications (up to 4 hours). Assuming the MIPS eligible clinician has received no training from his/her specialty society, we estimate it will take an MIPS eligible clinician's billing clerk up to 2 hours to review this list, review the submission method, and select a submission method and measures on which to report. If an MIPS eligible clinician has received training, then we believe this would take less time. We believe 4 hours is a reasonable estimate for an MIPS eligible clinician's billing clerk to review the measure specifications of measures they select to report and to develop a mechanism for incorporating submission of the selected measures or into the office work flows. Further, we estimate that it will take a physician up to 1 hour to review MIPS quality performance category measure specifications for each MIPS eligible clinician. Therefore, we believe that the start-up cost for a MIPS eligible clinician's billing clerk to report measures data may be calculated as: 6 hours × $34.20/hour = $205.20, and the start-up cost for a physician to review quality performance category measure specifications to be calculated as 1 hour × $182.46/hour = $182.46. These start-up costs pertain to the specific quality submission methods below, and hence appear in the burden estimate table.

We believe the burden associated with actually submitting the quality measures will vary depending on the submission method selected by the MIPS eligible clinician. As such, we break down the burden estimates by MIPS eligible clinicians and groups according to the submission method used. The revised quality performance requirements and burden estimates will be submitted along with all other ICRs listed below under a new OMB control number (0938-NEW).

1. Burden for Quality Performance Category Reporting by MIPS Eligible Clinicians: Claims-Based Submission

We anticipate the claims submission process for MIPS will be operationally similar as it was under the PQRS. MIPS eligible clinicians must gather the required information, select the appropriate quality data codes (QDCs), and include the appropriate QDCs on the claims they submit for payment. MIPS eligible clinicians will collect QDCs as additional (optional) line items on the CMS-1500 claim form or the electronic equivalent HIPAA transaction 837-P, approved by OMB under control number 0938-0999. This rule does not revise either of those forms. We note that the claims submission option is only available to individual MIPS eligible clinicians and is not available for groups.

The total estimated burden will vary along with the volume of claims on which the quality data is reported. Based on our experience with the PQRS, we estimate that the burden for submission of quality data will range from 7.22 hours to 17.8 hours per MIPS eligible clinician. The wide range of estimates for the time required for a MIPS eligible clinician to submit quality measures via claims reflects the wide variation in complexity of submission across different clinician quality measures. As shown in Table 48 we also estimate that the cost of quality data submission will range from $18.47 (.22 hours × $83.96) to $906.77 (10.8 hours × $83.96). The total estimated annual cost per MIPS eligible clinician ranges from the minimum burden estimate of $406.13 to a maximum burden estimate of $1,294.43. The burden will involve becoming familiar with MIPS data submission requirements. Therefore, we believe that the start-up cost for a MIPS eligible clinician's billing clerk to report measures data may be calculated as: 6 hours × $34.20/hour = $205.20, and the start-up cost for a physician to review quality performance category measure specifications to be calculated as 1 hour × $182.46/hour = $182.46. Therefore, total annual burden cost is estimated to range from a minimum burden estimate of $121,501,865 (299,169 × $406.13) to a maximum burden estimate of $387,252,730 (299,169 × $1294.43).

Based on the assumptions discussed above, Table 48 summarizes the range of total annual burden associated with MIPS eligible clinicians using the claims submission mechanism.

2. Burden for Quality Performance Category Reporting by MIPS Eligible Clinicians and Groups Using Qualified Registry and QCDR Submissions

For qualified registry and QCDR submissions, we estimate an additional time burden for MIPS eligible clinicians and groups to become familiar with MIPS submission requirements and, in some cases, new specialty measure sets. Therefore, we believe that the start-up cost for a MIPS eligible clinician's billing clerk to report measures data may be calculated as: 6 hours × $34.20/hour = $205.20, and the start-up cost for a physician to review quality performance category measure specifications to be calculated as 1 hour × $182.46/hour = $182.46. These start-up costs pertain to the specific quality submission methods below, and hence appear in the burden estimate table.

Little, if any, additional data will need to be reported to the qualified registry or QCDR solely for purposes of participation in MIPS. However, MIPS eligible clinicians and groups will need to authorize or instruct the qualified registry or QCDR to submit quality measures results and numerator and denominator data on quality measures to CMS on their behalf. We estimate that the time and effort associated with this will be approximately 5 minutes (0.083 hours) per MIPs eligible clinician for a total burden cost of $6.97, at a computer systems analyst's labor rate. Hence, we estimated 10.083 burden hours per MIPS eligible clinician, with annual total burden hours of 2,163,711 (10.083 burden hours × 214,590 MIPS eligible clinicians). The total estimated annual cost per MIPS eligible clinician is estimated to be approximately $646.51. Therefore, total annual burden cost is estimated to be $138,734,298 (214,590 × $646.51). Based on these burden requirements and the number of eligible clinicians historically using the Qualified Registry and QCDR submissions, we have calculated a burden estimate for quality performance category reporting for these submissions:

3. Burden for Quality Performance Category Reporting by Eligible Clinician and Groups: EHR Submission

Based on our experience with the PQRS, we estimate that the time needed to perform all the steps necessary for MIPS eligible clinicians to report quality performance measures includes the time to prepare for participating in quality performance category submissions for MIPS (calculated at 6 hours plus 1 hour of physician time for reviewing specifications), and an additional 3 hours for data submission through an EHR.

For EHR submission, the MIPS eligible clinician or group must review the quality measures on which we will be accepting MIPS data extracted from EHRs, select the appropriate quality measures, extract the necessary clinical data from his or her EHR, and submit the necessary data to the CMS-designated clinical data warehouse. To submit data to CMS directly from their EHRs, MIPS eligible clinicians must have access to a CMS-specified identity management system which we believe takes less than 1 hour to obtain. Once an MIPS eligible clinician has an account for this CMS-specified identity management system, he or she will need to extract the necessary clinical data from his or her EHR, and submit the necessary data to the CMS-designated clinical data warehouse. We estimate that obtaining a CMS-specified identity management system will require 1 hour per MIPS eligible clinician cost of $83.96 (1 × $83.96), and that submitting a test data file to CMS will also require 1 hour for a per MIPS eligible clinician for a cost of $83.96. With respect to submitting the actual data file for the respective reporting period, we believe that this will take an MIPS eligible clinician or group no more than 2 hours for a per MIPS eligible clinician cost of submission of $167.92 (2 × $83.96). The burden will involve becoming familiar with MIPS submission. In addition, we believe that the start-up cost for a MIPS eligible clinician's billing clerk to report measures data may be calculated as: 6 hours × $34.20/hour = $205.20, and the start-up cost for a physician to review quality performance category measure specifications to be calculated as 1 hour × $182.46/hour = $182.46. Hence, we estimated 11 burden hours per MIPS eligible clinician, with annual total burden hours of 849,651 (11 burden hours × 77,241 MIPS eligible clinicians). The total estimated annual cost per MIPS eligible clinician is estimated to be $723.50. Therefore, total annual burden cost is estimated to be $55,883,864 (77,241 × $723.50).

Based on these burden requirements and the number of eligible clinicians historically using the EHR submission mechanism, we have calculated a burden estimate for quality performance category reporting for this submission mechanism:

4. Burden for Quality Performance Category Reporting for Groups Using the CMS Web Interface

We estimate that 112,467 MIPS eligible clinicians submitting as 300 groups will participate in MIPS using the CMS Web Interface in the 2017 Performance Period. Groups interested in participating in the MIPS using the CMS Web Interface must complete a registration process. However, since a group using the CMS Web Interface would not need to determine which measures to report under MIPS, we believe that the registration process is handled by the group's administrative staff. We estimate that the registration process for groups under MIPS involves approximately 1 hour per group. We assume that the group staff involved in the group registration process has an average labor cost of $34.20 per hour. Therefore, assuming the total burden hours per group associated with the group registration process is 1 hour, we estimate the total cost to a group associated with the group registration process to be approximately $34.20 ($34.20 per hour × 1 hour per group).

