Civil No. 3:15-cv-01186
Before the court are three motions for judgment on the administrative record. Eastman Chemical Company and the Eastman Chemical Company Welfare Benefit Program (collectively, "Eastman"), have filed a Second Motion for Judgment on the Administrative Record After Remand (Docket No. 103), to which Gary Carty has filed a Consolidated Response (Docket No. 110), and Eastman has filed a Reply (Docket No. 112). The third defendant, Metropolitan Life Insurance Company ("MetLife"), has filed a Second Motion for Judgment on the Administrative Record, in which it adopts and incorporates by reference the arguments raised by Eastman (Docket No. 105) and which Carty addresses in his Consolidated Response, and MetLife filed a Reply (Docket No. 113) adopting and incorporating Eastman's Reply. Finally, Carty has filed a Second Motion for Judgment on the ERISA Record (Docket No. 106), to which Eastman has filed a Response (Docket No. 108) and MetLife has filed a Response adopting and incorporating Eastman's Response (Docket No. 109), and Carty has filed a Reply (Docket No. 114). For the reasons set out herein, the motions filed by Eastman and MetLife will be denied and Carty's motion will be granted.
This action—brought under the civil enforcement provisions of the Employee Retirement Income Security Act, 29 U.S.C. § 1132 et seq. ("ERISA")—concerns the discontinuation of Carty's long term disability ("LTD") benefits. Carty is a former Eastman Chemical Company employee and a beneficiary under the self-funded LTD plan of the Eastman Chemical Company Welfare Benefit Program. (Docket No. 1 ¶¶ 4, 6, 9.) MetLife is the claims administrator of the LTD plan. (Id. ¶ 5.) Carty stopped working for Eastman on November 28, 2013 and was approved for LTD benefits beginning on June 8, 2013. (Id. ¶¶ 14, 17.) MetLife originally terminated Carty's benefits by letter dated April 17, 2015. (Id. ¶ 18.) This is the second time that decision has come before this court for review on the administrative record. (See Docket No. 55.) A. Applicable Plan Provisions
The LTD plan states that its purpose is to "provide continuing income when you are unable to work due to an extended disability." (Administrative Record ("AR") (Docket Nos. 96, 96-1 to -11) at 1070.) The plan lists a number of requirements, all of which must be met "[t]o be considered disabled under the LTD [plan]." (Id. at 1071.) For the first eighteen months of LTD benefits, a claimant will be deemed to be disabled if he is unable to make 80% of his pre-disability salary performing the same essential job functions as he had performed pre-disability. (Id.) To continue to be considered disabled after the first eighteen months, "you may not earn more than 50% of your pre-disability earnings from any employer in your local economy at any gainful occupation for which you are reasonably qualified taking into account your training, education and experience." (Id.) MetLife interprets this provision to provide that a beneficiary is no longer disabled when he is capable of earning more than 50% of his pre-disability earnings, even if he is not actually earning that amount in practice. In a prior Memorandum, the court upheld MetLife's interpretation under an arbitrary and capricious standard but noted that the plan's definition of disability was indeed poorly drafted and would, read literally and in isolation, allow an individual to continue to qualify as disabled merely by keeping his income below the 50% threshold for any reason. (Docket No. 55 at 15-16.)
The plan provides for ongoing review and reconsideration of eligibility: "The Claims Administrator periodically reviews your continued eligibility for LTD plan benefits. You must cooperate with and respond to any request that the Claims Administrator makes relating to your claim." (AR at 1077.) "If the Claims Administrator determines that you are no longer eligible for LTD benefits, your benefits will terminate at the end of the month that determination is made." (Id.) However, "benefits will terminate immediately on . . . the date you are no longer disabled because you are engaging in gainful work[, or] the date the Claims Administrator determines that you are not receiving Appropriate Care and Treatment." (Id. at 1078.) Benefits will also be terminated if the beneficiary "fail[s] to provide the required proof of continuing disability." (Id.)
The plan grants MetLife discretion in construing the plan and determining eligibility:
[MetLife] shall have the complete discretionary authority to control the operation and administration of this Program, with all powers necessary to enable it to properly carry out such responsibilities, including, but not limited to, the power to construe the terms of this Program, to determine status, coverage and eligibility for Benefits and to resolve all interpretive, equitable and other questions that shall arise in the operation and administration of the Program.(Id. at 1062.) Among the powers expressly granted MetLife are the power "to exercise discretion in interpreting the Program, its interpretation thereof in good faith to be final and conclusive on all persons claiming Benefits under the Program." (Id.) Eastman is ultimately responsible for funding any benefits awarded under the program. (Id. at 1014.) B. Carty's Termination of Benefits and Complaint
Carty is a former manager in Eastman's information technology department. In March of 2013, he applied for LTD benefits, citing his bipolar disorder, anxiety, and depression. (Id. at 628.) He was approved for LTD benefits beginning on June 8, 2013. (Docket No. 1 ¶ 17; Docket No. 6 ¶ 17.) A MetLife log shows that its claim specialists continued to monitor Carty's treatment and status at least every few months through early 2015. (AR at 731, 736, 747, 758, 771, 779-80, 788, 796, 803, 810, 819.) In March of 2015, a MetLife senior psychological clinical specialist recommended a full file review of Carty's claim, due apparently to the infrequency of Carty's visits with one of his treating professionals. (Id. at 818-19.) By a letter dated April 17, 2015, MetLife informed Carty that his LTD benefits were being terminated because his level of impairment no longer satisfied the plan's definition of disability. (Id. at 231-32.) Carty appealed the decision. (Id. at 81-190.) MetLife informed Carty by a letter dated October 23, 2015, that it was upholding the termination of his benefits. (Id. at 4.) In the October 23, 2015 letter, MetLife echoed its earlier conclusion that Carty was no longer disabled, but also introduced a ground for termination that had not been clearly identified as a basis for the initial denial: that Carty was in violation of a provision of the plan requiring him to receive "appropriate care and treatment." (Id. at 7.) Shortly thereafter, Carty filed his Complaint in this case, challenging MetLife's determination under ERISA. (Docket No. 1.) Each of the parties moved for judgment on the administrative record. (Docket Nos. 38, 40-41.)
