Joshua Ray LANIGAN, Plaintiff-Appellant, v. Nancy A. BERRYHILL, Acting Commissioner of Social Security, Defendant-Appellee, 865 F.3d 558

SYKES, Circuit Judge. Joshua Lanigan injured his back at his job in 2009. That same year he hurt his neck in a ear accident, and in 2011 he was diagnosed with diabetes. Since then his medical impairments have been complicated by mental illness. Lanigan applied for Supplemental Security Income and Disability Insurance Benefits in March 2012 when he was 38 years old. An Administrative Law Judge (“ALJ”) found his physical and mental impairments to be severe but not disabling and denied benefits. The Appeals Council denied review, and the district court upheld the ALJ’s decision. See 42 U.S.C. § 405(g).

We conclude that the case must be returned to the agency for further proceedings because the ALJ misinformed a vocational expert about Lani-gan’s residual functional capacity, thus undermining the expert’s testimony that Lanigan could engage in competitive employment. I. Background Lanigan asserts that he became disabled in May 2009 after injuring his lower back while working as a general laborer. He received worker’s compensation for that injury and tried various forms of physical therapy, but none proved to be effective. Eventually he was diagnosed with degenerative disc disease in his lower back.

In 2009 he also injured his neck in a car crash. Then in December 2011 he was diagnosed with Type 2 diabetes. At one time Lanigan was a body builder, but since 2009 he has been physically inactive and bounced between part-time or seasonal jobs and periods of unemployment. The Commissioner of Social Security concedes that Lanigan’s physical impairments, by themselves, would limit him to light work, so our focus is on his mental illness and its effect on his ability to engage in competitive employment.

After Lanigan learned of his diabetes diagnosis, he fell into a state of depression, anxiety, and suicidal preoccupation. He explained to a psychiatrist that he has an “ingrained fear of diabetes” because as a child he watched his grandfather struggle with and eventually die from the disease.

In January 2012, a month after receiving the diagnosis, Lanigan reported to his physician that he had been in a “very low mood,” rarely leaving his apartment, and “sitting for hours at a time with a loaded pistol on his lap.” The following year after a psychiatric evaluation, Lanigan’s treating psychiatrist documented a history of alcohol abuse, bipolar affective disorder, major depression, attention deficit/hyperactivity disorder, oppositional defiant disorder, and kleptomania.

The psychiatrist also noted that Lanigan had complained of intermittent episodes of visual hallucinations in which he saw animals or people in his periphery and five- to ten-minute episodes of palpitations, sweating, and tremors when in public places or in the midst of family members. Lanigan attributes the episodes to his belief that he is being stigmatized because he is mentally ill. In addition to the stress of being diagnosed with diabetes, Lanigan’s inability to maintain full-time work has also contributed to his anxiety.

He and his "wife divorced in 2009, and in 2012 he moved in with his mother and stepfather after being evicted from his apartment.

In the months following that move, Lanigan complained to his psychiatrist about the added stress of living with his verbally abusive stepfather, prompting the doctor to opine that the living arrangement had “been tough on him” and that “a lot of his pain and stress level would be improved ultimately if he can acquire the finances to get his own living quarters back.” In June 2012, three months after applying for benefits, Lanigan was examined by a state-agency psychologist who concluded that he suffers from severe affective and anxiety disorders.

Lanigan told the psychologist that he “can only pay attention for a few minutes” and is limited in his ■ability to “complete tasks, concentrate, understand, follow instructions, and get along [with] others.” Based on her examination, the psychologist concluded that Lanigan’s mental impairments could cause moderate limitations in his ability to (1) understand, remember, and carry out detailed instructions; (2) maintain attention and concentration for extended periods; (3) perform activities within a schedule, maintain regular attendance, and be punctual; and (4) work in coordination with or in proximity to others without being distracted by them'.

A second state-agency psychologist who reviewed Lanigan’s file several months later agreed with the first doctor about the limitations caused by Lanigan’s mental illness. One feature of Lanigan’s mental illness is unmentioned in the evaluations of the state-agency psychologists—apparently because the problem did not arise until 2013. That year Lanigan began experiencing recurring blackout episodes. In June he was hospitalized on an involuntary, emergency basis after being arrested for shooting out windows with a BB gun. At the hospital he claimed he could not recall his actions.