The burden associated with the group submission requirements under the CMS Web Interface is the time and effort associated with the group submitting the quality measures data. For physician groups, this would be the time associated with the physician group completing the CMS Web Interface. We estimate that, on average, it will take each group 79 hours to submit quality measures data via the CMS Web Interface at a cost of $83.96 per hour, for a total cost of $6,6632.84 (79 × $83.96). We also estimate that a physician for each group will need to spend 1 hour per year to review quality performance measure specifications, for a total cost of $182.46. As mentioned above, we estimate it will take 1 hour for a group to register to submit through the CMS Web Interface, for a total of cost of $34.20 (1 × $34.20). Therefore, the total estimated annual cost per group is estimated to be approximately $6,632.84. The total annual burden hours are estimated to be 24,300 (300 eligible groups × 81 annual hours), and the total annual burden cost is estimated to be $2,052,850 (300 × $6,849.50).

Based on the assumptions discussed above we have calculated the following burden estimate for groups submitting to MIPS with the CMS Web Interface.

D. ICRs Regarding Burden for Third Party Reporting and Data Validation (§ 414.1400 and § 414.1390)

1. Burden for Qualified Registry and QCDR Self-Nomination

For CY 2015, 98 qualified registries and 49 QCDRs were qualified to report quality measures data to CMS for purposes of the PQRS. Under MIPS we believe that the number of QCDRs and qualified registries will increase because (1) many MIPS eligible clinicians will be able to use the qualified registry and QCDR for all MIPS submission (not just for quality submission) and (2) QCDRs will be able to provide innovative measures that address practice needs. Qualified registries or QCDRs interested in submitting quality measures results and numerator and denominator data on quality measures to CMS on their participants' behalf will need to complete a self-nomination process in order to be considered qualified to submit on behalf of MIPS eligible clinicians or groups, unless the qualified registry or QCDR was qualified to submit on behalf of MIPS eligible clinicians or groups for prior program years and did so successfully. We estimate that the self-nomination process for qualifying additional qualified registries or QCDRs to submit on behalf of MIPS eligible clinicians or groups for MIPS will involve approximately 1 hour per qualified registry or QCDR to complete the online self-nomination process.

Please note that the self-nomination statement is an online form that entities will use to provide information on their business. The self-nomination statement will be available at https://jira.oncprojectracking.org/login.jsp .

In addition to completing a self-nomination statement, qualified registries and QCDRs will need to perform various other functions, such as meet with CMS officials when additional information is needed. In addition, QCDRs must benchmark and calculate their measure results. We note, however, that many of these capabilities may already be performed by QCDRs for purposes other than to submit data to CMS for MIPS. The time it takes to perform these functions may vary depending on the sophistication of the entity, but we estimate that a qualified registry or QCDR will spend an additional 9 hours performing various other functions related to being a MIPS qualified registry or QCDR.

We estimate that the staff involved in the qualified registry or QCDR self-nomination process will mainly be Computer Systems Analysts or the equivalent, at an average labor cost of $83.96/hour. Therefore, assuming the total burden hours per qualified registry or QCDR associated with the self-nomination process is 10 hours, the annual burden hours is 1,500 (150 QCDRs × 10 hours). We estimate that the total cost to a qualified registry or QCDR associated with the self-nomination process will be approximately $839.60 ($83.96 per hour × 10 hours per qualified registry). We also estimate that 150 new qualified registries or QCDRs will go through the self-nomination process leading to a total burden of $125,940 ($839.60 × 150).

The burden associated with the qualified registry and QCDR submission requirements in MIPS will be the time and effort associated with the qualified registry calculating quality measures results from the data submitted to the qualified registry or QCDR by its participants and submitting the quality measures results, the numerator and denominator data on quality measures, and the advancing care information performance category and CPIA data to CMS on behalf of their participants. We expect that the time needed for a qualified registry or QCDR to review the quality measures and other information, calculate the measures results, and submit the measures results and numerator and denominator data on the quality measures and the advancing care information performance category and CPIA data on their participants' behalf will vary along with the number of MIPS eligible clinicians submitting data to the qualified registry or QCDR and the number of applicable measures. However, we believe that qualified registries and QCDRs already perform many of these activities for their participants. Therefore, there may not necessarily be an additional burden on a particular qualified registry or QCDR associated with calculating the measure results and submitting the measures results and numerator and denominator data on the quality measures to CMS on behalf of their MIPS participants. Whether there is any additional burden to the qualified registry or QCDR as a result of the qualified registry's or QCDR's participation in MIPS will depend on the number of measures that the qualified registry or QCDR intends to report to CMS and how similar the qualified registry's measures are to CMS' MIPS quality measures.

Based on the assumptions previously discussed, we provide an estimate of total annual burden hours and total annual cost burden associated with a qualified registry or QCDR self-nominating to be considered “qualified” for the purpose of submitting quality measures results and numerator and denominator data on MIPS eligible clinicians.

2. Burden for MIPS Data Validation

Under MIPS, a CMS contractor will conduct a data validation survey in order to identify and address problems with data handling, data accuracy, and incorrect payments for the MIPS Program. Because the data that will be submitted by, or on behalf of, MIPS eligible clinicians to the MIPS Program will be used to calculate payment adjustments, it is critical that this data be accurate. Additionally, the data will be used to generate Feedback Reports for MIPS eligible clinicians and groups and, in some cases, is posted publicly on the CMS Web site, further supporting the need for accurate and complete data. The CMS data validation contractor will conduct surveys of Groups, Registries, Qualified Clinical Data Registries (QCDRs), EHR Vendors, and MIPS eligible clinicians in support of evaluating the data submitted for MIPS. The MIPS Data Validation survey will be similar to the PQRS Data Validation Survey. The PQRS Data Validation Survey uses a series of approximately thirty questions, arranged by category, to gather information about data handling practices, training, and quality assurance, as well as the challenges that stakeholders faced in participating in the PQRS program. Under MIPS, the survey's topics will be expanded beyond validation of quality measures to include CPIA and potentially advancing care information performance category data.

The MIPS Data Validation Survey for Performance Year 2017 will be conducted in late 2018 for data reported in early 2018. Because the MIPS verification process is still under development, the precise sample size for respondents has not yet been determined. We anticipate that at most 500 entities would be contacted for MIPS data verification for Performance Year 2017. Based on the most recent year of the PQRS data validation survey, we will assume that the response rate will be 86 percent. Hence, we estimated the total number of respondents for Performance Year 1 will be 430 (500 entities contacted × 86 percent response rate).

We estimate the total annual burden for the ongoing MIPS data validation will be up to 750 hours each performance year (500 responses × 1.5 hours), and the data validation will be conducted at a clerk's labor rate of $34.20 per hour for a total burden cost of $25,650 ($34.20 × 1.5).

E. Burden for Reporting Quality Performance Category via CAHPS for MIPS Survey

Under MIPS, groups may elect to report on the CAHPS for MIPS Survey by contracting for survey administration with a CMS approved vendor. At this point, we do not believe that the groups that elect to report on CAHPS for MIPS will experience additional burden because CAHPS will cover one of their six Quality performance category measures. Beneficiaries will experience burden under the CAHPS survey; and because the survey will be similar to the CAHPS for PQRS survey, we are assuming that the burden per beneficiary will be the same.