On December 15, 2016, the court granted Carty's motion, denied the defendants' motions, and remanded the case to MetLife for further consideration. (Docket No. 56.) The court found a number of deficiencies in both MetLife's substantive rationale for discontinuing benefits and its conclusory, frequently confusing, communications explaining its reasoning and decisions to Carty. (Docket No. 55 at 16-21.) Specifically, the court held that MetLife improperly relied on opinions that ignored symptoms documented by Carty's psychiatrist, Dr. Ronald Smith, and his psychologist, Dr. Edward Latham. The court wrote that, "[w]hile MetLife ha[d] broad discretion in administering its plan and in choosing which professional opinions to credit," it "had an obligation to take all of Carty's documented symptoms into account." (Id. at 21.) Instead, MetLife relied on the opinions of two file reviewers, who focused on a narrow set of symptoms that excluded some of the impairments that had been most damaging to Carty's ability to obtain and maintain employment, namely his inability to cope with routine stressors or honor basic attendance obligations. (Id. at 18-19.) MetLife also failed to give Carty adequate notice before relying on the alleged inadequacy of his treatment, rather than solely his non-impairment, as a ground for termination of benefits. (Id. at 20.) The court concluded that, "[i]n light of the cumulative deficiencies in MetLife's process and reasoning, . . . MetLife acted arbitrarily and capriciously in concluding that Carty was no longer impaired and that he was not receiving appropriate care and treatment." (Id. at 21.) The court, however, denied Carty's request for an order immediately restoring his benefits and, instead, remanded the case to MetLife for reconsideration. (Id. at 22.) C. MetLife's Reconsideration of its Termination of Benefits
In January 2017, MetLife began taking steps to consider the case on remand. (See AR at 2071.) MetLife sent Carty's counsel a number of written questions and record requests. (Id. at 2069.) Carty provided answers to the questions, as well as records and an assessment by Dr. James Kyser, who was Carty's treating physician from June 2015 to February 2017. In response to a question regarding whether he had been working since having become disabled, Carty wrote:
Mr. Carty has attempted to work as a real estate agent, hoping that such a position would be possible because it does not have the same stress, deadlines, or work rules that his previous position had. The attempt has not been successful and Mr. Carty has been unable to perform sufficient sustained work to cover business expenses like marketing required by the job. When he is in a manic phase, he is able to perform some work, but when is in a depressive phase, he is completely unable to function.(Id. at 1990.)
Dr. Kyser provided a medical evaluation dated June 10, 2015, a date toward the beginning of his treatment of Carty. Dr. Kyser wrote that, on the date in question, Carty was alert and cooperative, his speech was coherent, and his judgment and insight were good. (Id. at 1993.) Dr. Kyser concluded that Carty's "[c]ognition [was] grossly intact" and that Carty exhibited "above average" intelligence. Dr. Kyser also observed, however, that Carty reported frequent, unpredictable panic attacks and that the day of the exam was a "good day." (Id. at 1992-93.)
Dr. Kyser also completed an Assessment of Mental Limitations dated February 13, 2017. (Id. at 1983-85.) The assessment form asked Dr. Kyser to rate Carty's abilities to perform certain tasks as either unlimited, good, fair, poor, or none. (Id. at 1983.) Dr. Kyser rated Carty's abilities as "unlimited" in the following areas: ability to perform activities of daily living independently and appropriately, free of supervision and direction; capacity to interact appropriately, communicate effectively, and engage in other aspects of social functioning; ability to follow work rules; and ability to maintain personal appearance. He rated Carty as "good" with regard to his ability to demonstrate reliability. He rated Carty as "fair" with regard to the following areas: abilities of concentration, persistence, and pace; ability to adapt to stressful circumstances in work or work-like settings where failure to adapt results in repeated episodes of deterioration or decompensation that causes him to withdraw or experience an exacerbation of symptoms; ability to deal with the public; ability to deal with stress of ordinary work; ability to persist at assigned tasks; ability to relate to supervisors and co-workers; ability to work at a consistent pace for acceptable periods of time; and ability to timely complete tasks commonly found in work setting. Dr. Kyser did not rate Carty's abilities as "poor" or "none" in any area. He also wrote that Carty had chronic mood instability with anxiety and that he struggled to deal with people without conflict, along with recurring thoughts and decreased concentration. He noted that Carty had reported being able to work about fifteen hours per week. (Id. at 1984-85.)
Carty further provided records of his treatment by Dr. Douglas Herr and Dr. Alexander Chalko at Hermitage Psychiatric in December 2016 and January 2017. Those records, unfortunately, are handwritten and mostly illegible. (Id. at 1972-79.) Carty provided Assessments of Mental Limitations from Dr. Herr and Dr. Chalko, using the same format as had been used with Dr. Kyser. Dr. Chalko did not rate any of Carty's abilities as "unlimited" or "good." He rated the following abilities as fair: ability to follow work rules; ability to maintain personal appearance; ability to persist at assigned tasks; and ability to relate to supervisors and co-workers. He rated Carty's abilities as "poor" in all remaining categories: ability to perform activities of daily living independently and appropriately, free of supervision and direction; capacity to interact appropriately, communicate effectively, and engage in other aspects of social functioning; abilities of concentration, persistence, and pace; ability to adapt to stressful circumstances in work or work-like settings where failure to adapt results in repeated episodes of deterioration or decompensation that causes him to withdraw or experience an exacerbation of symptoms; ability to deal with stress of ordinary work; ability to deal with the public; ability to deal with stress of ordinary work; ability to demonstrate reliability; ability to work at a consistent pace for acceptable periods of time; and ability to timely complete tasks commonly found in work setting. He did not rate any abilities as "none." (Id. at 1945-46.)
Similarly, Dr. Herr did not rate Carty's abilities as "unlimited" in any area. He rated them as "good" with regard to the following areas: ability to perform activities of daily living independently and appropriately, free of supervision and direction; ability to follow work rules; and ability to maintain personal appearance. He rated Carty's ability to deal with the public as "fair," and rated the following abilities as either "fair" or "poor," depending on the day: ability to persist at assigned tasks; ability to relate to supervisors and co-workers; and ability to timely complete tasks commonly found in work setting. Dr. Herr rated the following abilities as poor: capacity to interact appropriately, communicate effectively, and engage in other aspects of social functioning; abilities of concentration, persistence, and pace; ability to adapt to stressful circumstances in work or work-like settings where failure to adapt results in repeated episodes of deterioration or decompensation that causes him to withdraw or experience an exacerbation of symptoms; ability to deal with stress of ordinary work; ability to demonstrate reliability; and ability to work at a consistent pace for acceptable periods of time. He did not rate any abilities as "none." (Id. at 1916-17.)