When Lanigan was admitted, a doctor noted that he had attempted suicide four times in the prior six months and scored his global assessment of functioning (commonly known as “GAF”) at 30 to 35. In the discharge

summary, another doctor noted that Lanigan had been experiencing mood swings in which he would feel okay and then suddenly become very irritable and express fear that he was “going to hurt people.” Lanigan had told staff he was not having homicidal thoughts while hospitalized but, nevertheless, said he worried about his irritable episodes and thus avoided people and isolated himself. At an outpatient visit a week after Lanigan’s release from the hospital, a psychiatric nurse noted that he possibly suffers from a dissociative disorder.

At his hearing before the ALJ in December 2013, Lanigan testified about his mental-health limitations. He described having difficulty in social situations and said he cannot go to a grocery store without first taking medication .or having someone with him to help in case of a panic attack.

He also briefly testified about his stay in the hospital in July 2013 and other blackout episodes when he would “not remember what I was doing.” He said, “I’ve had episodes where [I’m] sitting down at dinner with my family, they start fighting, it’s 6:00 in the evening and next thing I know I’m sitting in my room and it’s 9:45 and I don’t remember anything.” When asked about his current employment, Lanigan testified that as of the date of the hearing, he had been working part-time at Michaels craft store for five to six weeks on the truck crew—unloading the truck, stocking the store, and setting up displays.

He described his limited interactions with customers as “very stressful” and said he must have another employee present with him when he encounters customers. Lanigan testified that even his five-hour shifts three or four days a week are hard but said his coworkers “are really nice” and make him “feel comfortable.” He also explained that he excuses himself three to five times during each shift, sometimes for up to 20 minutes, so that he can retreat to the bathroom and get his emotions “in check.” He said his boss knows his situation “so she’s tolerant of it.” A vocational expert also testified.

The ALJ asked him to assess whether competitive employment would be available to a person with the following hypothetical residual functional capacity (“RFC”): capable of performing low-stress jobs constituting light work so long as those jobs involve only routine tasks; does not require more than occasional interaction with coworkers or the public; does not involve piece work or a rapid assembly line; is limited to occasional stooping, crouching, kneeling, or crawling; and can be off task up to 10% of the workday in addition to regularly scheduled breaks. The ALJ did not explain the source of the 10% figure.

The vocational expert opined that a person with this RFC could not operate a buffing machine or work as a general laborer as Lanigan once had done. But, the vocational expert continued, a person with the RFC described by the ALJ could work as a hand packer, machine operator, or factory inspector.

The vocational expert acknowledged, however, that this hypothetical person would be unemployable if any of the following is true: he will be off task more than 10% of the workday, he must leave his work station and walk around when off task, he cannot interact at all with coworkers or the public, or he will miss work more than twice per month. The ALJ applied the five-step analysis for assessing disability, see 20 C.F.R. §§ 404.1520(a), 416.920(a), and concluded that Lanigan was not disabled.

At step 1 the ALJ determined that Lanigan had not engaged in substantial gainful activity since his alleged onset in May 2009. At step 2 the ALJ identified Lanigan’s severe impairments as “degenerative disc disease ..., obesity, affective disorder, and anxiety disorder.” And at step 3 the ALJ concluded that these impairments, individually or in combination, do not satisfy a listing for presumptive disability.

The ALJ concluded that Lanigan’s mental impairments do not cause two or more marked limitations or one such limitation coupled with repeated episodes of decompensation, and thus the paragraph B criteria of listings 12.04 and 12.06 were not satisfied. The ALJ acknowledged that Lanigan’s mental impairments do cause moderate restriction in his activities of daily living and moderate difficulty in social functioning.

The ALJ reasoned, however, that Lanigan must not have any marked limitations because he was working part-time in a retail environment, cares for his cat and dog, prepares meals daily, vacuums, helps with laundry, drives,. shops in stores, manages money, and sees family regularly. In assessing Lanigan’s RFC, the ALJ found Lanigan’s testimony to be “generally credible” but rejected his statements concerning the intensity, persistence, and limiting effects of his mental illness.

The ALJ reasoned that although Lanigan described severe problems being in public, he “actually works in a retail store[ ] about 15-22 hours a Week” and has sustained other work that would require some social interaction “with no evidence that the work activity caused his symptoms to flare-up or made him need emergency treatment.” The ALJ also noted that while the record provides “some support” for Lanigan’s testimony about his social limitations and anxiety in public, Brian Eggener, a treating psychiatrist, had assigned a GAF score of 65 to 70 in March 2012.