The usual practice in estimating the burden on public respondents to surveys such as CAHPS is to assume that respondent time is valued, on average, at civilian wage rates. As previously explained, the BLS data show the average hourly wage for civilians in all occupations to be $23.06. Although most Medicare beneficiaries are retired, we believe that their time value is unlikely to depart significantly from prior earnings expense, and have used the average hourly wage to compute the dollar cost estimate for these burden hours.

Under the first performance year of MIPS, we assume the number of groups that elect to report on the CAHPS for MIPS survey will be the same as 2014, when the CAHPS for PQRS survey was used. Table 54 shows the estimated annualized burden for beneficiaries to participate in the CAHPS for MIPS survey. We assume that all 300 groups submitting via the CMS Web Interface will elect to use the CAHPS for MIPS Survey. Based on historical information on the numbers of CAHPS for PQRS respondents, we assume that an average of 287 beneficiaries will respond per group. The CAHPS Survey for MIPS will be administered to approximately 86,100 beneficiaries per year (300 groups × an average of 287 beneficiaries per group responding). The survey contains 83 items and is estimated to require an average administration time of 18.4 minutes in English (at a pace of 4.5 items per minute) and 22 minutes in Spanish (assuming 20 percent more words in the Spanish translation), for an average response time of 20.24 minutes or 0.337 hours. These burden and pace estimates are based on CMS's experience with surveys of similar length that were fielded with Medicare beneficiaries. As indicated below, the annual total burden hours are estimated to be 29,106 hours (86,100 respondents × .337 burden hours per respondent to report).

F. ICRs Regarding Burden Estimate for Advancing Care Information Performance Category (§ 414.1375 and Section II.E.5.g. of This Preamble)

Advancing care information performance category data will not be submitted separately by MIPS eligible clinicians in most cases as was required under the Medicare EHR Incentive Program. MIPS eligible clinicians and clinician groups will submit this data using the same submission mechanism, or a similar submission mechanism they have selected for the other MIPS performance categories. For the purpose of submission advancing care information performance category objectives and measures under the MIPS, we have proposed in section II.E.1.f. of this proposed rule to allow for MIPS eligible clinicians to submit advancing care information performance category data through qualified registry, EHR, QCDR, and CMS Web Interface submission methods. Also, we have streamlined the submission requirements for advancing care information as part of the MIPS program. Compared to the reporting requirements in the 2015 Medicare EHR Incentive Program Final Rule, two objectives and their associated measures (Clinical Decision Support and Computerized Provider Order Entry) will no longer be required for submission purposes. We have also worked to align the advancing care information performance category with other MIPS performance categories, such as submitting eCQMs to the quality category, which will streamline submission requirements and reduce MIPS eligible clinician confusion. Hence, a MIPS eligible clinician's estimated burden for the advancing care information performance category is lower than the estimated 7 hours per MIPS eligible clinician in the Medicare EHR Incentive Program—Stage 3 PRA (OMB control number 0938-1278) currently under review at OMB. We are requesting that effective January 1, 2017, the MIPS Collection of Information Requirements replace those for eligible clinicians in the Medicare EHR Incentive Program Stage 3 PRA.

As noted above in Section B, a variety of entities will report advancing care information performance category data on behalf of clinicians. Based on historical data and 2015 Medicare EHR Incentive Program attestation, we estimate that approximately 436,500 clinicians not participating in APMs would submit advancing care information performance category data to MIPS.

Because Performance Year 2017 will be the first year for MIPS eligible clinicians to report the advancing care information performance category data as groups, there is considerable uncertainty about what number of MIPS eligible clinicians will report as part of a groups. For the purposes of our burden estimate, we conservatively estimate that all the clinicians that reported as individuals under the 2015 Medicare EHR Incentive Program will continue to report as individuals in MIPS Year 1, but may transition to group submission in future years. Because some participants in APM Entities will be required to report advancing care information performance category data to fulfill the requirements of submitting to MIPS, we have included them in our burden estimate for the advancing care information performance category. Further we anticipate that the 434 Shared Savings Program ACOs will submit data at the ACO participant billing TIN level, for a total of 25,925 submitting entities, and approximately 55,000 other APM participants will report as individual MIPS eligible clinicians. Hence, as shown in Table 56, we estimate that up to approximately 517,425 entities will be submitting data under the advancing care information performance category (436,100 MIPS eligible clinicians + 25,925 billing TINS within the Shared Savings Program ACOs + 55,000 APM participants). The total burden hours for a MIPS eligible clinician or group to report on the objectives and measures specified for the advancing care information performance category will be 4 hours. The total estimated burden hours are 1,552,275 (517,425 × 4). At a physician's hourly rate, the total burden cost is $283,228,097 (1,552,275,300 × $182.46).

G. ICRs Regarding Burden for Clinical Practice Improvement Activities Submission (§ 414.1355, § 414.1365, and Section II.E.5.d of This Preamble)

Requirements for submitting clinical practice improvement activities are new, and we do not have historical data which is directly relevant. As noted in section III.B, a variety of entities will report advancing care information performance category data on behalf of eligible clinicians. For eligible clinicians who are not part of APMs, we assume that the number of eligible clinicians submitting as part of a group will be approximately the same as the number of eligible clinicians submitting PQRS data through the GPRO Web Interface in 2014. We estimate that that there could be as many as 595,100 MIPS eligible clinicians submitting CPIA category data as individuals, which is equal to the number of eligible clinicians using the claims, QCDR or qualified registry, or EHR submission mechanisms under the 2014 PQRS. We estimate that approximately 112,500 MIPS eligible clinicians comprising 300 groups may report at the group level.

Because some APM Entities and participants will be required to report CPIA data to fulfill the requirements of submitting to MIPS, we have included them in our burden estimate for the CPIA submitting. As with the advancing care information performance category, participants in Shared Savings Program ACOs will report at the ACO participant billing TIN level, and other APM participants will report as individual MIPS eligible clinicians. We anticipate MIPS eligible clinicians, groups, APM billing TINs, will submit CPIA data using the same mechanism, or a similar mechanism as they select for submitting quality data. In addition to collecting necessary supporting documentation, each MIPS eligible clinician, group, ACO participant billing TIN, or APM participant will provide a yes/no attestation submitted during the data submission period for successfully completed CPIAs. We estimate that up to approximately 676,325 entities will be submitting data for CPIA. We estimate it will take no longer than 3 hours per entity to submit data for the CPIA category. The total estimated burden is 2,028,975 (676,325 entities × 3 hours each). At a physician's hourly rate, the total estimated burden cost is $370,206,779 (2,028,975 × $182.46).

H. ICRs Regarding Burden for Resource Use (§ 414.1350 and Section II.E.5.c of This Preamble)

The resource use performance category relies on administrative claims data. For claims-based submitting, the Medicare Parts A and B claims submission process is used to collect data on resource measures from MIPS eligible clinicians. MIPS eligible clinicians are not asked to provide any documentation by CD or hardcopy. Therefore, we do not anticipate any new or additional submitting for MIPS eligible clinicians as a result of this performance category within MIPS.

I. ICR Regarding Partial QP Elections for Advanced APMs

Section II.E.5.h. of this preamble discusses the MIPS-related submission requirements for participants in the Shared Savings Program and certain APMs. APM Entities participating in Advanced APMs will face an additional submission requirement under MIPS related to Partial Qualifying APM Participant (QP) elections. A representative from each APM Entity will log into the MIPS portal to indicate whether eligible clinicians would wish to participate in MIPS if the eligible clinicians participating in the APM Entity are later deemed to be Partial QPs. We estimate it will take each APM Entity representative 15 minutes to make this election, and an additional 15 minutes to register for the MIPS Portal. We estimate that 543 APM Entities will make this election on the MIPS Portal, for a total burden estimate of 272 hours (543 × .5). At a computer systems analyst's hourly labor cost, the total burden cost is estimated to be $22,795 (272 × $83.96).

