George Meuser v. Carolyn Colvin

• Court: Court of Appeals for the Seventh Circuit
• Decided: October 3, 2016
• Docket: 16-1052
• Status: Published

Opinion

In the

United States Court of Appeals For the Seventh Circuit ____________________

No. 16‐1052 GEORGE B. MEUSER, Plaintiff‐Appellant,

v.

CAROLYN W. COLVIN, Acting Commissioner of Social Security, Defendant‐Appellee. ____________________

Appeal from the United States District Court for the Southern District of Indiana, Evansville Division. No. 3:15‐cv‐32 — William G. Hussmann, Jr., Magistrate Judge. ____________________

ARGUED JULY 7, 2016 — DECIDED OCTOBER 3, 2016 ____________________

Before WOOD, Chief Judge, and BAUER and KANNE, Circuit Judges. PER CURIAM. George Meuser suffers from schizophrenia and applied for Disability Insurance Benefits principally be‐ cause of that impairment. But an administrative law judge concluded at Step 2 of the 5‐step disability analysis that Meuser’s schizophrenia was not a severe impairment and

2 No. 16‐1052

denied benefits on that basis. A magistrate judge presiding by consent, see 28 U.S.C. § 636(c), upheld that ALJ’s decision, but Meuser argues that it rests on a profound misunder‐ standing of the medical evidence and thus is not supported by substantial evidence. We agree. I. BACKGROUND Meuser, who is 46 years old, was diagnosed with schizo‐ phrenia in 1996, and for 15 years he managed his symptoms with the antipsychotic drug Zyprexa. Throughout that time, from 1995 through March 2012, Meuser worked in the mail‐ room at a publishing company. Although he described him‐ self as “socially impaired somewhat,” he generally got along well with his coworkers. But Meuser’s health began deteriorating in late 2011 after his pharmacist gave him the generic version of Zyprexa. Be‐ fore that switch, as evidenced by nearly two years of physi‐ cians’ progress notes, Meuser was sleeping eight to ten hours per night, was not experiencing side effects from his medica‐ tion, and was showing “good” and “euthymic” mood and affect, meaning neither elevated nor depressed, see DORLAND’S MEDICAL DICTIONARY 655 (32d ed. 2012). But after the drug substitution, Meuser started having insomnia and would sleep only four to six hours per night. He said that he could not focus at work. He told clinicians that an increased work load over the preceding six or seven months was causing him stress. Hoping that a break would improve his symptoms, Meuser took a leave of absence from his job in December 2011. He apparently was living with his parents at the time, and he continues to live with them.

No. 16‐1052 3

Meuser changed psychiatrists in January 2012. At that time he began seeing Dr. Charles Rhoton, who rediagnosed Meuser’s schizophrenia from “undifferentiated” to “para‐ noid type,” which involves “prominent delusions or audito‐ ry hallucinations,” see AM. PSYCHIATRIC ASS’N, DIAGNOSTIC AND STATISTICAL MANUAL OF MENTAL DISORDERS 313 (4th ed. text revision 2000) (DSM‐IV‐TR). Dr. Rhoton also noted “[p]roblems related to social environment” and assessed Meuser’s Global Assessment of Functioning (GAF) score1 as 61 to 70 with “mild symptoms.” Dr. Rhoton switched Meuser back to the brand‐name Zyprexa he was taking pre‐ viously and increased his dosage, but Meuser did not fully improve. Two weeks after Dr. Rhoton’s initial assessment, Meuser reported that the increased dosage was causing him to sleep at least 12 to 14 hours a night. Meuser started taking a lower dosage, which helped, but still his sleep was “errat‐ ic.” Dr. Rhoton noted that Meuser’s mood was “mildly dys‐ thymic,” or mildly depressed, and his affect was “blunted,” meaning that Meuser exhibited a severe reduction in the in‐ tensity of his external expression of emotion, see DORLAND’S MEDICAL DICTIONARY at 582, 655. Dr. Rhoton now revised his diagnosis to be schizophrenia, undifferentiated type. The next month Meuser reported to Dr. Rhoton that he was “doing pretty well,” had more energy, was falling asleep easier, and had “been getting out of the house a little more.” Dr. Rhoton noted that Meuser was responding well to his medication and that his mood and affect were normal. Nev‐

1 The American Psychiatric Association abandoned reliance on GAF

scores in 2013 with the publication of the fifth edition of its Diagnostic and Statistical Manual of Mental Disorders. See Williams v. Colvin, 757 F.3d 610, 613 (7th Cir. 2014).

