Lynn Marie LARSON, Plaintiff-Appellant, v. Michael J. ASTRUE, Commissioner of Social Security, Defendant-Appellee, 615 F.3d 744

WOOD, Circuit Judge. Lynn Marie Larson contends that she is disabled by anxiety, depression, and ankle pain. She applied for Supplemental Security Income (“SSI”), but an Administrative Law Judge (“ALJ”) concluded that her impairments, although severe, are not disabling. The district court upheld the agency’s decision, Larson v. Astrue, No. 09-cv-067-bbc, 2009 WL 3379144, at *1 (W.D.Wis. Oct.19, 2009), and Larson appeals. Among other things, she argues that the ALJ erred by discrediting her testimony and not giving controlling weight to the opinion of her long-term treating psychiatrist.

We agree with her that the evidence supports an award of benefits. I Now 38, Larson was educated through three years of college and has past work experience as a bartender. She has been under the care of mental health specialists since at least 1998, when she began seeing Dr. Bruce Rhoades, a psychiatrist. He diagnosed Larson with “major depression (recurrent) moderate.” His treatment notes from 1999 through 2003 show that he prescribed and regularly adjusted the dosages of several anti-depressants and anxiety medications.

Matters went from bad to worse for Larson in January 2004, when she was raped by the grandfather of one of her children and suffered a broken hand and injured thumb. She dates the onset of her disability from that incident. After the assault a social worker provided therapy for depression and post-traumatic stress disorder (“PTSD”). The social worker scored Larson at 50 on the Global Assessment of Functioning (“GAF”), which measures a person’s overall ability to function.

American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders, 30 (4th ed.1994). (A GAF of 50 indicates serious symptoms or functional limitations. Id. at 32.) Larson also consulted Dr. Rhoades, who observed that her mood was depressed though she appeared “pleasant and settled.” He diagnosed Larson with generalized anxiety disorder and possible PTSD, renewed her prescriptions for anti-depressants, and increased the dosage of her anti-anxiety medication. A few months later Dr.

Rhoades concluded that Larson was doing much better and scored her at 70 on the GAF; nevertheless, he confirmed his diagnosis of PTSD. In April 2004, Larson tripped outside a bar after five or six drinks and fractured her ankle in three places. The same orthopedist who had treated her after the rape surgically repaired the ankle fractures. A month later he concluded that the ankle was healing well. Around the same time, Larson confessed to Dr. Rhoades that she had started drinking more heavily and questioned whether her depression was the reason. Dr.

Rhoades responded by adjusting her medications; he decreased the dosage of her anti-depressants but, gauging her anxiety level as “fairly high,” he increased her anti-anxiety medication. Her GAF was back down to 50. Dr. Rhoades later reported that Larson’s anxiety was “under reasonable control,” a view that prompted him to change her medication again. In June 2004 she applied for benefits. Her initial application was limited to allegations relating to the pain from her broken ankle; later she added allegations of disability stemming from mental impairments.

A month later, Larson’s stepfather beat her and re-injured her ankle. X-rays showed no evidence of a new fracture, but Larson told her orthopedist that she was having difficulty walking without an ankle brace. She also saw Dr. Rhoades, who reconfirmed the diagnosis of major depression and prescribed additional anti-depressants and anti-anxiety medications. Other issues in 2004 and 2005 led to further consultations with Dr. Rhoades and Jennifer Herink, a psychotherapist. In August 2004 Dr. Rhoades noted that Larson was “not doing very well” and prescribed additional medication to treat her depression and anxiety.

She had a “nervous breakdown” and missed almost two weeks of work at the Head Start program where she had been working part-time as a bus driver. A nephew she had been raising was placed in foster care after a social services agency investigated an allegation of child neglect. And she was arrested for driving while under the influence. She reported to Herink that she had stopped taking her prescription medication and started (or, it seems, continued) self-medicating with marijuana and alcohol. Larson reported to Dr. Rhoades that she was not getting out of bed, and so he prescribed two additional anti-depressants.

Throughout the last half of 2004 and 2005, Dr. Rhoades documented that Larson was depressed, assessed her GAF at 50 to 60, and prescribed additional medications to control her anxiety and mood. Larson’s application for SSI was denied initially in August 2004 and upon reconsideration in August 2005. A state-agency psychologist had reviewed the medical record shortly before the second denial and had assessed Larson’s mental impairments using a standard form “Psychiatric Review Technique,” see 20 C.F.R. § 404.1520a.