J. Summary of Annual Burden Estimates

The total gross burden estimate includes the total burden of recordkeeping and data submission under MIPS. Table 60 provides an estimate of the total annual burden of MIPS of 12,493,654 hours and a total annual burden cost of $1,327,177,683. Some of the information collection burden under MIPS does not represent an additional burden to the public, but replaces information collection burden that existed under two of its predecessor programs, the PQRS and the Medicare EHR Incentive Program. The estimated total existing burden approved for information collections related to PQRS and the Medicare EHR Incentive Program (for EPs) was 9,969,514 hours for a total annual burden cost of $1,199,257,029. The net burden estimate reflects only the incremental burden associated with this rule, and excludes the burden of existing recordkeeping and data submission under the PQRS, the Medicare EHR Incentive Program, CAHPS for PQRS, and PQRS Data Validation. Mindful of the combined data submission burden of MIPS, we have sought to avoid duplication of data submission efforts and simplified data submission structures within the unified program.

K. Submission of PRA-Related Comments

To obtain copies of the supporting statement and any related forms for the proposed collections discussed above, please visit CMS's Web site at www.cms.hhs.gov/PaperworkReductionActof1995 , or call the Reports Clearance Office at 410-786-1326.

We invite public comments on these potential information collection requirements. If you wish to comment, please submit your comments electronically as specified in the ADDRESSES section of this proposed rule and identify the rule (CMS-5517-P), the ICR's CFR citation, CMS ID number, and OMB control number.

ICR-related comments are due July 8, 2016.

IV. Response to Comments

Because of the large number of public comments we normally receive on Federal Register documents, we are not able to acknowledge or respond to them individually. We will consider all comments we receive by the date and time specified in the DATES section of this preamble, and, when we proceed with a subsequent document, we will respond to the comments in the preamble to that document.

V. Regulatory Impact Analysis

A. Statement of Need

This proposed rule is necessary to make payment and policy changes under the Medicare PFS and to make statutorily-required changes under the Medicare Access and CHIP Reauthorization Act of 2015 (MACRA). The MACRA's enactment consolidated certain aspects of physician quality reporting and performance programs into the new Merit-Based Incentive Payment System, including using certified EHR technology (Section 1848(o) of the Act), the PQRS (Section 1848(k) and (m) of the Act), and the value-based payment modifier (Section 1848(p) of the Act). These programs have been developed and most recently implemented by CMS as the Medicare EHR Incentive Program (80 FR 62761), the PQRS (80 FR 71135), and the VM (80 FR 71273). The MACRA's enactment altered the Medicare EHR Incentive Program such that the existing Medicare payment adjustment for EPs under section 1848(a)(7)(A) of the Act will end in CY 2018. Similarly, MACRA ends the separate PQRS Program in CY 2018 and provides for the inclusion of various aspects of PQRS in MIPS, and sunsets the VM program, ending it in CY 2018 and establishing certain aspects of the VM as a component of MIPS in CY 2019. Finally, the MACRA introduces incentive payment to eligible clinicians who become Qualifying APM Participants (QPs) through participation in Advanced APMs.

This consolidated program for physicians and other eligible clinicians represents a new approach to the delivery of health care in this care setting aimed at reducing burden on Medicare-enrolled eligible clinicians, improving population health, lowering growth in overall health care costs, and providing clear incentives for the provision of the best quality care for Medicare beneficiaries. MIPS provides payment adjustments for eligible clinicians for providing value-driven health care services to their patients, and APMs offer a variety of opportunities that substantially alter the methods of payment for health care and enable clinicians to make fundamental changes to their day-to-day operations to improve the quality and reduce the cost of health care.

B. Overall Impact

We have examined the impact of this rule as required by Executive Order 12866 on Regulatory Planning and Review (September 30, 1993), Executive Order 13563 on Improving Regulation and Regulatory Review (February 2, 2013), the Regulatory Flexibility Act (RFA) (September 19, 1980, Pub. L. 96-354), section 1102(b) of the Act, section 202 of the Unfunded Mandates Reform Act of 1995 (March 22, 1995; Pub. L. 14-04), Executive Order 13132 on Federalism (August 4, 1999) and the Congressional Review Act (5 U.S.C. 804(2)).

Executive Orders 12866 and 13563 direct agencies to assess all costs and benefits of available regulatory alternatives and, if regulation is necessary, to select regulatory approaches that maximize net benefits (including potential economic, environmental, public health and safety effects, distributive impacts, and equity). A regulatory impact analysis (RIA) must be prepared for major rules with economically significant effects ($100 million or more in any 1 year). We estimate, as discussed below in this section, that the PFS provisions included in this proposed rule will redistribute more than $100 million in 1 year. Therefore, we estimate that this rulemaking is “economically significant” as measured by the $100 million threshold, and hence also a major rule under the Congressional Review Act. Accordingly, we have prepared a RIA that, to the best of our ability, presents the costs and benefits of the rulemaking. The RFA requires agencies to analyze options for regulatory relief of small entities. For purposes of the RFA, small entities include small businesses, nonprofit organizations, and small governmental jurisdictions. Most hospitals, practitioners and most other providers and suppliers are small entities, either by nonprofit status or by having annual revenues that qualify for small business status under the Small Business Administration (SBA) standards. (For details, see the SBA's Web site at http://www.sba.gov/content/table-smallbusiness-size-standards (refer to the 620000 series)). Individuals and States are not included in the definition of a small entity.

The RFA requires that we analyze regulatory options for small businesses and other entities. We prepare a regulatory flexibility analysis unless we certify that a rule would not have a “significant economic impact on a substantial number of small entities.” The analysis must include a justification concerning the reason action is being taken, the kinds and number of small entities the rule affects, and an explanation of any meaningful options that achieve the objectives with less significant adverse economic impact on the small entities.

There are over 1 million physicians, other practitioners, and medical suppliers that receive Medicare payment under the PFS. Approximately 95 percent of practitioners, other providers and suppliers are considered to be small entities, based upon the SBA standards. As shown later in this analysis, however, potential losses to these practitioners under the MIPS program are a small percentage of their total Medicare Part B PFS revenue—4 percent in the first year—though rising to as high as 9 percent in subsequent years. On average, practitioners' Medicare billings are only about 22 percent of total revenue, so even those practitioners adversely affected by MIPS would rarely face losses in excess of 3 percent of revenues, the HHS standard for determining whether an economic effect is “significant.” (In order to determine whether a rule meets the RFA threshold of “significant” impact HHS has for many years used as a standard adverse effects that exceed 3 percent of either revenues or costs.) However, because there are so many affected eligible clinicians, even if only a small proportion is significantly adversely affected, the number could be “substantial.” Therefore, we are unable to conclude that an Initial Regulatory Flexibility Analysis (IRFA) is not required. Accordingly, the analysis and discussion provided in this section, as well as elsewhere in this proposed rule, together meet the requirements for an IRFA. We note that whether or not a particular eligible clinician is adversely affected would depend in large part on the performance of that eligible clinician and that CMS will offer significant technical assistance to eligible clinician in meeting the new standards.