This means, the ALJ posited, that Lanigan has only mild symptoms and limitations. And even when Lanigan was having suicidal ideations, the ALJ noted, his GAF score was 71 upon discharge from the hospital. Moreover, the ALJ asserted, the physician who approved Lanigan’s release said he was not showing signs of depression or anxiety.

The ALJ commented that the record lacks “opinions from treating or examining physicians indicating that the claimant is disabled or even has limitations greater than those determined in this decision,” and said he placed “great weight on the opinions of the State agency medical consultants that the claimant can sustain light to medium work” because those opinions are “consistent with the record as a whole.” The ALJ concluded that Lanigan’s RFC matched the hypothetical given to the vocational expert.

At step 4, then, the ALJ found that Lanigan could not perform his past work but at step 5 concluded that he could work one of the jobs identified by the vocational expert. II. Discussion Because the Appeals Council denied review, the ALJ’s decision is the final word of the Commissioner of Social Security. Scrogham v. Colvin, 765 F.3d 685, 695 (7th Cir. 2014). The district court’s decision is reviewed de novo, so we’re directly reviewing the ALJ’s decision. See Yurt v. Colvin, 758 F.3d 850, 856 (7th Cir. 2014). We will uphold that decision if it is supported by substantial evidence in the. record. Moore v.

Colvin, 743 F.3d 1118, 1120-21 (7th Cir. 2014). On appeal Lanigan argues that the hypothetical RFC posed to the vocational expert—“hypothetical” in name only because the ALJ assigned that very same RFC to Lanigan—is flawed and caused the vocational expert to overstate the available jobs that Lanigan can perform.

Lanigan identifies two flaws: First, the ALJ failed to lay a foundation for certain limitations described in the hypothetical, including that he might be off task up to 10% of the workday (but not more) and that he is able to maintain “frequent” (rather than just “occasional”) contact with coworkers and the public. Second, Lanigan says, the hypothetical failed to account for his moderate limitations in concentration, persis-ténce, and pace. See O’Connor-Spinner v. Astrue, 627 F.3d 614, 619-20 (7th Cir. 2010). We agree with Lanigan that the ALJ’s hypothetical is not supported by substantial evidence.

See Ehrhart v. Sec’y of Health & Human Servs., 969 F.2d 534, 540 (7th Cir. 1992) (stating the rule that a hypothetical question -must be supported by medical evidence in the record). The ALJ instructed the vocational expert to evaluate a hypothetical person who would “be off task up to 10% of the work day, in addition to regularly scheduled breaks.” Significantly, the vocational expert testified that persons who will be off task less than 10% of the workday are capable of maintaining full-time employment.

But those who will be off task more than 10% of the time, the expert acknowledged, will be incapable of maintaining competitive employment and thus are disabled. Lani-gan argues that the ALJ had no basis to conclude that he wouldn’t be off task more than 10% of the time, especially given his unrebutted testimony that he was taking unscheduled breaks (sometimes for 20 minutes) three to five times during his five-hour shifts at Michaels. At that rate it’s likely Lanigan would be off task more than 10% of a typical workday.

The Commissioner responds that the ALJ’s 10% calculation was supported by the state-agency psychologists, who opined that Lanigan demonstrated adequate ability to sustain concentration and had only moderate—not marked—difficulty in various functional areas. But the Commissioner’s position is unpersuasive for two reasons. First, the ALJ made no effort to “build an accurate and logical bridge,” see Beardsley v.

Colvin, 758 F.3d 834, 837 (7th Cir. 2014), between the “no more than 10%” finding and the psychologists’ general assessment that Lanigan exhibits moderate difficulty in areas like the “ability to maintain attention and concentration for extended periods” and the “ability to perform activities within a schedule.” An ALJ need not address every piece of evidence, but he must establish a logical connection between the evidence and his conclusion. O’Connor-Spinner, 627 F.3d at 618. That did not happen here.

Second, the ALJ did not explain why he gave more weight to the opinions of the state-agency psychologists than he did to Lanigan’s long-time counselor, Carrie Pai-sar. Edmund Musholt, one of the reviewing state-agency psychologists, opined that Lanigan’s claim that he cannot pay attention for more than a few minutes is “only partially credible because exams show his activities and interests involve adequate ability to sustain concentration and social interaction.” But Dr. Musholt did not identify, and thus the ALJ had no way of knowing, what “activities and interests” Lanigan supposedly was tackling adequately.