He diagnosed Larson with an “affective disorder” — specifically depression — under Listing 12.04 and an “anxiety-related disorder” under Listing 12.06, see 20 C.F.R. Pt. 404, Subpt. P., App. 1. In his opinion neither of the impairments was severe. He concluded that Larson had not suffered an extended episode of decompensation (a somewhat vague term whose meaning we explore below) and was experiencing only “mild” restrictions on daily living activities and “moderate” difficulties in the realms of social functioning and concentration, persistence, or pace.

He thought that Larson could perform simple, repetitive, low-stress work even though she would probably have trouble dealing with large groups of people or stressful situations. In December 2005, Dr. Rhoades completed a Mental Impairment Questionnaire. He reported there that he had been treating Larson since 1998 on roughly a monthly basis. His diagnosis was severe, recurrent depression and dissociate identity disorder. Her current GAF score, he said, was 50. Observing that she avoids most social situations, he noted that Larson was experiencing repeated (i. e., three or more) episodes of decompensation.

He also checked a box indicating that she had “slight” restrictions in activities of daily living, “marked” difficulties in social functioning, and “frequent” deficiencies of concentration, persistence, or pace. In January 2006, Larson reported to Herink that she was having increased thoughts of suicide. Herink encouraged her to go to the hospital, but she did not follow that advice. Herink later asked the police to check on Larson. They did so and, according to Herink’s progress notes, they took her to the hospital. The administrative record contains no other information about this hospitalization.

Larson briefly testified before the ALJ at her hearing in March 2007. Twice during the questioning she said that she wanted to “go home.” Much of her testimony focused on her efforts to hold a job since her alleged onset date. The month before the hearing, Larson had quit a part-time job at a gas station, where she occasionally had to hide in the bathroom, apparently to avoid customers. Since 2004 she also had been working about two hours per week at a restaurant, tending bar, cooking, and waiting tables.

Although she drove the Head Start school bus for a short period, she was fired from that job after suffering a breakdown that alerted her employer to the drugs she was taking. Larson insisted that she could not work full-time because she suffers panic attacks and uncontrollable crying spells that last as little as 15 minutes to as long as several hours. She had succeeded in keeping the restaurant job, she explained, only because the owner, Paul Calliss, was a friend.

Larson explained that she was caring for her four children (then ranging in age from 6 to 17), cooking for the family, and doing laundry and other household chores. If she needed to shop, she went with Calliss or else waited until late in the evening to avoid encountering other people. Finally, she reported that she was taking medications (she did not say what or for which condition) that made her drowsy and required her to nap during the day.

Calliss, Larson’s friend and employer, confirmed that Larson worked for him “very part time,” sometimes all day if there was a special event but on average less than two hours per week. He explained that Larson typically needed a break after a short time because she was nervous around strangers. The ALJ called Dr. Steven Carter, a psychologist, as a medical expert to testify about Larson’s mental impairments. Dr. Carter considered whether Larson met the criteria for a per se disability under Listing 12.04 for depression.

A claimant suffering from an affective disorder meets the listed severity level for a depressive syndrome if enough listed factors (the “A criteria”) are present. 20 C.F.R. Pt. 404, Subpart P., App. 1, § 12.04(A); see Rabbets v. Comm’r Soc. Sec. Admin., 582 F.3d 647, 653 (6th Cir.2009); Holohan v. Massanari, 246 F.3d 1195, 1203 (9th Cir.2001).

There is no dispute that Larson’s depression qualified as severe for purposes of the “A criteria.” But the “A criteria” alone are not enough; in order to be considered per se disabled, at least two of the following “B criteria” must also be present: (1) “marked” restriction of activities in daily living, (2) “marked” difficulties in maintaining social functioning, (3) deficiencies of concentration, persistence or pace, or (4) “repeated episodes of decompensation each of extended duration.” 20 C.F.R. Pt. 404, Subpart P., App. 1, § 12.04(B); see Craft v. Astrue, 539 F.3d 668, 674-75 (7th Cir.2008); Randolph v.

Barnhart, 386 F.3d 835, 840 (8th Cir.2004). Dr. Carter testified that Larson met the “A criteria” but had not satisfied the “B criteria.” He thought (erroneously, as we have pointed out) that Larson had not been hospitalized or been in a group home, and on that basis he concluded that she had never experienced an extended episode of decompensation. He was willing to concede that she had “marked” restrictions in social activities, but he saw only “mild” restrictions on daily activities and no significant limitations with respect to concentration, persistence, or pace.