In addition, section 1102(b) of the Act requires us to prepare an RIA if a rule may have a significant impact on the operations of a substantial number of small rural hospitals. This analysis must conform to the provisions of section 603 of the RFA. For purposes of section 1102(b) of the Act, we define a small rural hospital as a hospital that is located outside of a Metropolitan Statistical Area for Medicare payment regulations and has fewer than 100 beds. We are not preparing an analysis for section 1102(b) of the Act because we have determined, and the Secretary certifies, that this proposed rule would not have a significant impact on the operations of a substantial number of small rural hospitals.

Section 202 of the Unfunded Mandates Reform Act of 1995 (UMRA) also requires that agencies assess anticipated costs and benefits on State, local, or tribal governments or on the private sector before issuing any rule whose mandates require spending in any 1 year of $100 million in 1995 dollars, updated annually for inflation. In 2016, that threshold is approximately $146 million. This proposed rule would impose no mandates on state, local, or tribal governments or on the private sector because participation in Medicare is voluntary and because physicians and other professionals have multiple options as to how they will participate under MIPS and discretion over their performance. Moreover, HHS interprets UMRA as applying only to “unfunded” mandates. We do not interpret Medicare payment rules as being “unfunded mandates,” but simply as conditions for the receipt of payments from the Federal government for providing services that meet Federal standards. This interpretation applies whether the facilities or providers are private, state, local, or tribal.

Executive Order 13132 establishes certain requirements that an agency must meet when it promulgates a proposed rule (and subsequent final rule) that imposes substantial direct requirement costs on state and local governments, preempts state law, or otherwise has Federalism implications. Since this regulation does not impose any costs on state or local governments, the requirements of Executive Order 13132 are not applicable.

We have prepared the following analysis, which together with the information provided in the rest of this preamble, meets all assessment requirements. The analysis explains the rationale for and purposes of this proposed rule; details the costs and benefits of the rule; analyzes alternatives; and presents the measures we would use to minimize the burden on small entities. As indicated elsewhere in this proposed rule, we are proposing to implement a variety of changes to our regulations, payments, or payment policies to implement statutory provisions. We provide information for each of the policy changes in the relevant sections of this proposed rule. We are unaware of any relevant federal rules that duplicate, overlap, or conflict with this proposed rule. The relevant sections of this proposed rule contain a description of significant alternatives if applicable.

C. Changes in Medicare Payments

Section 101 of the, (1) repeals the Sustainable Growth Rate formula for physician payments in Medicare, and (2) requires that we establish a Merit-based Incentive Payment System for eligible clinician under which the Secretary must use an eligible clinician's composite performance score (CPS) to determine and apply a MIPS adjustment factor to the professional for a year.

Repealing the Sustainable Growth Rate formula eliminated significant and immediate problems with Medicare's physician fee schedule payments, including implausible payment reductions (such as the 21.2 percent decrease that was scheduled for April 1, 2015). The Office of the Actuary estimated that avoiding those payment reductions results in a budgetary cost of $150.5 billion for fiscal years 2015 through 2025 compared to the prior-law baseline. However, that cost is partially offset by other MACRA provisions that are estimated to have a net reduction in Federal expenditures of $47.7 billion.³⁷ The largest component of the MACRA costs is its replacement of scheduled reductions in physician payments with payment rates first frozen at 2015 levels and then increasing at a rate of 0.5 percent a year during calendar years 2016 through 2019. The estimates in this RIA take those legislated rates as the baseline for the estimates we make as to the costs, benefits, and transfer effects of the regulation, with some data collection provisions taking effect in 2017 and substantial payment reforms first taking effect in 2019.

As required by the MACRA, overall payment rates for services for which payment is made under the PFS would remain at the 2019 level through 2025, but starting in 2019, the amounts paid to individual eligible clinicians would be subject to adjustment through one of two mechanisms, depending on whether the eligible clinician meets the threshold for participation in Advanced APMs to be considered a Qualifying APM Participant (QP) or Partial QP, or is instead evaluated under MIPS.

For APMs, from 2019 through 2024, eligible clinicians receiving a substantial portion of their revenue through Advanced APMs and meeting other applicable requirements to become QPs would receive a lump-sum payment after each year equal to 5 percent of their Medicare covered professional services for services reimbursed according to the PFS in the preceding year. The APM Incentive Payment is separate from, and in addition to, the reimbursement for services furnished by an eligible clinician during that year. Eligible clinicians who become QPs would not receive a MIPS performance adjustment under the PFS. Eligible clinicians who do not become QPs, but meet a slightly lower threshold, would be deemed Partial QPs for that year, and may elect to report to and be scored under MIPS. In QP Performance Period 2017, we define Partial QPs to be Advanced APM participants that have at least 20 percent but less than 25 percent, of their Medicare Part B payments for covered professional services through an Advanced APM Entity, or at least 10 percent, but less than 20 percent, of their Medicare patients served through an Advanced APM Entity. If the Partial QP elects to be scored under MIPS, they would be subject to all MIPS requirements and would receive a MIPS payment adjustment. This adjustment may be positive or negative. If an eligible clinician does not meet either of those standards, the eligible clinician would be subject to MIPS and would report to MIPS and receive the corresponding MIPS payment adjustment.

Beginning in 2026, payment rates for eligible clinicians who achieve QP status for a year would be increased each year by 0.75 percent, while payment rates for eligible clinicians who do not achieve QP status would be increased each year by 0.25 percent. MIPS eligible clinicians would receive positive, neutral, or negative adjustments to their PFS payments in a payment year based on performance during a prior performance period. Although the legislation establishes overall payment rate and procedure parameters until 2026 and beyond, this impact analysis covers only the initial payment year (2019) in detail. After 2019, while overall payment levels will be partially bounded, we have also acknowledged in the preamble that the Department will revise its quality and other payment measures and overall payment thresholds and other parameters as eligible clinicians' behavior changes.

As discussed further in the preamble to this proposed rule, we are proposing requirements for MIPS that may result in the exclusion of certain eligible clinicians for various reasons. For example, MACRA requires us to exclude eligible clinicians from MIPS participation if they are QPs, or if they are a type of eligible clinician whose specialty is excluded from MIPS for the 2019 and 2020 MIPS payment years. Additionally, we are proposing above to exclude low volume eligible clinicians, or those with less than $10,000 in allowable claims and fewer than 100 Medicare patients.

We estimated the number of physicians and other professionals that would be excluded from MIPS due to their being QPs using data from APM entities that existed in 2014. First, we identified APM entities that participated in APMs that have similar design characteristics to those proposed for Advanced APMs in section II.F.4.b. of this proposed rule. In 2014, those models included the Pioneer Accountable Care Organization (ACO) Model (which is scheduled to end in 2016), Comprehensive ESRD Care (CEC) (which began in 2015, but used historical data from 2014), and Comprehensive Primary Care Initiative (CPC). Further, we assigned Shared Savings Program ACOs that existed in 2014 their 2016 track assignments because several ACOs have since transitioned to higher risk tracks. Next, we analyzed 2014 claims data to identify the APM Entities within each of those APMs to determine which of those APM Entities met the criteria for having at least 25 percent of their beneficiaries or allowable charges through the APM Entity.

Using those procedures, we arrived at a lower bound estimate that approximately 30,658 physicians and other professionals would become QPs, representing an estimated total incentive payment amount of approximately $146,000,000. However, we expect that the number of QPs may be significantly higher than the estimate based on 2014 data. CMS has continued to introduce new APMs since 2014, and intends to continue to introduce more APMs in future years. We base this expectation on prior experience with increased enrollment in current models and targets for new models that are expected to be adopted in the future. Additionally, CMS anticipates increased participation in currently existing APMs. Our upper bound estimate of QPs, based on the same estimating procedures, is 90,000 and the corresponding estimated total incentive payment is $429,000,000. In this regard, it is longstanding HHS policy not to attempt to predict the effects of future rulemakings, in order to maximize future Secretarial discretion over whether, and if so how, payment or other rules would be changed.