MetLife arranged for Carty to take part in an independent medical exam ("IME") performed by neuropsychologist Dr. Pamela Auble. On August 16, 2017, Dr. Auble performed the IME over the course of seven to eight hours. Dr. Auble also reviewed records of Carty's past treatments and assessments. (Id. at 1877-78.) Dr. Auble's IME report states that Carty reported that he worked 10 to 20 hours per week, but not every day. (Id. at 1883.) Dr. Auble concluded that Carty's "[i]ntelligience was estimated to fall within the low average range (21st percentile)," with his skills being particularly poor with regard to visual pattern completion, where he was in the 9th percentile. (Id. at 1884.) She noted that his "memory for new information was lower than expected given his other abilities." Dr. Auble suggested that Carty's psychiatric medications likely had some effect on his memory. (Id. at 1886.) She concluded that her exam, along with the records she reviewed, supported a finding of "some current mild limitations," but not of "consistent and persistent limitations from May 1, 2015, to the present." (Id. at 1887.) She suggested that "[t]he most likely cause of Mr. Carty's mild memory difficulties are his medications" and that, therefore, his limitations related to learning new information might be resolved by changing his medication regimen. (Id. at 1890.)
MetLife sent Dr. Auble's report to Carty, Dr. Herr, and Dr. Kyser. (Id. at 1836.) In response, Carty provided a report by Dr. Eric Engum, a neuropsychologist who had reviewed Carty's records and the report but did not test or examine Carty. (Id. at 1739-47.) Dr. Engum compared the complex responsibilities that Carty's former job at Eastman had required with his cognitive ability apparent from his records, particularly Dr. Auble's IME report. He concluded that, "[w]hile it should be appreciated that individuals suffering from a bipolar II disorder may be expected to experience fluctuations in their cognitive functions incident to the variations in their emotional, behavioral, and physical status," Carty's "decline in cognitive status is both more severe than would be expected and is incompatible with the performance of [the] essential job functions" of his pre-disability position. (Id. at 1740.) Dr. Engum posited that, based on the available descriptions of Carty's responsibilities at Eastman and Carty's educational and professional history, Carty "must have possessed superior cognitive and intellectual skills" prior to the onset of his disability. (Id.) Dr. Engum estimated that Carty could conservatively be assumed to have previously had an IQ in at least the "Bright Normal" range, corresponding with a standard IQ score of 111 to 120. (Id. at 1741.) Based on Dr. Auble's assessment, Dr. Engum placed Carty's current standard IQ at 88, at the upper end of the "Dull Normal" range. Id. According to Dr. Engum, Carty's cognitive abilities were consistent with work "at the semi- skilled, maximally at the skilled[,] level." (Id. at 1742.) Carty's limitations, Dr. Engum concluded, "would specifically disqualify [him] from most, if not all, supervisory and/or management positions." (Id.) Carty would also be unable to perform various types of computer-related work, including any work with "systems that require rapid modification and update" or any job that "requires mastery and understanding of new computer software." (Id.) In addition to those cognitive limitations, Carty's memory was, in Dr. Engum's view, "woefully inadequate for an individual who is expected to perform advanced technical and managerial functions within a large manufacturing company." (Id. at 1743.)
Dr. Engum acknowledged that, while Dr. Auble had similarly concluded that Carty now exhibited lower than average intelligence, she, unlike Dr. Engum, had not attributed that finding to cognitive decline since the advent of his disability. Dr. Engum explained that Dr. Auble had used a method known as the Test of Premorbid Functioning ("TOPF"). The TOPF is a tool for estimating a patient's cognitive functioning prior to the onset of an alleged disability, based on the patient's current ability to read and pronounce words. The TOPF is based on the assumption that the ability to pronounce words is, compared to other cognitive skills, resistant to deterioration based on neurological insult. Dr. Engum conceded that the TOPF is "certainly an accepted method of estimating premorbid cognitive function" but explained that the test had limitations that could lead to errors in its conclusions. He concluded, therefore, that his evaluation of Carty's premorbid functioning based on Carty's educational and professional achievements was more reliable. He expressed skepticism that a person with cognitive function in the low average range could, as Carty at one point did, obtain a bachelor's in science degree, with a double major in mathematics and computer science. (Id. at 1744.)
Dr. Engum also noted that the reliability of Dr. Auble's neuropsychological testing was limited by the cyclical nature of Carty's bipolar II disorder. While Dr. Auble's testing would have provided an accurate snapshot of Carty's functioning at the time of the tests themselves, Dr. Engum argued, Carty's functioning appeared to be cyclical and highly variable, meaning that the snapshot might not necessarily capture him at either his lowest- or highest-functioning. (Id. at 1744-45.) Although Dr. Engum's report was mostly dedicated to Carty's cognitive functioning, he also noted the emotional and behavioral obstacles to productive work noted by other healthcare providers. (Id.)
Carty also provided MetLife with a questionnaire letter sent by his counsel to Dr. Herr. Carty's counsel first asked whether Carty's condition was variable from day to day, and Dr. Herr agreed that it was. Carty's counsel next asked whether Carty's limitation was related to his prescribed medications. Dr. Herr refused to provide a yes-or-no answer, answering that he could not know whether that was the case without taking Carty off the medications, which Dr. Herr concluded would likely be unsafe. Finally, Carty's counsel asked whether Carty's medications were necessary for the treatment of his bipolar II disorder, and Dr. Herr stated that they were. Dr. Herr also added a handwritten note, disagreeing with Dr. Auble's conclusion that Carty had not suffered a cognitive decline. Dr. Herr, like Dr. Engum, concluded that Carty's current documented cognitive function was inconsistent with his history of academic attainment. (Id. at 1604-05.)
Carty's counsel sent a different questionnaire to Dr. Chalko. First, Carty's counsel asked whether Carty's medications were necessary, and Dr. Chalko stated that they were. Second, Carty's counsel asked whether Dr. Chalko agreed with a portion of Dr. Auble's assessment describing Carty's interpersonal difficulties and low frustration tolerance. Dr. Chalko confirmed that he agreed with that portion of the assessment. (Id. at 1599-1600.)