4 No. 16‐1052

ertheless, Meuser said, he was not feeling well enough to re‐ turn to work. Faced with the choice of returning to work or being fired, Meuser decided to quit his job. At his next ap‐ pointment with Dr. Rhoton at the end of March 2012, Meuser said he was doing “okay,” and Dr. Rhoton assessed his mood as dysthymic. Meuser had applied for DIB on February 27, 2012, and based on medical records through March 2012, psychologist F. Kladder, an agency consultant, opined that Meuser’s schizophrenia was not a severe impairment. He noted that Meuser’s symptoms had been well controlled for over 15 years and that, although his symptoms were not well con‐ trolled with generic medication, Meuser was again “doing well’ after switching back to Zyprexa. Dr. Kladder checked boxes indicating only mild difficulties in activities of daily living, social functioning, and concentration, persistence, and pace. A month later William Shipley, an agency consult‐ ant with a Ph.D. in an unspecified field, agreed with Dr. Kladder’s assessment without explanation. Meuser’s application for benefits then was denied initial‐ ly and on reconsideration. For the next year and a half, until his hearing before the ALJ, Meuser visited Dr. Rhoton six times as his symptoms waxed and waned. Although at each visit Meuser reported to Dr. Rhoton that he was “well,” “sta‐ ble,” “fairly well,” or “okay,” Dr. Rhoton noted during four of those visits that Meuser was exhibiting a blunted affect, and on a fifth visit that he was dysthymic. Meuser continued to complain about erratic sleep. In November 2012, for ex‐ ample, Meuser reported that sleep was “an issue” but said he did not want to change his medication. Then in February 2013 he reported sleeping only three or four hours some

No. 16‐1052 5

nights and other nights twelve to fourteen hours. And as late as August 2013 he reported sleeping “poorly” and having difficulty “falling asleep and staying asleep.” But Meuser did not report “psychotic symptoms” or “positive symp‐ toms” such as hallucinations, paranoia, or “thought broad‐ casting/insertion,” see AM. PSYCHIATRIC ASS’N, DIAGNOSTIC AND STATISTICAL MANUAL OF MENTAL DISORDERS 94 (5th ed. 2013) (DSM‐V). At Meuser’s last appointment before the Sep‐ tember 2013 hearing, Dr. Rhoton noted that his “negative symptoms” (i.e., limited emotional expression and disinter‐ est in work or social activities) “remain prominent,” see DSM‐V at 88. A week after this appointment, Dr. Rhoton completed a mental residual functional capacity assessment, marking boxes for the degree of Meuser’s limitations in various cate‐ gories. Dr. Rhoton opined that Meuser, in addition to being “moderately” limited in many areas, was “markedly” lim‐ ited in his abilities to (1) maintain attention and concentra‐ tion for extended periods, (2) regularly attend scheduled ac‐ tivities and be punctual, (3) “sustain an ordinary routine without special supervision,” (4) “make simple work‐related decisions,” (5) “ask simple questions or request assistance,” (6) accept instructions and criticism, and (7) set realistic goals and make plans. Dr. Rhoton further opined that Meuser was experiencing “extreme” limitations in the abili‐ ties to (1) “work in coordination with or proximity to others without being distracted,” (2) “complete a normal workday and workweek without interruptions from psychologically based symptoms and … perform at a consistent pace with‐ out an unreasonable number and length of rest periods,” and (3) “respond appropriately to changes in the work set‐

