To be cited only in accordance with Fed. R. App. P. 32.1 Before FRANK H. EASTERBROOK, Circuit Judge ILANA DIAMOND ROVNER, Circuit Judge AMY C. BARRETT, Circuit Judge Appeal from the United States District Court for the Eastern District of Wisconsin. No. 17-cv-1489 David E. Jones, Magistrate Judge.
Kurt Marquardt challenges the denial of his application for disability benefits. Marquardt contends that, in assessing his capacity for work, an administrative law judge improperly discredited medical opinions issued after he was last insured in 2013. These opinions diagnosed him with a longstanding cognitive impairment that limits him to jobs involving one task at a time, breaks, and no distractions. Because the judge did not adequately justify discounting these opinions, we vacate the judgment.
Marquardt was diagnosed in 2012 with systemic lupus erythematosus, a chronic disease that causes the immune system to attack virtually any organ, producing diverse symptoms. Treatment focuses on managing symptoms, as there is no cure. Lupus, MAYO CLINIC, https://www.mayoclinic.org/diseases-conditions/lupus/symptoms-causes/syc-20365789?p=1 (last visited Mar. 6, 2020). After his diagnosis, Marquardt's doctors treated him for various physical ailments and tested him regularly. He reported fatigue and short-term memory loss, which required evaluation. His doctors attributed the fatigue to attention-deficit disorder, for which he received helpful medicine. As for the memory loss, his rheumatologist did not see evidence of "lupus-type cerebritis" (brain inflammation), but wanted "further evaluation," and referred him to a neurologist.
Two reports followed, in 2015 and 2016. The first, in April 2015, came about two years after the date Marquardt was last insured (July 2013). A neurologist wrote that "Marquardt reportedly began to experience gradual-onset, worsening problems with concentration about 7-8 years ago" and concluded that "[i]n the context of his clinical history of cognitive concerns and intact activities of daily living, impressions are consistent with a diagnosis of Mild Cognitive Impairment ...." The impairment caused "Moderate" weaknesses in Marquardt's "mental flexibility." The neurologist also opined that lupus may contribute to his cognitive deficits. To "maximize daily cognitive functioning," the doctor recommended that Marquardt:
Reduce visual and verbal distractions by ensuring that the environment is quiet, workspaces are clear, and clutter is minimized[.]
Focus on one thing at a time[.] Minimize interruptions by turning off the ringer on the phone when trying to focus on a task, and communicate with others that interruptions should be minimized.
Take breaks when completing a long task ...
The neurologist reaffirmed this diagnosis in 2016. She reported that Marquardt still had "Borderline" limitations in mental flexibility, and "severe" limitations in motor speed. The doctor opined that Marquardt's "[c]linical history of cognitive concerns is significant for gradual-onset, worsening problems with concentration beginning about 8-9 years ago in the context of worsening fatigue." She repeated her earlier list of recommendations "to maximize daily cognitive functioning."
In Marquardt's application for benefits, he contended that he became disabled before 2013. He cited his lupus, and lupus-related symptoms—including "short-term memory and cognitive problems" as well as chronic fatigue, rashes, and arthritis. Marquardt also submitted his treatment records and a brain scan from November 2013 that showed areas of brightness known as white-matter hyperintensities—a condition consistent with cognitive deficits. José G. Merino, M.D., White Matter Hyperintensities on Magnetic Resonance Imaging: What Is a Clinician to Do?, 94 MAYO CLINIC PROC. 380 (2019). Marquardt also wrote that although he could do all his own house and yard work, he "often" delayed completing it because of his "fatigue and mental fog," which makes him "often turn circles before getting my task focus back."
After the Commission initially denied benefits (its consulting doctors did not think that the record showed cognitive deficits), Marquardt asked for a hearing and submitted new evidence. In addition to the neuropsychological evaluations from 2015 and 2016, he submitted a report from Angela Dellise, a licensed professional counselor who had treated him for nearly two years, beginning in late 2014. Dellise opined that Marquardt had "marked" limitations in his abilities to "understand ... and [c]arry out simple instructions," to "make judgments on simple work-related decisions," and to "[r]espond appropriately to usual work situations and to changes in a routine work setting." She also stated that "[r]umination keeps him from moving forward with tasks" and that he "gets distracted by details." Dellise saw these behaviors at her first session with Marquardt in 2014; he told her that they arose "much earlier." Marquardt (proceeding pro se) also testified. He repeated that he was able to do housework, but often lost his focus. "I'll start three other projects ... and I just never seem to get anything accomplished" because of the fatigue, he explained. On a normal day, he would take a brisk morning walk then return home "to get as much done as I can before my entire cognitive shuts down, and I lose the rest of my day."