Perhaps Lanigan was succeeding at “activities and interests” relevant to competitive employment, or he might have been excelling at wholly irrelevant tasks, e.g., caring for his pets or vacuuming the house. See Stark v. Colvin, 813 F.3d 684, 688 (7th Cir. 2016) (concluding that the claimant’s “persistence in struggling through household chores despite her pain does not mean, as the ALJ extrapolated, that she can manage the requirements of the work-place” because a person performing chores, unlike an employee, has more flexibility “and is not held to a minimum standard of performance”); Hill v.

Colvin, 807 F.3d 862, 865, 869 (7th Cir. 2015) (warning against equating the activities of daily living—like babysitting, caring for pets, going to church, visiting with family members, and doing household chores— with those of a full-time job). On this record there is no way to know. In addition, ,Dr. Musholt’s assessment of Lanigan’s credibility differs from that of Esther Le-fevre, the psychologist who actually examined Lanigan in June 2012. True, Dr.

Le-fevre did not fully credit Lanigan’s account of his physical limitations, but she did not say anything suggesting skepticism about his report of significant mental limitations. Further, Dr. Musholt’s assessment dates to March 2013, before Lanigan’s involuntary commitment in June 2013, and nothing is said about how long Dr. Lefevre spent examining Lanigan in 2012.

In contrast, the records from Paisar, who had been seeing Lanigan weekly or biweekly for roughly two years, recount that in July 2013 Lanigan had “talked about being fearful of what he might do -if he gets angry and ‘blacks out’ ” again and was “frustrated with not being able to remember time periods.” The counselor’s opinion is more recent than those of the state-agency consultants, and it corroborates Lanigan’s own account of serious difficulties with concentration. See Clifford v.

Apfel, 227 F.3d 863, 870 (7th Cir. 2000) (“[M]ore weight is generally given to the* opinion of a treating physician because of his greater familiarity with the claimant’s conditions and circumstances.”). Yet the ALJ seized on just one of Lanigan’s many sessions with Paisar—when Lanigan had reported enjoying a day outdoors with a friend—and from that single session concluded that “increasing his activity outside of the house” had improved Lanigan’s symptoms “rather than caused anxiety attacks or other problems.” That selective reading of two years of treatment notes is not persuasive. See Larson v.

Astrue, 615 F.3d 744, 751 (7th Cir. 2010) (having “symptoms that ‘wax and wane’ [is] not inconsistent with a diagnosis of recurrent, major depression”); Bauer v. Astrue, 532 F.3d 606, 609 (7th Cir. 2008) (a claimant with a chronic disease like bipolar disorder “is likely to have better days and worse days,” and even if “half the time she is well enough that she could work,” she still “could not hold down a full-time job”). Similarly, the ALJ’s conclusion that Lanigan could maintain frequent (instead of only occasional) contact with coworkers is not supported by the record.

The distinction matters because the vocational expert testified that someone who is limited to occasional contact with coworkers and has the hypothetical’s other limitations cannot maintain competitive employment.

The ALJ reasoned that although Lanigan described severe problems being in public, he “actually works in a retail store[ ] about 15-22 hours a week” and has sustained other work that would require some social interaction “with no evidence that the work activity caused his symptoms to flare-up or made him need emergency treatment.” We have cautioned ALJs not to draw conclusions about a claimant’s ability to work full time based on part-time employment. See Jelinek v.

Astrue, 662 F.3d 805, 812 (7th Cir. 2011) (explaining that a claimant’s “brief, part-time employment” did not support the conclusion “that she was able to work a full-time job, week in and week out, given her limitations”); Larson, 615 F.3d at 752 (“There is a significant difference between being able to work a few hours a week and having the capacity to work full time.”). That is especially true when, as here, the claimant’s employer is accommodating him. See Larson, 615 F.3d at 752.

At the hearing Lanigan testified that his supervisor at Michaels was aware of his mental illness and “tolerant of it.” Instead of acknowledging the employer’s commendable generosity, the ALJ assumed that Lanigan’s work performance was no different than any other employee’s. Equally troubling, the ALJ failed to mention that Lanigan had been on the job for just six weeks, far too short a time to infer anything about his prospects of maintaining even part-time employment. See Jelinek, 662 F.3d at 812; Larson, 615 F.3d at 752.

And, finally, Lanigan’s testimony directly contradicts the ALJ’s assertion that his work activity never “caused his symptoms to flare-up.” Lanigan testified that the very purpose of his frequent breaks was to make sure his “emotions [were] in check.” And he said that on Black Friday he “melted down on the floor” and “went in the back for over a half hour just to get away from everybody.” Thus, as Lanigan contends, the ALJ’s hypothetical includes assumptions about his RFC that simply lack support in the record.