6 No. 16‐1052

ting.” Dr. Rhoton wrote that these limitations had been pre‐ sent for over a year and a half, since February 2012. Prior to the hearing Meuser submitted reports completed by himself and his parents describing his functional limita‐ tions. Meuser’s mother described how Meuser’s sleep “is a major problem.” She reported that he “only sleeps a couple of hours at a time,” which “causes great anxiety.” She said that Meuser’s chores included doing laundry, taking out the trash, caring for their two cats, and doing “odd jobs as need‐ ed around the house.” Weekly he also prepares meals such as canned pasta or soup, frozen pizza, burgers, and cereal. She said that he “only goes out to eat or out in public about every two weeks when he sees his psychiatrist” and, even then, needs someone with him. And, his mother noted, Meuser had “become extremely reclusive” and “more ob‐ sessed with TV and with NASCAR” since his symptoms re‐ turned. She said that Meuser respects authority and “gets along with others when with them, but hardly ever goes near them.” Meuser’s father echoed these observations, say‐ ing that Meuser spends no time with others and has become more withdrawn. In Meuser’s own report, he recounted that his sleep is “not consistent,” and even if he does get a lot of sleep, he still feels tired. Meuser also reported difficulty with his short‐term memory and concentration. Meuser then testified about his schizophrenia at the hear‐ ing before the ALJ in September 2013. He said that he cannot sleep more than three to five hours a night. This sleep depri‐ vation left him so tired that he “couldn’t function” at work, and after quitting his job, Meuser said, his condition had not improved. The lack of sleep also affects his short‐term memory, Meuser explained, and thus he must write remind‐

No. 16‐1052 7

ers to himself or rely on his parents to help him remember things like doctor’s appointments. He also said that his par‐ ents help him with the laundry because he forgets clothes in the dryer. He said he would “really be in a bad spot” with‐ out his parents. Meuser described his daily activities as do‐ ing house chores, doing laundry, and “sometimes” cooking. He said he could drive but had not ventured beyond Hen‐ derson, Kentucky, in the previous year (about 37 miles from his home in Princeton, Indiana). He stays “at home most of the time,” said Meuser, and sometimes does not leave the house for days. He testified that he had become more social‐ ly withdrawn, and that he had no close friends, only ac‐ quaintances. And when asked about his reaction to social situations, Meuser said: “I just don’t even want to be there. It’s not anything specific with nervousness, or anything like that. It’s just, shut it down, don’t even go. Don’t even think about it.” The ALJ, in finding at Step 2 that Meuser’s schizophrenia was not a severe impairment, asserted that “[s]igns and find‐ ings upon objective examination simply do not reflect the presence of a severe physical or mental impairment that more than minimally affects claimant’s ability to perform basic work functions.” The ALJ reasoned, first, that Meuser had been “very stable” on Zyprexa for 14 years, with no side effects from the medication and GAF scores of 61 to 70. Sec‐ ond, after acknowledging what he described as “some com‐ plaints” from Meuser about his sleep while taking the gener‐ ic form of Zyprexa, the ALJ implied (without saying directly) that Meuser had gotten better after switching back to the brand‐name drug. As support for this apparent conclusion that Zyprexa effectively had controlled Meuser’s symptoms, the ALJ asserted that “[t]hroughout the record” Meuser had

8 No. 16‐1052

“denied any psychotic symptoms” and in February 2013 had reported having “no hallucinations or other thought prob‐ lems.” The ALJ suggested that, although Meuser had report‐ ed “some” problems with concentration and sleep in August 2013, his condition was not severe at the time because “his schizophrenia was negative and he was doing okay.” The ALJ gave “little weight” to Dr. Rhoton’s assessment that Meuser had a number of marked or extreme limitations be‐ cause, the ALJ said, the assessment was “completely in con‐ trast” with the psychiatrist’s treatment record. According to the ALJ, Meuser himself had “reported to Dr. Rhoton he was doing okay with negative symptoms remaining prominent, no positive symptoms with Zyprexa.” The ALJ asserted that medical records from after Meuser’s onset date show “essen‐ tially normal findings” by his treatment providers and “did not support the presence of a severe mental impairment.” Instead, the ALJ declared, those records evidence only mild limitations, and thus the ALJ gave “significant weight” to the agency consultants’ corresponding views. And having con‐ cluded that Meuser did not satisfy the requirement of a se‐ vere impairment at Step 2, the ALJ denied benefits without addressing the three remaining steps, see 20 C.F.R. § 404.1520(a)(4)(ii). The Appeals Council denied review, thus making the ALJ’s ruling the final decision of the Commissioner of Social Security. See Liskowitz v. Astrue, 559 F.3d 736, 739 (7th Cir. 2009). II. ANALYSIS We will uphold an ALJ’s decision if it is supported by substantial evidence, but that standard is not satisfied unless