Carty also provided a report from a vocational consultant, Mark Boatner, who had reviewed the various assessments and reports that had been prepared regarding Carty. Boatner was asked whether there would be any occupations in Carty's geographic area in which Carty could make 50% of his $126,200 pre-disability salary—in other words, any jobs in which Carty could, in light of his limitations, make at least $63,100 per year. Boatner replied that, based on the assessments of Carty's cognitive functioning made by Dr. Auble and Dr. Engum, Carty did not appear capable of performing work that would offer a salary at that level. (Id. at 1614-15.) He provided a list of job titles from the Dictionary of Occupational Titles that were defined as consistent with Carty's current learning ability, such as data entry clerk, nurse assistant, and cutting machine operator, as well as a range of salaries in the 10th and 50th percentile for those jobs in both Kingsport, Tennessee, where Carty had worked for Eastman, and in Davidson County, where he now lived. In neither metropolitan area did the 50th percentile salary for any of the listed positions approach $63,100. The highest of the 50th percentile salaries was below $39,000 per year. (Id. at 1616-17.) Boatner also concluded that the assessments of Carty's limitations and abilities by Dr. Herr and Dr. Chalko were inconsistent with performing his old job or a similar job. (Id. at 1618.)
Boatner concluded that Carty's part-time work as a real estate agent for a company selling rural real estate had no bearing on his ability to make the salary level required to be non-disabled under the plan. He noted that Carty's real estate work was done by appointment only and required only "modest effort," given the low volume of work he was doing and the amount of activity he could perform online. According to Boatner, "the real estate activity is part time and he is able to schedule meetings on his better days when his symptoms are not completely preclusive of at least a few hours of activity outside his home." (Id. at 1619.)
MetLife provided the additional materials from Dr. Herr, Dr. Chalko, and Boatner to Dr. Auble for her review and response. On November 7, 2017, Dr. Auble provided MetLife with an addendum to her report. She wrote that the materials provided did not affect her earlier conclusions, because they provided no additional treatment information that would affect her own analysis of Carty, his functioning, and his history. With regard to Dr. Engum's disagreement with her regarding whether Carty had experienced a cognitive decline, she wrote that Dr. Engum's own methodology—relying on self-reported educational history—was itself flawed, as clinicians often fail to appreciate that intelligence levels vary widely even within the same level of educational attainment. (Id. at 1579-80.) She defended her use of the TOPF but conceded that pre-disability testing or school records would have been a better source for analyzing his premorbid cognitive ability, had they been available. (Id. at 1580.) She agreed with Dr. Engum and Dr. Herr, however, that Carty's memory, at least, had deteriorated, and she reiterated her conclusion that the most likely cause of his memory difficulties was his medication. (Id.)
With regard to Boatner's analysis, Dr. Auble wrote, "I am not a vocational expert and would defer review of Mr. Boatner's report to a more qualified [reviewer], as this is outside the scope of a psychological expert. I do understand intelligence testing and would note that there is variability in intelligence that occurs within professions." (Id. at 1579.) She provided, as examples, wide ranges in documented verbal comprehension among managers and executives. (Id.)
On November 14, 2017, MetLife issued a denial of Carty's LTD benefits after April 30, 2015. (Id. at 1571-74.) MetLife wrote that the benefits were denied "because the new medical information received does not support consistent and persistent debilitation to totally preclude or severely limit [Carty] in all work related activity for the entire period from May 1, 2015." (Id. at 1571.) Although the letter included a bulk summary of plan provisions that included the definition of disability as relative to 50% of Carty's pre-disability salary, the denial letter included no direct discussion of that provision. To the contrary, the letter repeatedly discussed the determinative issue as whether Carty was "preclude[d]" from or "incapacitat[ed] from engaging in work related activity." (Id. at 1573.)
Carty filed an administrative appeal of the denial. (Id. at 1524.) MetLife ordered another IME, this time with neuropsychologist Dr. B. Charles Ihrig. (Id. at 1511.) Ihrig performed an in-person examination and reviewed Carty's medical records. MetLife asked Dr. Ihrig whether Carty's records demonstrated "functional limitations due to a cognitive condition, as of 5/1/15." (Id. at 1306.) He answered:
Yes and no. There is no documentation in the records from 5/1/15 until 8/15/17 that supports functional limitations due to a cognitive condition. However, on 8/16/17 neuropsychologist Dr. Auble does document a change in cognitive functioning, so from 8/16/17 forward, there is documentation to support functional limitations due to some mild to moderate cognitive changes . . . .(Id.) Dr. Ihrig noted that Carty's early treatment notes, from Dr. Latham and Dr. Smith, showed evidence of mood instability that caused Carty to be overwhelmed by tasks and attendance requirements, but those records did not portray Carty as having a limited cognitive capacity. (Id.) Dr. Ihrig concluded that, in his opinion, Carty had no functional cognitive limitations until August 16, 2017, and that he had limitations thereafter that were "largely based on sedation secondary to [his] medication regimen." (Id. at 1307.) Carty's cognitive weaknesses, Dr. Ihrig observed, were likely "cumulative over the course of his treatment." (Id. at 1308.) He noted, in particular, that Carty had a history of long-term treatment with benzodiazepines and was currently on large doses of Seroquel. Chronic benzodiazepine use, Dr. Ihrig explained, has been demonstrated to be a causal factor in cognitive decline, and Carty's use allegedly dated back over a period of twenty years. The effects of that cognitive decline could then be exacerbated by the sedating effects of the Seroquel. (Id. at 1308-09) As a result, Carty suffered from "sedation, confusion, executive dysfunction, reduced vigilance and attention, and reduced short term memory functioning." (Id. at 1309.) There was, moreover, no way to counteract those effects other than to change Carty's prescription regimen. (Id.) Other than summarizing Carty's records and self-reported history, Dr. Ihrig's report included minimal discussion of the effect of Carty's original bipolar II disorder symptoms on his ability to work. (Id. at 1299-1309.)