He opined that Larson should work in a low-stress, alcohol-free environment that did not involve large crowds. Dr. Carter did not try to reconcile his assumption that Larson had not been hospitalized with Herink’s progress note reporting that Larson had been taken to the hospital by the police. Nor did he confront Dr. Rhoades’s contradictory finding that Larson had suffered “repeated” episodes of decompensation. He also said nothing about the relevance to Listing 12.04 of the frequent adjustments to her medications. And although Dr.

Carter had been present for Larson’s testimony, he did not comment on her assertion that she had suffered a nervous breakdown while working as a bus driver or the fact that she had been fired from that position shortly thereafter. After the hearing, the ALJ denied Larson’s claim. At Step 1 of the five-step analysis prescribed in 20 C.F.R. § 404.1520, the ALJ found that Larson had not engaged in gainful employment since her 2004 onset date.

At Step 2 the ALJ concluded that Larson suffered from severe impairments, namely, left ankle pain, a left wrist fracture, right hand osteoarthritis, affective disorder, and anxiety disorder. At Step 3, however, the ALJ concluded that none of the impairments was medically equivalent to anything on the lists. After explaining that he was adopting the opinion of Dr.

Carter, the ALJ asserted that “there is no evidence in the record that the claimant has ever suffered an episode of decompensation of extended duration due to her psychological symptoms.” The ALJ acknowledged that Larson was experiencing restrictions in social functioning, but he reasoned that those restrictions were only “moderate” because she attended doctor’s appointments on a regular basis, was able to go to the grocery store alone if she went at night, and had good relationships with two friends. The ALJ pegged her difficulties maintaining concentration, persistence, and pace at the “moderate” level.

At Step 4 the ALJ stated that he had given “some weight” to Dr. Rhoades’s opinion that Larson met the criteria in the listing, but he found that Rhoades’s assessment of the severity of Larson’s symptoms was not sufficiently corroborated. Larson’s testimony about the severity of her impairments, the ALJ thought, was inconsistent with her account of her daily activities. The ALJ concluded that her “psychological symptoms wax and wane based on situational stressors.” At Step 5, the ALJ concluded based on the testimony of a vocational expert that Larson could work as a hand packager or electronics worker.

II On appeal, Larson first argues that the ALJ should have granted controlling weight to the opinion of Dr. Rhoades, who easily qualified as her treating psychiatrist, and found her disabled at Step 3. Larson maintains that the assessment given by Rhoades is consistent — or at least not inconsistent — with the evidence in the record. A treating physician’s opinion is entitled to “controlling weight” if it is “well-supported by medically acceptable clinical and laboratory diagnostic techniques and is not inconsistent with other substantial evidence.” 20 C.F.R. § 404.1527(d)(2); see Schaaf v.

Astrue, 602 F.3d 869, 875 (7th Cir.2010); Denton v. Astrue, 596 F.3d 419, 424 (7th Cir.2010). An ALJ who does not give controlling weight to the opinion of the claimant’s treating physician must offer “good reasons” for declining to do so. 20 C.F.R. § 404.1527(d)(2); see Schaaf, 602 F.3d at 875; Schmidt v. Astrue, 496 F.3d 833, 842 (7th Cir.2007). If the ALJ had given Dr. Rhoades’s opinion controlling weight, Larson’s condition would have been recognized as a listed impairment and she would have been found disabled at Step 3. See 20 C.F.R. § 404.1520a(d)(1); Craft, 539 F.3d at 675.

All medical experts agreed that Larson met the “A criteria” for depression, and Dr. Rhoades opined that she satisfied the “B criteria” as well. With regard to the latter, Dr. Rhoades saw “marked” difficulties in Larson’s ability to function socially. The ALJ offered several reasons for rejecting Rhoades’s view. First, he found Dr. Carter’s opinion more persuasive. In addition, he was impressed by the evidence showing that Larson attended doctors’ appointments, went grocery shopping at night, and had good relationships with two friends.

All of this indicated to him that her difficulties were just “moderate.” But the ALJ mischaracterized Dr. Carter’s opinion; Dr. Carter, like Dr. Rhoades, had concluded that Larson experienced “marked,” not “moderate,” limitations in this area. Whether by mistake or design, the ALJ disregarded this medical evidence and improperly substituted his own opinion. See Murphy v. Astrue, 496 F.3d 630, 634 (7th Cir.2007) (“[A]n ALJ cannot disregard medical evidence simply because it is at odds with the ALJ’s own unqualified opinion.”); Clifford v. Apfel, 227 F.3d 863, 870 (7th Cir.2000).