No. 16‐1052 9

the ALJ has adequately supported his conclusions. See Jelinek v. Astrue, 662 F.3d 805, 811 (7th Cir. 2011). On appeal, Meuser first argues that a diagnosis of schiz‐ ophrenia is enough, standing alone, to satisfy the “severe impairment” requirement at Step 2 because that diagnosis, by definition, requires “marked” social or occupational dys‐ function, see DSM‐V at 99. We have difficulty imagining how an uncontested diagnosis of schizophrenia (which describes Meuser’s situation) could not survive Step 2. An impairment is “not severe” only if it is “a slight abnormality” that has “no more than a minimal effect on the ability to do basic work activities,” SSR 96‐3p, 1996 WL 374181, at *1 (July 2, 1996), such as “[u]nderstanding, carrying out, and remem‐ bering simple instructions,” “[r]esponding appropriately” to supervisors and co‐workers, and “[d]ealing with changes in a routine work setting,” 20 C.F.R. § 404.1521. We recently emphasized in rejecting an ALJ’s determination at Step 2 that a claimant’s impairments were not severe, “[t]he Step 2 de‐ termination is ‘a de minimis screening for groundless claims.’” O’Connor‐Spinner v. Colvin, 2016 WL 4197915, at *6 (7th Cir. Aug. 9, 2016) (quoting Thomas v. Colvin, 826 F.3d 953, 960 (7th Cir. 2016)); see Newell v. Comm’r of Soc. Sec., 347 F.3d 541, 546 (3d Cir. 2003); Smolen v. Chater, 80 F.3d 1273, 1290 (9th Cir. 1996); McDonald v. Sec’y of Health & Human Servs., 795 F.2d 1118, 1122 (1st Cir. 1986). Essentially, the ALJ has conflated Steps 2, 4, and 5. Alt‐ hough the ALJ could still find that Meuser, with medication, can perform work and is not disabled, an assessment of the functional limitations caused by an impairment is more ap‐ propriate for Steps 4 and 5, not Step 2. See, e.g., Spiva v. Astrue, 628 F.3d 346, 348 (7th Cir. 2010) (schizophrenia severe

10 No. 16‐1052

but not disabling at Step 4); Milcanovic v. Colvin, 572 F. App’x 587, 590–91 (10th Cir. 2014) (schizophrenia severe but not disabling at Step 5); Ferguson v. Comm’r of Soc. Sec., 628 F.3d 269, 271 (6th Cir. 2010) (schizophrenia severe but not disa‐ bling at Step 4); Sultan v. Barnhart, 368 F.3d 857, 861–62 (8th Cir. 2004) (schizophrenia severe but not disabling at Step 5); Matlock v. Barnhart, 90 F. App’x 208, 209–10 (9th Cir. 2004) (schizophrenia severe but not disabling at Step 4); Moore v. Sullivan, 895 F.2d 1065, 1068 (5th Cir. 1990) (schizo‐ phrenia severe but not disabling at Step 4). But see Bunch v. Heckler, 778 F.2d 396, 400–01 (7th Cir. 1985) (upholding an ALJ’s Step 2 determination that claimant’s schizophrenia was not severe despite hospitalization due to “bizarre and with‐ drawn behavior,” reports of “auditory hallucinations and … frequent, intrusive and disturbing thoughts,” and a period of only three months in which she had not “heard voices”). Meuser’s argument that his diagnosis alone is enough to sat‐ isfy the severity requirement is similar to an issue we recent‐ ly considered in O’Connor‐Spinner v. Colvin, 2016 WL 4197915 (7th Cir. Aug. 9, 2016), concerning a diagnosis of major de‐ pression. In that case we rejected the ALJ’s determination that the claimant’s major depression was not a severe im‐ pairment at Step 2, noting that the determination “strikes us as nonsensical given that the diagnosis, by definition, reflects a practitioner’s assessment that the patient suffers from ‘clin‐ ically significant distress or impairment in social, occupa‐ tional, or other important areas of functioning.’” Id. at *6 (quoting DSM‐IV‐TR at 356). We need not resolve this issue, however, because the ALJ made other errors that alone warrant a conclusion that the ALJ’s decision is not supported by substantial evidence. First, Meuser argues, correctly, that the ALJ fundamentally