MetLife provided a copy of Dr. Ihrig's report to Carty. In response, Carty resubmitted his records, as well as an additional report by Dr. Herr dated May 3, 2018. (Id. at 1164.) Dr. Herr stressed that Carty continued to suffer from limitations related to his bipolar II symptoms that were in addition to any limitations related to his cognitive decline. Dr. Herr wrote that "Carty's Bipolar II condition with mixed hypomania and rapid cycling is one of the most severe non-psychotic cases I have seen and causes significant limitations in his ability to function in any work or social setting." (Id.) Dr. Herr also wrote that more recent meetings with and treatment of Carty had led him to believe that some of his earlier assessments of Carty's abilities were incorrect. For example, he would now rate Carty's ability to follow work rules as "fair," rather than "good," and his ability to deal with the public as "poor," rather than "fair." (Id.) In Dr. Herr's opinion, Carty "could [not] function in a work capacity on a sustained basis" other than in a "highly controllable and part time environment." (Id. at 1165.) Dr. Herr agreed that Carty had suffered cognitive decline, although he argued that the decline was likely attributable, at least in part, to Carty's bipolar II disorder itself. Dr. Herr agreed, however, that medication may have played a role. (Id. at 1166-67.) He noted that the medications on which Carty relied "are unpleasant at best and very disruptive to many patients' wellbeing, and the only reason that they are used is that they are better than the alternative, which would be to let the disease progress unchecked while the patient suffers more acute episodes and a worse decline over time." He posited that, if Carty had been left untreated by medication, he "would very likely be dead due to suicide or incarcerated due to misbehavior of some kind." (Id. at 1167-68.)
MetLife provided Dr. Herr's report to Dr. Ihrig, who issued a supplemental report stating that Dr. Herr's report did not change his earlier opinion, with some caveats. Regarding Carty's ability to work, Dr. Ihrig wrote the following:
I have little doubt that Mr. Carty is capable of functional employment[,] and [he] admits to being employed for an extended period of time. He has found a job that allows for some flexibility in scheduling. He has engaged with enough clients in a positive manner that he has been able to produce income and maintain his job. There is a functional reality here that is hard to ignore. If he were so impaired and performing so poorly, no company would be risking their reputation to keep him around. I feel that the record supports some limitation and inconsistency that reasonably may reduce his number of hours worked.(Id. at 1110-11.) He complained, however, that the nature of the questions posed by MetLife for his earlier report had led him to offer an analysis based on whether Carty was capable of productive work at all:
I believe that my prior assessment was limited to an all or nothing questioning and the reality is that Mr. Carty is impaired but capable of being produc[tive] and working in some fashion. He would likely not currently be able to maintain full employment given the descriptions offered.(Id. at 1111.) Dr. Ihrig also admitted that his original report's focus on cognitive functioning "may have been confining in my scope and specificity of my responses." (Id.) He conceded that "[t]he reality is that Mr. Carty does appear to suffer from a major mental illness that has affected him broadly and likely cognitively." (Id. (emphasis added).)
Finally, Dr. Ihrig noted that, based on MetLife's formulation of the questions to be addressed by his reports, he had focused on Carty's level of impairment in May of 2015, despite the fact that the evidence of poor cognitive functioning—the admitted focus of his report—did not arise until later:
The currently posed question . . . is . . . limiting as it states "as of 5/1/2015" and the provided information states a level of impaired functioning that is current. If the question had been "subsequent to that date" or "currently" the answer might be different. Earlier documentation still falls short. and this date is not clearly addressed in the current provided documents. Instead, it outlines impairment in the last year. This new evidence provides no sense of retrospective documentation or reference to any time period around 5/1/2015. Therefore, I stand by my prior conclusions regarding timeframe, level of impairment, and documentation of his illness and associated reduction of capacities but with the caveat that the documentation in the last year appears to support the possibility that he may only be capable of part time employment.(Id.)
By a letter dated May 20, 2018, MetLife informed Carty that it was upholding the discontinuation of his LTD benefits. (Id. at 1104-08.) Unlike MetLife's initial denial letter, this letter described its decision in reference to the correct definition of disability under the plan:
For the reasons detailed below, we are upholding the termination of his claim. The available medical documentation did not support [that] Mr. Carty would have been prevented from earning more than 50% of his pre-disability earnings from any employer in his local economy at any gainful occupation for which he was reasonably qualified taking into account his training, education and experience, as of May 1, 2015 ongoing.(Id. at 1104.) MetLife reiterated its conclusion in terms of the 50% of pre-disability earnings benchmark near the end of the letter. (Id. at 1107.) At no point in between did it perform an analysis under that standard or explain, specifically, how or why Carty could be expected to make at least $63,100 per year. The letter did, however, cite Dr. Ihrig's conclusions that Carty "was able to contribute and be productive" and "had been able to produce income and maintain his job" selling real estate. The letter acknowledged that Dr. Ihrig had eventually agreed that Carty was not currently able to maintain full employment but did not explain how being "able to contribute and be productive" for significantly fewer than forty hours per week would translate to an over $60,000 annual salary for a person with Carty's present level of mood instability and/or cognitive function. (Id. at 1106.)
On July 23, 2018, MetLife filed a Motion to Quash Subpoena and/or Motion for Protective Order with the court, seeking to prevent Carty from obtaining a deposition of Dr. Ihrig. (Docket No. 81). On July 25, 2018, Carty filed a Motion for Sanctions (Docket No. 87), arguing that MetLife's denial of his appeal without allowing him a chance to review and respond to Dr. Ihrig's second report violated the post-remand Scheduling Order entered by the court on December 19, 2017 (Docket No. 80). The court granted both motions in part. The court held that Carty had not shown that he was entitled to depose Dr. Ihrig but stayed proceedings to allow Carty the opportunity to supplement the administrative record below with any response he had to Dr. Ihrig's report. (Docket No. 91 at 11.) Shortly thereafter, Carty informed the court that he had concluded that no supplementation was necessary (Docket No. 92), and the court lifted the stay (Docket No. 93). The parties' respective motions for judgment on the record followed. (Docket Nos. 103, 105-06.)