To estimate the number of physicians and other professionals ineligible or excluded due to the proposed low-volume exclusion, ineligible specialties, and newly-enrolled eligible clinicians, we began with a sample of clinicians participating in Medicare B in 2014. We then estimated the number of ineligible clinicians by applying the low-volume exclusion proposed for MIPS—that is, eligible clinicians with less than $10,000 in allowable charges and fewer than 100 Medicare patients—and number of clinicians ineligible for MIPS in Year 1 based on their specialty. We then removed eligible clinicians that were newly enrolled in Medicare.

We have estimated the effects of these various exclusions in Table 61.

We have also estimated the number of clinicians that we believe will be excluded from MIPS in CY 2017 by specialty. Our estimates follow in Table 62. We note that the estimates in Table 62 are based on clinicians in our 2014 data that were in ineligible specialties, newly enrolled, or met the proposed low-volume exclusion. However, due to data limitations, the estimates include only a portion of the 30,658-90,000 QPs that are listed in Table 61 above.

Based on the estimates of excluded providers in Table 61, we estimate that between approximately 687,000 and 746,000 clinicians will be assigned a CPS score in MIPS Year 1. They are clinicians in eligible specialties that (a) are not QPs participating in Advanced APMs (b) exceeded the low volume threshold (c) have been enrolled as Medicare physicians for more than one year, (d) had measures that met or exceeded the relevant case size thresholds.

According to National Health Expenditure data, in 2013, physicians and other professionals received a total of $586.7 billion from all sources. Medicare paid $130.3 billion of that amount. Based on the lower bound total in Table 61 of $13,909 billion in allowed charges for professionals excluded from MIPS, we estimate that less than 11 percent of professionals' Medicare Part B spending for services covered under the Medicare PFS will be excluded from MIPS, and less than 3 percent of all professionals' spending from all sources will be excluded.

We used 2014 VM, PQRS, and other available data to model the scoring provisions described in this regulation. First, we arithmetically calculated a hypothetical CPS for each eligible clinician. Then, we implemented an exchange function based on the provisions of this proposed rule to translate the hypothetical CPS into a negative payment adjustment or positive payment adjustment. This entailed modifying parameters of the exchange function iteratively in order to achieve distributions in payment adjustments that meet requirements related to budget neutrality and aggregate exceptional performance payment amounts. However, because of the lack of historical data for the proposed advancing care information and CPIA measures, this version of the model does not estimate scores for the advancing care information and CPIA performance categories. Based on 2015 Medicare EHR Incentive Program data, we estimate that approximately 226,514 Medicare attesters would receive a 90 percent score in the advancing care information performance category and thereby receive an estimated 23 more points to their CPS, and that 209,000 eligible clinicians receiving a negative adjustment for 2016 would receive an advancing care information performance category score of 0. We also estimate that approximately 412,678 clinicians are non-eligible provider types, and therefore, would not be measured on the advancing care information performance category. Hence we estimate the CPS using only quality and resource use performance category scores, but recognize the scores would adjust by the advancing care information characteristic estimates described above. The model also set a hypothetical performance threshold, and estimated a MIPS payment adjustment associated with each CPS.

The costs for implementation and complying with the advancing care information performance category requirements could potentially lead to higher operational expenses for MIPS eligible clinicians. However, we believe that the combination of payment adjustments and long-term overall gains in efficiency will likely offset the initial expenditures. Additionally, because we are proposing above to reweight the advancing care information performance category scores for eligible clinicians that were exempt from the Medicare EHR Incentive Program or received hardship exemptions, these proposals would not impose additional requirements for EHR adoption during the first MIPS performance period. Health IT vendor may face additional costs in the first year of MIPS if they choose to develop additional capabilities in their systems in order to submit advancing care information and CPIA performance category data on behalf of eligible clinicians.

Additionally, we believe a majority of MIPS eligible clinicians who are able to report the advancing care information performance category of MIPS have already adopted an EHR during Stage 1 and 2 of the prior Medicare EHR Incentive Program. As we have stated with respect to the Medicare EHR Incentive Program, we believe that future retrospective studies on the costs to implement an EHR and the return on investment (ROI) will demonstrate efficiency improvements that offset the actual costs incurred by eligible clinicians participating in MIPS and specifically in the advancing care information performance category, but we are unable to quantify those costs and benefits at this time.

At present, evidence on EHR benefits in either improving quality of care or reducing health care costs is mixed. This is not surprising since the adoption of EHR as a fully functioning part of medical practice is still in its infancy. Even physician offices and hospitals that can meet Medicare EHR Incentive Program standards have not necessarily fully implemented all the functionality of their systems or fully exploited the diagnostic, prescribing, and coordination of care capabilities that these systems promise. Moreover, many of the most important benefits of EHR depend on interoperability among systems and this functionality is still lacking in many EHR systems. A recent RAND report prepared for the ONC reviewed 236 recent studies that related the use of health IT to quality, safety, and efficacy in ambulatory and non-ambulatory care settings and found that—

A majority of studies that evaluated the effects of health IT on healthcare quality, safety, and efficiency reported findings that were at least partially positive. These studies evaluated several forms of health IT: metrics of satisfaction, care process, and cost and health outcomes across many different care settings. The relationship between health IT and [health care] efficiency is complex and remains poorly documented or understood, particularly in terms of healthcare costs, which are highly dependent upon the care delivery and financial context in which the technology is implemented.

Other recent studies have not found definitive quantitative evidence of benefits. We request comments providing better evidence concerning EHR benefits in reducing the costs or increasing the value of EHR-supported health care.

Similarly, the costs for implementation and complying with the CPIA performance category requirements could potentially lead to higher expenses for MIPS eligible clinicians. Costs per full-time equivalent primary care clinician for CPIA will vary across practices, including for some activities or patient-centered medical home practices, in incremental costs per encounter, and in estimated costs per member per month. Costs may vary based on panel size and location of practice among other variables. For example, Magill (2015), conducted a study of PCMH in two states. Magill (2015), found that costs associated with a full-time equivalent primary care clinician, who were associated with PCMH functions, varied across practices. Specifically, Magill (2015) found an average of $7,691 per month in Utah practices, and an average of $9,658 in Colorado practices. Consequently, PCMH incremental costs per encounter were $32.71 in Utah and $36.68 in Colorado (Magill, 2015). Magill (2015) also found that the average estimated cost per member, per month, for an assumed panel of 2,000 patients was $3.85 in Utah and $4.83 in Colorado. However, given the lack of comprehensive historical data for proposed CPIA, we are unable to quantify those costs in detail at this time. We request public comments on the costs associated with CPIA from practices that have implemented clinical practice improvements in the past.

Payment impacts in this proposed rule reflect averages by specialty based on Medicare utilization. The payment impact for an individual eligible clinician could vary from the average and would depend on the mix of services that the eligible clinician furnishes. The average percentage change in total revenues would be less than the impact displayed here because eligible clinicians generally furnish services to both Medicare and non-Medicare patients. In addition, eligible clinicians may receive substantial Medicare revenues for services under other Medicare payment systems that would not be affected by MIPS adjustment factors.