No. 16‐1052 11

misunderstood the diagnosis and symptoms of schizophre‐ nia by focusing solely on the “positive symptoms,” such as hallucinations, and thinking that “negative symptoms” meant no symptoms. For example, the ALJ reasoned that Meuser’s description of his limitations is not fully credible in part because, although Meuser had problems with sleep, he “had no hallucinations or other thought problems” and “his schizophrenia was negative and he was doing okay” (empha‐ sis added). The ALJ’s mistaken understanding of the medical evidence led him to conclude that Meuser’s mental status was “essentially normal” when, in fact, Dr. Rhoton’s finding that Meuser’s “[n]egative symptoms remain prominent” meant just the opposite—that his emotional expression was impaired and he showed disinterest in work or social activi‐ ties. Thus, the ALJ improperly played doctor when he ig‐ nored expert opinions to arrive at his own, incorrect, inter‐ pretation of the medical evidence. See Stage v. Colvin, 812 F.3d 1121, 1125 (7th Cir. 2016); Moon v. Colvin, 763 F.3d 718, 722 (7th Cir. 2014). The Commissioner does not explicitly deny that the ALJ misunderstood the meaning of “negative symptoms” but contends that any misunderstanding is harmless because the reference to “negative symptoms” in the medical record was merely a “single notation” on which the ALJ did not place undue weight. To the contrary, the ALJ clearly misunder‐ stood the medical evidence and repeatedly relied on the lack of “positive” symptoms and the reference to negative symp‐ toms to conclude that Meuser’s schizophrenia was not se‐ vere. The Commissioner further argues that any error with respect to this issue was harmless because Meuser did not show that any limitations lasted for 12 months as required to move on to Step 3, see 20 C.F.R. § 404.1509. But this duration

12 No. 16‐1052

requirement is not part of the definition of “severe.” See id. § 404.1520. Regardless, despite a perfunctory reference to the duration requirement, the ALJ did not rely on this rationale in his opinion, so the Commissioner cannot now rely on it. See SEC v. Chenery Corp., 318 U.S. 80, 87–88 (1943); Kastner v. Astrue, 697 F.3d 642, 648 (7th Cir. 2012). The ALJ’s misunderstanding of the symptoms of schizo‐ phrenia was compounded, Meuser argues, when the ALJ ig‐ nored several notations that Meuser had a blunted or flat af‐ fect and a dysthymic mood, which are negative symptoms of schizophrenia, see DSM‐V at 88. We agree. An ALJ cannot recite only the evidence that supports his conclusion while ignoring contrary evidence. See Moore, 743 F.3d at 1124; Bates v. Colvin, 736 F.3d 1093, 1099 (7th Cir. 2013). This “cherry‐ picking” is especially problematic where mental illness is at issue, for “a person who suffers from a mental illness will have better days and worse days, so a snapshot of any single moment says little about [his] overall condition.” Punzio v. Astrue, 630 F.3d 704, 710 (7th Cir. 2011). Here the ALJ ig‐ nored all six of Dr. Rhoton’s notations that Meuser exhibited a blunted affect or a dysthymic mood from February 2012 through August 2013 (a period, incidentally, much longer than 12 months). Next, Meuser argues that the ALJ incorrectly rejected Dr. Rhoton’s opinion that he was markedly or extremely lim‐ ited in several areas of functioning. Because Dr. Rhoton was Meuser’s treating psychiatrist, his opinion was entitled to controlling weight if “well‐supported and not inconsistent with other substantial evidence.” Stage, 812 F.3d at 1126; see Bauer v. Astrue, 532 F.3d 606, 608 (7th Cir. 2008). The ALJ rejected Dr. Rhoton’s opinion because it did not align with

No. 16‐1052 13

his own, incorrect interpretation of the medical evidence; as such the ALJ’s refusal to give controlling weight to the opin‐ ion is not supported by substantial evidence. Moreover, even assuming that there had been a reason to deny controlling weight to Dr. Rhoton’s opinion, the ALJ was “not permitted simply to discard it.” Scrogham v. Colvin, 765 F.3d 685, 697 (7th Cir. 2014). Rather, the ALJ was re‐ quired to explicitly consider the details of the treatment rela‐ tionship and explain the weight he was giving the opinion. See 20 C.F.R. § 404.1527; Yurt v. Colvin, 758 F.3d 850, 860 (7th Cir. 2014). He did not. The Commissioner contends that the ALJ “properly weighed” Dr. Rhoton’s opinion “under the regulatory factors.” This contention is frivolous. The ALJ did not mention any “regulatory factors” when evaluating Dr. Rhoton’s opinion; the ALJ said only that the opinion mer‐ ited “little weight.” Looking at the factors, this conclusion is not supportable. Dr. Rhoton is a psychiatrist and for a year and a half was personally responsible for treating Meuser’s schizophrenia. See 20 C.F.R. § 404.1527(c) (explaining that more weight is given to opinions by specialists, to examining and treating relationships, to longer treatment relationships where there are frequent examinations); Bauer, 532 F.3d at 608. The ALJ instead gave “significant weight” to the agency consultants’ opinions, but these consultants did not examine Meuser, they are not psychiatrists (though one was a psychologist), and they had reviewed only a fraction of Meuser’s treatment records that were available before Meuser submitted additional evidence. See Goins v. Colvin, 764 F.3d 677, 680 (7th Cir. 2014) (criticizing ALJ’s reliance on consulting physicians’ conclusions that were based on in‐ complete medical record); Campbell v. Astrue, 627 F.3d 299,