II. LEGAL STANDARD
A denial of benefits challenged under ERISA is subject to de novo review unless the benefits plan gives the administrator discretionary authority in interpreting the plan and determining employee eligibility. Firestone Tire & Rubber Co. v. Bruch, 489 U.S. 101, 115 (1989). When a plan administrator has such discretionary authority, a court reviews a decision to deny benefits only for whether it was arbitrary or capricious. Miller v. Metro. Life Ins. Co., 925 F.2d 979, 983 (6th Cir. 1991) (citing Firestone, 489 U.S. at 115). "The arbitrary or capricious standard is the least demanding form of judicial review of administrative action." Davis ex rel. Farmers Bank & Capital Trust Co. of Frankfort, Ky. v. Ky. Fin. Cos. Ret. Plan, 887 F.2d 689, 693 (6th Cir. 1989) (quoting Pokratz v. Jones Dairy Farm, 771 F.2d 206, 209 (7th Cir. 1985)). Under this standard, the determination of an administrator will be upheld if it is "rational in light of the plan's provisions." McClain v. Eaton Corp. Disability Plan, 740 F.3d 1059, 1064 (6th Cir. 2014) (quoting Marks v. Newcourt Credit Grp., Inc., 342 F.3d 444, 457 (6th Cir. 2003)). Stated differently, a claim administrator's decision is not arbitrary and capricious if it "is based on a reasonable interpretation of the plan." Shelby Cty. Health Care Corp. v. S. Council of Indus. Workers Health & Welfare Trust Fund, 203 F.3d 926, 933-34 (6th Cir. 2000).
Courts use both "arbitrary and capricious" and "arbitrary or capricious," interchangeably, to refer to this standard of review, sometimes in the same case. See, e.g., Cochran v. Trans-Gen. Life Ins. Co., 12 F. App'x 277, 281 (6th Cir. 2001). Which term is used has no bearing on the substance of the standard, at least in the ERISA context.
While this review is "not without some teeth, it is not all teeth." McClain, 740 F.3d at 1064. "A decision reviewed according to the arbitrary and capricious standard must be upheld if it results from a deliberate principled reasoning process and is supported by substantial evidence." Id. (quoting Schwalm v. Guardian Life Ins. Co. of Am., 626 F.3d 299, 308 (6th Cir. 2010)). The court must review "the quality and quantity of the medical evidence and the opinions on both sides of the issues" to determine whether a reasoned explanation exists to support an administrator's decision. McDonald v. W.-S. Life Ins. Co., 347 F.3d 161, 172 (6th Cir. 2003). "When it is possible to offer a reasoned explanation, based on the evidence, for a particular outcome, that outcome is not arbitrary or capricious." McClain, 740 F.3d at 1065 (citing Shields v. Reader's Digest Ass'n, Inc., 331 F.3d 536, 541 (6th Cir. 2003)). Moreover, a court must accept an administrator's rational decision, if it is not arbitrary or capricious, even in the face of an equally rational interpretation of a plan offered by a participant. Gismondi v. United Techs. Corp., 408 F.3d 295, 298 (6th Cir. 2005) (citing Morgan v. SKF USA, Inc., 385 F.3d 989, 992 (6th Cir. 2004)).
When analyzing whether an administrator's decision was arbitrary or capricious, a court should consider a potential conflict of interest arising when a plan administrator both evaluates claims for benefits and pays benefit claims. Metro. Life Ins. Co. v. Glenn, 554 U.S. 105, 112 (2008). Even where the claims administrator does not directly fund the benefits, a conflict may nevertheless exist to the extent that a company may be inclined to contract with an administrator more likely to deny claims. Id. at 114-15. In addition, "a plan administrator, in choosing the independent experts who are paid to assess a claim, is operating under a conflict of interest that provides it with a clear incentive to contract with individuals who are inclined to find in its favor that [a claimant] was not entitled to continued [disability] benefits." Kalish v. Liberty Mut./Liberty Life Assur. Co., 419 F.3d 501, 508 (6th Cir. 2005) (quoting Calvert v. Firstar Fin., Inc., 409 F.3d 286, 292 (6th Cir. 2005)). Any potential conflict of interest does not change the standard of review, but it is a factor to consider in determining whether an administrator's decision was arbitrary and capricious. Glenn, 554 U.S. at 115-16.
Carty argues that MetLife's decision was arbitrary and capricious for four reasons: (1) the conclusion that Carty was not disabled was arbitrary and capricious in light of the evidence presented; (2) MetLife arbitrarily and capriciously failed to perform a vocational review to determine Carty's earning capacity; (3) MetLife improperly relied on Dr. Ihrig's conclusions regarding Carty's level of impairment in 2015, which ignored the impairments documented by Dr. Smith and Dr. Latham; and (4) MetLife improperly limited the opinions it obtained to the topic of Carty's impairment in 2015, failing to account for his impairments since. These four issues can ultimately be collapsed into two questions. First, was it arbitrary and capricious to conclude, based on the evidence provided, that Carty is, or at some point during the relevant time period was, capable of earning an annual salary of $63,100? And, second, did MetLife act arbitrarily and capriciously in its analysis regarding the nature of Carty's impairment over time? A. Evidence of Carty's Ability to Earn 50% of His Pre-Disability Salary
MetLife argues that its conclusion that Carty was not disabled is supported by evidence, for which it cites various portions of Carty's medical records that described aspects of his mental health or cognition in positive terms. For example, Dr. Chalko, in the standard checklist that accompanied the notes of his visits with Carty, regularly described Carty as being "well groomed" and "fully oriented," with "appropriate" affect, "normal" mood, "intact" thought processes, "intact" judgment, and "good" insight. (AR at 1137-43 (capitalization altered throughout).) As the court explained in its first opinion in this case, however, a finding of non-disability cannot be based solely on a laundry list of symptoms that a claimant does not have. (Docket No. 55 at 18.) Carty's disability or lack thereof must be considered in light of the symptoms he does have, and the very records that MetLife cites also depict Carty as chronically anxious, heavily medicated, and continuing to suffer from symptoms related to his bipolar depression. (AR at 1137-43.) Just as importantly, Carty's claim must be considered in light of his symptoms' bearing on his ability to perform paid labor at the level defined by the plan as distinguishing between disability and non-disability. In this case, that line was drawn at the ability to earn an annual salary of $63,200. For MetLife's decision not to be arbitrary and capricious, it must have had some basis for concluding that such a salary was attainable by Carty.