Table 63 shows the estimated payment impact on PFS services of the proposals contained in this proposed rule. To the extent that there are year-to-year changes in the volume and mix of services provided by eligible clinicians, the actual impact on total Medicare revenues will be different from those shown in Table 63. We conducted sensitivity analyses with low, high, and midpoint estimates, and we believe the midpoint estimate represents our best projection of the effects of the MIPS program on Medicare charges. As noted above, given the limitations on the data used for this simulation, differences between specialties are attributable to different performance levels on the quality and resource use performance category measures available from historical PQRS and VM data. Our midpoint estimate, with a performance threshold set at 50, follows as Table 63. Additionally, using the same data, we have estimated the impact on PFS services of the proposals contained in this proposed rule by practice size. That estimate follows as Table 64.

Based on National Health Expenditure data, total Medicare payments for physicians and clinical services expenditures in 2013 reached $130.3 billion. Payments from all sources reached $586.7 billion. Table 63 shows that the aggregate negative payment adjustment for all eligible clinicians under MIPS is estimated at $833 million, which represents less than 1 percent of eligible clinicians' Medicare payments and less than 0.2 percent of eligible clinicians' payments from all sources. Table 63 also shows that the aggregate positive payment adjustment for eligible clinicians under MIPS is estimated at $1.333 billion (including exceptional performance adjustments), which represents approximately 1.02 percent of eligible clinicians' Medicare payments and 0.23 percent of payments from all sources.

D. Impact on Beneficiaries

There are a number of changes in this proposed rule that would have an effect on beneficiaries. In general, we believe that the proposed changes will have a positive impact and improve the quality and value of care provided to Medicare beneficiaries.

More broadly, we expect that over time both the overall MIPS program and increasing participation in APMs will increasingly result in improved quality of care, resulting in lower morbidity and mortality, and in reduced spending, as physicians respond to the incentives offered by MIPS and APMs and adjust their clinical practices in order to maximize their performance on specified quality measures and activities. The various shared savings initiatives already operating have had modest success but have demonstrated that all three outcomes are possible. For example, in August of 2015, we issued 2014 quality and financial performance results showing that Medicare ACOs continue to improve the quality of care for Medicare beneficiaries while generating financial savings. Additionally, in their first years of implementation, both Pioneer and Shared Savings Program ACOs had higher quality care than Medicare fee for service (FFS) providers on measures for which comparable data were available. Shared Savings Program patients with multiple chronic conditions and with high predicted Medicare spending received better quality care than comparable FFS patients. Between the first and third performance periods, Pioneer ACOs improved their average quality score from 73 percent to 87 percent. Taken together, Pioneer and Shared Savings Program ACOs yielded $411 million in cost savings in 2014.

Results from the first year of the CPC Initiative indicate that it has generated nearly enough savings in Medicare health care expenditures to offset care management fees paid by CMS.

• The primary sources of the savings were reduced rates of hospital admissions and emergency department visits.
• The bulk of the savings was generated by patients in the highest-risk quartile, but favorable results were also seen in other patients.
• Over 90 percent of practices successfully met all first-year transformation requirements.
• The expenditure impact estimates differ across the seven regions.
• Additional time and data are needed to assess impact on care quality.

These results should be interpreted cautiously as effects are emerging earlier than anticipated, and additional research is needed to assess how the initiative affects cost and quality of care beyond the first year. Because the effects of the CPC Initiative are likely to be larger in subsequent years, these early results suggest it is likely the model will eventually break-even or generate savings.

Basing reimbursement in part on performance metrics is still an evolving art and, as discussed throughout this preamble, there are multiple variables and as yet no definitive answers as to what combinations of measures, benchmarks, and other variables will achieve the best results over time. Accordingly, we are unable at this time to provide specific dollar estimates of these benefits and cost reductions.

E. Impact on Other Health Care Programs and Providers

The MIPS program is aimed at Medicare FFS physicians and other professionals paid under the PFS. These physicians and other professionals are almost all engaged in serving patients covered by other payers as well. Because Medicare covers only about one person in seven (though a considerably higher share of total healthcare spending, since older persons incur far higher expenses on average than younger persons), for most of those services that will be subject to MIPS payment adjustments, Medicare provides only a fraction of practice revenues. Moreover, it is unlikely that many insurance payers will adopt MIPS or MIPS-like payment models in the short run. Hence, MIPS incentives are necessarily attenuated. On the other hand, changing practices for one group of patients will possibly lead to changes for other patients (for example, EHR systems are almost always used for all patients served by a physician). Physicians and other professionals may find it simpler and more efficient to adopt clinical practice improvements for all patients, regardless of payer, in response to MACRA's incentives, through the use of both MIPS measures and activities and alternative payment models. Furthermore, since MACRA eventually rewards participation in APMs beyond those in Medicare, other payers may start to develop more models in which clinicians and patients can participate. Hence, there are likely to be beneficial effects on a far broader range of patients in the health care system than simply Medicare patients, and we believe those effects would include improved health care quality and lower costs over time. However, we have no basis at this time for quantifying such effects.

We note that large proportions of the Medicare and Medicaid programs are already delivered through capitated insurance payments to HMOs, PPOs, and related organizations. The Medicare Advantage program and related State programs therefore already have substantial incentives to improve quality and reduce costs. MIPS does not affect provider payments under those programs directly, which have their own reimbursement mechanisms for physicians and other professionals. In many but not all cases, those insurance carriers do use incentive mechanisms that are similar in purpose and design to the kinds of APMs that we expect will arise under the new payment adjustments. We would not expect major near-term changes in HMO and PPO payment arrangements, or performance, from any MIPS or APM spillover effects. Regardless, we have no basis at this time for quantifying any such effects.

There are other potentially affected provider entities, including hospitals, skilled nursing facilities, Critical Access Hospitals (largely small rural hospitals), and providers serving unique populations, such as providers of tribal health care services. In none of these cases do we believe that MIPS would have significant effects on substantial numbers of providers. But to the extent that MIPS and increasing participation in APMs over time succeed in improving quality and reducing costs, there may be some beneficial effects not only on patients but also on some providers.

As noted previously in this section of the preamble, and as discussed in this subsection, we have concluded that financial effects on either directly or indirectly affected small entities, including rural hospitals, will be minimal. We welcome comments on these conclusions.

F. Alternatives Considered

This proposed rule contains a range of policies, including many provisions related to specific statutory provisions. The preceding preamble provides descriptions of the statutory provisions that are addressed, identifies those policies where discretion has been exercised, presents our rationale for our proposed policies and, where relevant, analyzes alternatives that we considered. While it is hard to single out any one alternative for public comment, we particularly call attention to the performance threshold and the level at which it is set for scoring purposes under MIPS.

As described above, pursuant to section 1848(q)(6)(D)(i) of the Act, for each year of the MIPS, the Secretary shall compute a performance threshold with respect to which the CPSs of MIPS eligible clinicians are compared for purposes of determining the MIPS adjustment factors under section 1848(q)(6)(A) of the Act for a year. The performance threshold for a year must be either the mean or median (as selected by the Secretary, which may be reassessed every 3 years) of the CPSs for all MIPS eligible clinicians for a prior period specified by the Secretary. Section 1848(q)(6)(D)(iii) of the Act outlines a special rule for the initial 2 years of MIPS, which requires the Secretary, prior to the performance period for such years, to establish a performance threshold for purposes of determining the MIPS adjustment factors under paragraph (A) and an additional performance threshold for purposes of determining the additional MIPS adjustment factors under paragraph (C), each of which shall be based on a period prior to the performance periods and take into account data available with respect to performance on measures and activities that may be used under the performance categories and other factors determined appropriate by the Secretary.