And, as Lanigan further argues, there is a second serious flaw in the hypothetical: it fails to account for his moderate limitations in concentration, persistence, and pace. See O’Connor-Spinner, 627 F.3d at 619 (“Among the limitations the [vocational expert] must consider are deficiencies of concentration, persistence and pace.”).

We have said that an ALJ must explicitly address those limitations in the hypothetical unless one of three exceptions applies: (1) the vocational expert was independently familiar with the claimant’s medical file; (2) the hypothetical adequately apprised the vocational expert of the claimant’s underlying mental conditions; or (3) the hypothetical otherwise accounted for the limitations using different terminology. Id. at 619-20. None of the exceptions applies here.

As for the first, the Commissioner does not contend that the vocational expert examined Lanigan’s medical records, even if he might have reviewed information about Lanigan’s employment history. And though the vocational expert was present when Lanigan testified at the hearing, that testimony was too limited to provide a complete and full picture of his mental limitations.

Regarding the next two exceptions, the Commissioner argues that the ALJ’s hypothetical effectively communicated, in different words, the idea that Lani-gan had experienced “moderate difficulties in concentration, persistence, or pace.” We cannot agree. The hypothetical begins by positing a person capable of performing “simple, routine, and repetitive tasks.” These terms refer to “unskilled work,” which the regulations define as work that can be learned by demonstration in less than 30 days. See 20 C.F.R. §§ 404.1568, 404.1520.

We have explained that the speed at which work can be learned is unrelated to whether a person with mental impairments—i.e., difficulties maintaining concentration, persistence, or pace—can perform such work. See Yurt, 758 F.3d at 858-59 (rejecting the notion that “confining the claimant to simple, routine tasks and limited interactions with others adequately captures temperamental deficiencies and limitations in concentration, persistence, and pace”); Stewart v. Astrue, 561 F.3d 679, 685 (7th Cir. 2009); Craft v. Astrue, 539 F.3d 668, 677-78 (7th Cir. 2008).

The Commissioner also argues that the hypothetical about “off-task” behavior informed the vocational expert about Lanigan’s moderate difficulties in the domain of concentration, persistence, or pace. As we’ve noted, however, to the extent that the 10% calculation was flawed, so was the hypothetical. See Young v. Barnhart, 362 F.3d 995, 1004 (7th Cir. 2004) (“For all of the same reasons that the RFC fell short, the hypothetical question, which was based entirely on that RFC[,] did as well.”).

Additionally, the Commissioner argues that even if the ALJ’s hypothetical was flawed, Lanigan waived any challenge to it by not objecting to the vocational expert’s testimony during the hearing. The Commissioner relies on Donahue v. Barnhart, which holds that claimants must raise objections to a vocational expert’s testimony at the hearing. 279 F.3d 441, 446 (7th Cir. 2002). But Lanigan is not challenging the vocational expert’s testimony.

Rather, he is challenging the lack of a substantial basis for the ALJ’s characterization of Lanigan’s mental RFC in the hypothetical questions he posed to the vocational expert. Accordingly, we Reverse the decision upholding the ALJ’s denial of benefits and Remand to the agency for further proceedings. . The GAF is a 100-point metric used to rate overall psychological, social, and occupational functioning, with lower scores corresponding to lower functioning. Am. Psychiatric Ass'n, Diagnostic and Statistical Manual of Mental Disorders 32-34 (4th ed., Text Rev. 2000).

A GAF score of 31 to 40 reflects "[s]ome impairment in reality testing or communication (e.g., speech is at times illogical, obscure, or irrelevant) OR major impairment in several areas, such as work or school, family relations, judgment, thinking, or mood (e.g., depressed man avoids friends, neglects family, and is unable to work...).” Id. at 34 (bold removed). The American Psychiatric Association eliminated use of the GAF in 2013, before the ALJ’s decision in this case, citing a "conceptual lack of clarity” and "questionable psychometrics in routine practice.” See Am.

Psychiatric Ass’n, Diagnostic and Statistical Manual of Mental Disorders 16 (5th ed. 2013). Although the American Psychiatric Association no longer uses this metric, at the time of Lanigan's psychological evaluations, clinicians still used GAF scores to report their assessment of a person’s overall level of functioning.