The Sixth Circuit has held that a plan administrator does not necessarily act arbitrarily and capriciously by basing its denial on medical records alone, without obtaining a separate vocational review. See Judge v. MetLife Co., 710 F.3d 651, 661-62 (6th Cir. 2013) (collecting cases); Burge v. Republic Engineered Prods., Inc., 432 F. App'x 539, 550 (6th Cir. 2011) ("[The plan administrator] was . . . not required to consider vocational evidence, as opposed to medical evidence, in analyzing [the claimant's] claim."); Douglas v. Gen. Dynamics Long Term Disab. Plan, 43 F. App'x 864, 870 (6th Cir. 2002) (holding that the plan administrator was not required to offer testimony from a vocational expert as to the types of jobs plaintiff could perform). For example, in Judge v. MetLife, the claimant's LTD plan, unlike Carty's, provided benefits only if the claimant was "expected never again to be able" to perform his job or "[a]ny other job for which [he was] fit by education, training or experience." The claimant's medical records demonstrated that he was only permanently impaired with regard to heavy lifting, and the Sixth Circuit held that, given the limited nature of the claimant's restrictions and the narrow definition of disability under the plan, the administrator was permitted to infer, without the aid of specific vocational evidence, that the claimant was able to perform some job that would preclude a disability award. Judge, 710 F.3d at 654, 660-62.
In Douglas v. General Dynamics, the Sixth Circuit relied on a similarly demanding definition of disability to uphold a plan administrator's denial of benefits, without vocational evidence, for a claimant suffering from depression. 43 F. App'x at 867, 871.
Eastman's plan, however, has a broader definition of disability, and Carty's condition has a more complex relationship with his employability than a simple restriction on a few physical tasks. The court will not repeat its analysis from its first opinion, but the evidence prior to the remand included documented symptoms that would inhibit participating in a workplace requiring significant interpersonal interaction or regular, reliable attendance, as well as an inability to cope with routine stressors. The contrary opinions on which MetLife originally relied, as the court held, did not meaningfully rebut those impairments. Although the evidence assembled since the remand shows a changed picture, it does not show an end to impairments. To the contrary, current evidence shows that, in addition to his bipolar II disorder, Carty now exhibits significant cognitive deficits that would impede his ability to perform many jobs or to learn new skills that might be used for others. Carty's originally documented symptoms, alone, would have prevented Carty from performing his own job or any job very similar to it. His now-documented cognitive symptoms have added an additional hurdle. The determinative question, then, is whether he is able to make the threshold salary for non-disability, despite those limitations.
Plans that define disability in terms of earning a particular sum or working in a specific field are particularly likely to require more than a conclusory analysis regarding a claimant's occupational prospects. For example, in Elliott v. Metro. Life Ins. Co., 473 F.3d 613 (6th Cir. 2006), the claimant's plan defined disability for the time period in question in terms of whether she was "unable to earn more than 80% of [her] Predisability Earnings or Indexed Predisability Earnings at [her] Own Occupation for any employer in [her] Local Economy." Id. at 617. Her plan administrator, based solely on medical data and no occupational analysis, concluded that she was not disabled. The Sixth Circuit ruled that the decision was arbitrary and capricious because "medical data, without reasoning, cannot produce a logical judgment about a claimant's work ability." Id. Therefore, even though the plan relied on "numerous medical evaluations," its decision was arbitrary and capricious because it "did not rely on an application of the relevant evidence to the occupational standard when it denied her claim initially and on internal appeal." Id.
A "focus on the written terms of the plan is the linchpin of" ERISA. Heimeshoff v. Hartford Life & Acc. Ins. Co., 571 U.S. 99, 108 (2013) (citing Varity Corp. v. Howe, 516 U.S. 489, 497 (1996)). MetLife, therefore, had an obligation to consider whether Carty was disabled, not in some generic sense or as the term is used in other plans, but as the term is defined in the Eastman LTD plan in which Carty participated. Carty is certainly not the most disabled person to seek disability benefits. He is capable of doing some work and making some money. His LTD plan, however, does not require him to be so disabled that he cannot perform any paid work in order to receive benefits. It requires only that his disability has rendered him unable to earn 50% of his $126,200 pre-disability salary. MetLife has simply failed to identify any basis for assuming that Carty, now or in 2015, is or was capable of making that much money. In 2015, he was incapable of meeting the basic attendance and behavioral obligations of the work for which he was most qualified. Now, those symptoms persist, albeit possibly with more effective treatment, but he also has been shown to suffer from a below-average IQ with learning and memory impairments, possibly related to the medications he requires to treat his condition. The court fails to see how MetLife could justify assuming that a man with an IQ in the high 80s, who cannot reliably work a full work week (and may not be able to do even that unless he has control over his own schedule and work environment), could be expected to make over $60,000 a year in Middle Tennessee.
MetLife's initial denial letter avoided that issue by ignoring the standard set forth in Carty's plan and applying an entirely different standard of disability. Its denial letter on administrative appeal corrected that error superficially but not in its reasoning. Indeed, MetLife relied, in large part, on the IME by Dr. Ihrig, despite Dr. Ihrig's having specifically complained that MetLife had limited him to "all or nothing" analysis regarding the extent of Carty's disability. Dr. Ihrig's comments on Carty's ability to work seem to assume the same erroneous standard initially used in the initial denial letter. Dr. Ihrig discussed whether Carty was "capable of functional employment." He based that conclusion on the "functional reality" that Carty had performed part-time work selling real estate—work for which there is no evidence suggesting that Carty was being paid even close to $63,100 per year. Dr. Ihrig, moreover, agreed that the "record supports some limitation and inconsistency that reasonably may reduce [Carty's] number of hours worked." (AR at 1110-11.) Nevertheless, MetLife treated Dr. Ihrig's IME as supportive of its decision. While MetLife eventually paid lip service to the correct standard of disability, the reasoning on which it relied treated Carty as non-disabled as long as he was able to work at all. Applying a "total disability" standard where none exists in a claimant's plan is arbitrary and capricious, regardless of whether the administrator was required to rely on vocational evidence in doing so. See Burge, 432 F. App'x at 550.