14 No. 16‐1052

309 (7th Cir. 2010) (same). This error, too, is reason enough to remand the case. See Bauer, 532 F.3d at 608–09. The preceding errors are sufficiently serious to warrant remanding the case, but we note additional issues that the ALJ must address on remand. Meuser argues that the ALJ’s misunderstanding of the medical evidence led to a flawed finding that Meuser’s testimony about the severity of his schizophrenia was “not fully credible.” The Commissioner contends that Meuser forfeited any challenge to the ALJ’s credibility finding by not raising it in the district court. But the credibility finding rests entirely on the ALJ’s misunder‐ standing of the medical evidence. That foundation has col‐ lapsed, and with it the adverse credibility assessment. There was no reason for Meuser to belabor the point by stating the obvious. Meuser further contends that the ALJ overstated both the effectiveness of his medication and the extent of his daily ac‐ tivities. Regarding the effectiveness of the medication, it is true that, about a month after switching back to Zyprexa, Meuser reported that he was “doing pretty well,” had more energy, was sleeping better, and was getting out of the house more. But “[t]here can be a great distance between a patient who responds to treatment and one who is able to enter the workforce.” Scott v. Astrue, 647 F.3d 734, 739–40 (7th Cir. 2011). The ALJ acknowledged that Meuser reported “some” problems with sleep; in fact, Meuser consistently com‐ plained about his sleep for over a year and a half, reporting as late as February 2013 that he was sleeping only three or four hours some nights and twelve to fourteen hours other nights. The ALJ also emphasized that Meuser was “very sta‐ ble” on Zyprexa for 14 years, but this evidence was from be‐

No. 16‐1052 15

fore the alleged onset date and therefore irrelevant. See Wiese v. Astrue, 552 F.3d 728, 731 (8th Cir. 2009). And regarding the extent of Meuser’s daily activities, the ALJ determined that Meuser was only mildly limited in ac‐ tivities of daily living and in social functioning. But “[a]n ALJ cannot disregard a claimant’s limitations in performing” daily activities. Moss v. Astrue, 555 F.3d 556, 562 (7th Cir. 2009); see Craft v. Astrue, 539 F.3d 668, 680 (7th Cir. 2008). When assessing Meuser’s activities of daily living, the ALJ relied on Meuser’s ability to, for example, do laundry, cook meals, and complete chores, but the ALJ ignored evidence that Meuser gets help with the laundry from his parents when they retrieve the clothes he leaves in the dryer, that the meals he cooks are simple and prepackaged, and that he rarely leaves the house. Meuser, after all, is not claiming that he is disabled by physical limitations, so the proper focus— ignored by the ALJ—is the effect of Meuser’s schizophrenia on his mental functioning, including his abilities to concen‐ trate, perform work tasks without constant supervision, maintain a consistent pace, and work around and with other persons. Whether or not he can pop a frozen dinner into the microwave or occasionally clean the litterbox is irrelevant. Moreover, the ALJ asserted that Meuser was only mildly lim‐ ited in social functioning in part because he was able to spend time with his parents and he got along well with coworkers when he did work. But the ALJ disregarded evi‐ dence that Meuser was “extremely reclusive,” that he seemed to spend time only with his parents, and that his on‐ ly interactions with coworkers were before his alleged onset date.

16 No. 16‐1052

III. CONCLUSION Because the ALJ misunderstood the medical evidence and improperly rejected the treating psychiatrist’s opinion, the ALJ’s conclusion that Meuser did not have a severe im‐ pairment is not supported by substantial evidence. We there‐ fore REVERSE the judgment of the district court and REMAND the matter to the agency for further proceedings.