The administrative law judge then heard testimony from a vocational expert. The judge asked the expert whether jobs existed for someone who could "perform simple, routine tasks with simple short instructions, make simple decisions and have few workplace changes" in a setting that was not fast-paced. The judge also added restrictions about physical exertion. She did not, however, ask the expert to restrict the job to one that was "quiet," did not require "multitasking," and granted "breaks" during tasks—as the neurologist had advised. The expert opined that the person the judge described could work as a mail clerk in a non-postal setting, a cashier, or sales attendant. But if the person were off task for 20% of the workday, the worker would likely be fired.
Following the five-step evaluation process, see 20 C.F.R. § 404.1520(a)(4), the judge ruled that Marquardt could perform light work. Marquardt passed step one, as he had not worked regularly since his onset date; his medical condition was severe (step two); but it was not presumptively disabling (step three). At step four, the judge ruled—given the caveats posed to the vocational expert that Marquardt could perform only simple tasks with simple instructions, make simple decisions, and tolerate few workplace changes—that Marquardt could perform light work. The judge did not discuss the 2015 and 2016 neuropsychological assessments (limiting Marquardt to single-tasks jobs with breaks and no distractions) or the 2013 brain scan that suggested cognitive impairment. Nor did she address the vocational expert's opinion that someone who was off task for more than 20% of the day would likely be fired. She gave "little weight" to Dellise's opinion about Marquardt's cognitive abilities because "it was rendered three years after the date last insured, and Ms. Dellise did not see the claimant until more than two years after the date last insured." "Further," the judge continued, "evidence of the claimant's activities of daily living show that his mental limitations are not as severe as assessed by Ms. Dellise." The judge then concluded that Marquardt could do the jobs identified at the hearing (step five). The district court affirmed this ruling.
On appeal, Marquardt argues that the administrative law judge erroneously ignored the two reports from the neurologist. He explains that, even though they were created after the date that he was last insured, they show that he had a cognitive impairment before that date. We agree. An administrative law judge should consider retrospective medical opinions (created after the date last insured) that are consistent with past symptoms. See Bjornson v. Astrue, 671 F.3d 640, 642 (7th Cir. 2012); Estok v. Apfel, 152 F.3d 636, 640 (7th Cir. 1998). A judge need not mention every piece of evidence in the record, Denton v. Astrue, 596 F.3d 419, 425 (7th Cir. 2010), but she "must confront the evidence that does not support her conclusion and explain why that evidence was rejected." Moore v. Colvin, 743 F.3d 1118, 1123 (7th Cir. 2014). The two neuropsychological assessments from 2015 and 2016 meet this standard. They opine that, for several years before he was last insured, Marquardt had feeble concentration, weak "mental flexibility," and frequent fatigue. In order to function, they concluded, he needed to avoid "multitasking," receive "breaks," and have a "quiet" place. These reports are also consistent with the brain scan. The scan was taken just a few months after the last insured date; we have no reason to think that it depicts something new.
In addition, the two reasons that the judge gave for discounting Dellise's opinion are unsound. The judge first explained that the opinion "was rendered three years after the date last insured." But as mentioned above, that is not a valid basis for discounting a medical opinion that is consistent with evidence of disability during the insured period. See Estok, 152 F.3d at 640. And Dellise's opinion that Marquardt's "[r]umination keeps him from moving forward with tasks" and that he "gets distracted" is consistent with Marquardt's reports about his "fatigue and mental fog" during the insured period. The second reason for discounting the opinion—that the "evidence of the claimant's activities of daily living show that his mental limitations are not as severe as [Dellise] assessed"—is also flawed. We have cautioned against reading too much into a claimant's activities of daily living because people have more flexibility in scheduling and executing them than they do for the activities of a full-time job. See, e.g., Bjornson, 671 F.3d at 647. And the neuropsychological reports specifically accounted for these activities in concluding that Marquardt suffered from a mild cognitive limitation.
We have considered the additional argument that the ALJ should have incorporated the limitations that the neurologist identified (jobs should avoid "multitasking" but include "breaks" and take place in a "quiet" setting) into the hypothetical posed to the vocational expert. On remand, the ALJ may consider these limitations, which are set forth in the 2015 and 2016 reports, when posing a new hypothetical to a vocational expert.
Therefore, we VACATE the judgment and REMAND for further proceedings.