While MetLife was not specifically required to rely on a vocational expert, it was required to base its conclusion on at least some combination of evidence and reasoning that rationally addressed Carty's ability to make an annual salary of at least $63,100. It failed to do so, relying instead on a rote recitation of the medical evidence followed by a conclusory statement of its vocational impact. MetLife's decision, therefore, was arbitrary and capricious with regard to its conclusion that Carty was not and is not disabled. B. Treatment of the Evidence Available in 2015 vs. the Evidence Later Arising
Carty raises a number of issues related to MetLife's treatment of the evidence at various stages of his alleged disability—for example, that MetLife improperly relied on reports that had been limited to Carty's condition in May 2015 and that Dr. Ihrig's analysis of Carty's impairments at that time ignored the actual contemporaneous evidence. Given that Carty contends that he has been disabled continuously up to and including the present, whereas MetLife claims he has been non-disabled for the past nearly four years, there is nothing inherently wrong with MetLife's seeking and relying on opinions that isolate the question of his condition at specific points, such as at the time of denial. Nevertheless, there are some aspects of MetLife's analysis, particularly with regard to the Dr. Ihrig's IME, that cast further doubt on the rationality of its process.
For example, Dr. Ihrig's analysis was, he has now admitted, focused on cognition. (AR at 1111.) That focus was specifically requested by MetLife, which requested Dr. Ihrig's opinion on whether "the medical information support[ed] functional limitations due to a cognitive condition, as of 5/1/15." (Id. at 1306.) There is, however, ample evidence suggesting that Carty's cognitive impairment has been a progressive condition, possibly related to his medication, that did not present itself to healthcare providers or evaluators until after 2015. More importantly, Carty's claim of disability has never been—and, in those earlier years, decidedly was not—solely about impaired cognition. In other words, MetLife sought an opinion on what was likely a later-arising symptom but specifically limited that opinion to a date when that symptom had not yet been detected. That is not to say that Dr. Ihrig's insight on the matter lacked all value—it is certainly relevant to the question of Carty's disability that he may not have been cognitively impaired in 2015. Nevertheless, the temporal and subject-matter restrictions placed on Dr. Ihrig's initial report severely constrained its probative value, as Dr. Ihrig himself explained to MetLife.
MetLife's subject-matter limitations on Dr. Ihrig also explain the alleged short shrift that Dr. Ihrig gave to the opinions and observations of Dr. Latham and Dr. Smith. The issue is not that Dr. Ihrig ignored those perspectives. The issue is that Dr. Latham and Dr. Smith were treating Carty for potentially disabling symptoms other than impaired cognition. At the time of his initial disability award and the time of the discontinuation of benefits, Carty was suffering from a number of bipolar II-related symptoms that interfered with his ability to function in a workplace. Dr. Smith and Dr. Latham both provided relevant information with regard to those symptoms. Based on the cognitive focus that MetLife had requested from Dr. Ihrig, however, that information had minimal bearing on his analysis.
MetLife was free to limit the scope of the questions it asked Dr. Ihrig as it saw fit. In order to exhibit a rational decision-making process, however, it needed to recognize those limitations in its own analysis. MetLife's final denial letter failed to do so. To the contrary, MetLife wrote that Dr. Ihrig concluded that Carty's records "did not support cognitive or psychiatric functional limitations, as of May 1, 2015." (AR at 1105.) But that is not what Dr. Ihrig was asked; the question and answer section of his report includes no question about psychiatric limitations. This is not some distinction noticed by Carty in the context of this judicial review. Dr. Ihrig himself warned MetLife that his report's "focus on cognitive functioning may have been confining in [the] scope and specificity of [his] responses." (Id. at 1111 (internal quotation marks omitted).) Dr. Ihrig followed that warning by stating that, in his opinion, "Carty does appear to suffer from a major mental illness that has affected him broadly," as well as "likely cognitively." (Id.) Dr. Ihrig's analysis, by his own characterization, simply does not support the broad conclusions that MetLife drew from it. MetLife's failure to account for the temporal and subject-matter limitations of Dr. Ihrig's report further supports the conclusion that it did not engage in a rational, principled decision-making process in concluding that Carty was no longer entitled to benefits. C. Remedy
Carty asks the court to order his claim approved and to require MetLife to pay retroactive benefits, with interest. The Sixth Circuit has suggested that, "where the 'problem is with the integrity of [the plan's] decision-making process," rather than 'that [a claimant] was denied benefits to which he was clearly entitled,' the appropriate remedy generally is remand to the plan administrator." Elliott, 473 F.3d at 622 (quoting Buffonge v. Prudential Ins. Co. of America, 426 F.3d 20, 31 (1st Cir. 2005)). That is especially true when complex medical issues are involved; in such a case, the courts "ought to be doubly reluctant to don the white coats ourselves and say that the record permits just one medical conclusion." Cooper v. Life Ins. Co. of N. Am., 486 F.3d 157, 174 (6th Cir. 2007) (Sutton, J., concurring in part and dissenting in part). The records here clearly permit more than one medical conclusion; indeed, the various examiners and reviewers reached different conclusions about a wide range of topics related to Carty's symptoms, from the severity of his psychiatric limitations, to the cause of his cognitive limitations, to the steepness of his cognitive decline. Those disagreements, however, are largely irrelevant to MetLife's primary error—its failure to engage in at least some reasoning tied to Carty's vocational prospects.
Even taking a conservative view of Carty's disability and crediting the analyses that found him the least impaired, it is clear that Carty's psychological and cognitive limitations circumscribe his job prospects considerably. The only evidence in the record regarding what those job prospects are likely to be is Boatner's assessment, and Boatner did not find that Carty is even close to being able to earn half of his former salary. MetLife, of course, might have rationally concluded that Carty was less impaired than Boatner assumed for the purposes of his analysis. Even based on those rosier assumptions, however, the record supporting a post-May 2015 earning power sufficient to qualify Carty as non-disabled is essentially nonexistent. The court, therefore, sees no reason to remand Carty's case to MetLife a second time, to perform what would be a fundamentally non-medical analysis on an issue where MetLife itself admits the evidence already in the record is sufficiently complete to allow one to draw a conclusion. The court will order MetLife to restore Carty's benefits and pay the past benefits to which he was entitled.
Carty has also requested attorney's fees in his motion, but he has not briefed the issue with the level of detail that would be necessary for the court to consider it. That request, therefore, will be denied without prejudice to a future motion. --------
For the foregoing reasons, Carty's motion (Docket No. 106) will be granted, and the defendants' motions (Docket Nos. 103, 105) will be denied. MetLife will be ordered to pay Carty benefits retroactive to its initial discontinuation of payments and to treat Carty as currently eligible for benefits.
An appropriate order will enter.
ENTER this 5th day of March 2019.
ALETA A. TRAUGER
United States District Judge