Depending on where the threshold is set within those parameters, the proportions and distributions of MIPS eligible clinicians receiving payment reductions versus positive payment adjustments can change dramatically from our estimates. For example, in Table 63, we estimated (based on available data) that 40.0 percent of Colon/Rectal Surgery specialists will receive a negative payment adjustment under MIPS. Setting the performance threshold at a lower level would enable more Colon/Rectal Surgery specialists to avoid negative adjustments and potentially qualify for more positive adjustments. Conversely, we estimated above that 59.2 percent of Interventional Radiology specialists would receive a positive adjustment under the current proposal. Setting the performance threshold at a higher level would result in fewer Interventional Radiology specialists qualifying for positive adjustments, and potentially more of them receiving negative adjustments. But any payment changes resulting from changes to the performance threshold policy will depend primarily on changes to practices and other responses from MIPS eligible clinicians.

We request comment on these alternatives, on all previous estimates of effects, and on any other issues or options that might improve the substantive effects of this proposed rule, or our estimates of those effects. We are particularly interested in comments on any aspects of this proposed rule that might inadvertently or unintentionally create adverse effects on the delivery of high quality and high value health care, and on options that might reduce such effects.

G. Assumptions and Limitations

We would like to note several limitations to the analyses that estimated eligible clinicians' eligibility, negative payment adjustments, and positive payment adjustments based for the first MIPS performance period (2017) based on 2014 data described above:

• The scoring model cannot reflect that eligible clinicians' behavioral responses to MIPS will be different than their responses to the 2014 PQRS requirements. As with all scoring models based on historical data, the model assumes that the measures reported and the distribution of scores on those measures would be the same under MIPS' first performance period as they were under the 2014 PQRS program. However, the intent of the MIPS program is to incentivize eligible clinicians both in terms of the reporting of measures and in terms of improving the quality of patient care.
• Limited historical data for two performance categories. Because we have limited historical data for the proposed advancing care information and CPIA performance categories, the modeled scoring estimates do not include advancing care information or CPIA performance category scores. The model also set a hypothetical performance threshold and estimated a MIPS payment adjustment for each CPS.
• Some of the MIPS scoring provisions could not be applied because MIPS will have different reporting requirements than PQRS. For example, the proposed MIPS scoring provisions require at least one cross-cutting quality measure, whereas the 2014 PQRS program did not have such a requirement.
• The scoring model does not reflect the growth in Advanced APM participation between 2014 and 2017. Due to data limitations, the scoring model could only identify clinicians that participated in Advanced APMs and would have exceeded the QP threshold in 2014. Several new Advanced APM have been implemented or will be implemented between 2014 and 2017. Further, some clinicians will join the successors of Advanced APMs already in existence in 2014.

Due the limitations above, there is considerable uncertainty around our estimates that is difficult to quantify in detail.

H. Accounting Statement

As required by OMB Circular A-4 (available at http://www.whitehouse.gov/omb/circulars/a004/a-4.pdf ), in Table 65 (Accounting Statement), we have prepared an accounting statement.

We have not attempted to quantify the benefits of this rule because of the many uncertainties as to both provider behaviors and resulting effects on patient health and cost reductions. For example, the applicable percentage for MIPS incentives changes over time, increasing from 4 percent in 2019 to 9 percent in 2022 and subsequent years, and we are unable to estimate precisely how physicians will respond to the increasing incentives. As noted above, in CY 2019, we estimate that we will distribute approximately $833 million in payment adjustments on a budget-neutral basis, which represents the applicable percent for 2019 required under section 1848(q)(6)(B)(i) of the Act and excludes $500 million in exceptional performance payments. In 2020, section 1848(q)(6)(B)(ii) of the Act specifies that the applicable percent will be 5 percent, which we estimate would mean that we will distribute approximately $1,041 million in payment adjustments on a budget-neutral basis, ignoring changes in clinical practice, volume growth, or other changes that may affect Medicare physician payments. Finally, in 2021, section 1848(q)(6)(B)(iii) of the Act specifies that the applicable percent will be 7 percent, which we estimate would mean that we will distribute approximately $1,458 million in payment adjustments on a budget-neutral basis, again ignoring changes in clinical practice, volume growth, or other changes that may affect Medicare physician payments, as well as the $500 million in exceptional performance payments.

Further, the addition of new APMs and participants over time will affect the pool of MIPS eligible clinicians, and for those that are MIPS eligible clinicians, may change their relative performance. The $500 million available for exceptional performance and the 5 percent incentive for QPs are only available from 2019 through 2024. Beginning in 2026, QPs will receive a higher conversion factor than non-QPs. However, we are unable to estimate the number of QPs in those years, as we cannot project the number or types of Advanced APMs that will be made available in those years through future CMS initiatives proposed and implemented in those years, nor the number of QPs for those models.

The percentage of the CPS attributable to each performance category will change over time, and we will incorporate improvement scoring in future years. The CPIA category represents an entirely new category for measuring eligible clinicians' performance. We may also propose policy changes in future years as we continue implementing MIPS and as eligible clinicians accumulate experience with the new system. Moreover, there are interactions between the MIPS and APM incentive programs and other shared savings and incentive programs that we cannot model or project. Nonetheless, even if ultimate savings and health benefits represent only low fractions of current experience, benefits are likely to be substantial in overall magnitude.

The table that follows includes our estimate for MIPS payment adjustments ($833 million), the exceptional performance payments under MIPS ($500 million), and payments to QPs (using the lower bound estimate described in the preceding analysis, $146 million). However, of these three elements, only the budget-neutral MIPS payment adjustments are shown as estimated decreases.

List of Subjects

42 CFR Part 414

• Administrative practice and procedure
• Biologics
• Drugs
• Health facilities
• Health professions
• Kidney diseases
• Medicare
• Reporting and recordkeeping requirements

42 CFR Part 495

• Administrative practice and procedure
• Health facilities
• Health maintenance organizations (HMO)
• Health professions
• Health records
• Medicaid
• Medicare
• Penalties
• Reporting and recordkeeping requirements

For the reasons set forth in the preamble, the Centers for Medicare & Medicaid Services proposes to amend 42 CFR chapter IV as set forth below:

PART 414—PAYMENT FOR PART B MEDICAL AND OTHER HEALTH SERVICES

1. The authority citation for part 414 continues to read as follows:

Authority: Secs. 1102, 1871, and 1881(b)(l) of the Social Security Act (42 U.S.C. 1302, 1395hh, and 1395rr(b)(l)).

2. In § 414.90—

a. Amend paragraph (e) introductory text by removing the phrase “and subsequent years” and adding in its place the phrase “through 2018”.

b. Amend paragraph (e)(1)(ii) by removing the phrase “and each subsequent year” and adding in its place the phrase “through 2018”.

3. Subpart O is added to part 414 to read as follows:

Subpart O—Merit-Based Incentive Payment System and Alternative Payment Model Incentive

PART 495—STANDARDS FOR THE ELECTRONIC HEALTH RECORD TECHNOLOGY INCENTIVE PROGRAM

4. The authority citation for part 495 continues to read as follows:

Authority: Secs. 1102 and 1871 of the Social Security Act (42 U.S.C. 1302 and 1395hh).

5. Section 495.4 is amended by revising the definition of “Meaningful EHR user” to read as follows:

6. Section 495.40 is amended by—

a. Revising paragraph (a) introductory text.

b. Revising paragraphs (a)(2)(i)(E) and (F).

c. Adding paragraphs (a)(2)(i)(G), (H) and (I) and (b)(2)(i)(H) and (I).

The revision and additions read as follows:

7. Section 495.102 is amended by—

a. Revising paragraph (d)(1).

b. Revising paragraph (d)(2)(iv).

c. Revising paragraph (d)(3).

The revisions read as follows:

8. Section 495.316 is amended by revising paragraph (g)(2) and adding paragraph (g)(3) to read as follows:

Note:

The following Appendix will not appear in the Code of Federal Regulations.

Appendix