The Commissioner tries to salvage the ALJ’s conclusion by pointing to one instance in 2004 when Dr. Rhoades described Larson’s demeanor as “pleasant and settled” and by recalling that the state-agency psychologist thought that Larson had only “moderate” limitations in social functioning. But these are not reasons that appear in the ALJ’s opinion, and thus they cannot be used here. See SEC v. Chenery Corp., 318 U.S. 80, 87-88, 63 S.Ct. 454, 87 L.Ed. 626 (1943); Parker v. Astrue, 597 F.3d 920, 922 (7th Cir.2010); Stewart v. Astrue, 561 F.3d 679, 684 (7th Cir.2009). Dr.

Rhoades also stated that Larson had experienced repeated (by which he meant at least three) episodes of decompensation. The ALJ disagreed, insisting that there was “no evidence in the record that the claimant has ever suffered an episode of decompensation of extended duration due to her psychological symptoms.” The ALJ’s conclusion followed Dr. Carter’s. But, as we noted earlier, both the ALJ and Dr. Carter overlooked the evidence from Herink of Larson’s suicidal thinking and trip to the hospital in 2006.

Although everyone seemed to think that he or she knew what is meant by “episodes of decompensation,” this is not a self-defining phrase. Dr. Carter took an approach that was too narrow in light of the

definitions that the Social Security Administration uses. The listing defines “episodes of decompensation” as “exacerbations or temporary increases in symptoms or signs accompanied by a loss of adaptive functioning, as manifested by difficulties in performing activities of daily living, maintaining social relationships, or maintaining concentration, persistence, or pace.” 20 C.F.R. Pt. 404, Subpart P., App. 1, § 12.00; see also Stedman’s Medical Dictionary, 497 (28th ed.2006) (defining decompensation as the “appearance or exacerbation of a mental disorder due to failure of defense mechanisms”); Zabala v.

Astrue, 595 F.3d 402, 405 (2d Cir.2010) (stating that decompensation is a temporary increase in symptoms); Kohler v. Astrue, 546 F.3d 260, 266 n. 5 (2d Cir.2008) (same). An incident — such as hospitalization or placement in a halfway house— that signals the need for a more structured psychological support system would qualify as an episode of decompensation, 20 C.F.R. Pt. 404, Subpart P., App. 1, § 12.00, but so would many other scenarios. The listing recognizes that an episode may be inferred from medical records showing a significant alteration in medication, see 20 C.F.R. Pt. 404, Subpart P., App. 1, § 12.00.

Larson has a long history of problems that have led to significant alterations in her medications. See Rubbers, 582 F.3d at 660 (observing that treating physician’s testimony that side effects of medication affects claimant’s ability to function is consistent with a finding of repeated episodes of decompensation); Natale v. Comm’r of Soc.

Sec., 651 F.Supp.2d 434, 451-53 (W.D.Pa.2009) (stating that it was error for ALJ to reject treating physician’s conclusion that claimant suffered repeated episodes of decompensation where claimant had history of adjustments to medication and fluctuating mood); 3 Social Security Law & Practice § 42:124 (2010) (“Episodes of decompensation may be demonstrated by an exacerbation in symptoms or signs that would ordinarily require increased treatment or a less stressful situation (or a combination of the two).”). Dr.

Carter, despite being present for Larson’s testimony, never mentioned the nervous breakdown that caused her to miss almost two weeks of work; that incident qualified as an episode of decompensation, see Lankford v. Sullivan, 942 F.2d 301, 307-08 (6th Cir.1991).

In addition, although the listing defines “repeated episodes of decompensation” as three episodes within one year or an average of one every four months (each lasting for at least two weeks), the listing also states that for claimants who experience more frequent episodes of shorter duration, the ALJ should determine if the duration and the functional effects are of equal severity. 20 C.F.R. Pt. 404, Subpart P., App. 1, § 12.00. A fair reading of the record indicates that Dr. Rhoades reached exactly the latter conclusion about Larson. In response, the Commissioner makes much of the fact that Dr.

Rhoades did not explain on the questionnaire his conclusion that Larson had experienced repeated episodes of decompensation. But in every

section on the questionnaire that allowed for comments, Dr. Rhoades made them; the question dealing with Larson’s functional limitations and episodes of decompensation did not invite further explanation or include space for comments. Although by itself a check-box form might be weak evidence, the form takes on greater significance when it is supported by medical records. Mason v. Shalala, 994 F.2d 1058, 1065 (3d Cir.1993); see also Johnson v. Apfel, 189 F.3d 561, 564 (7th Cir.1999) (upholding ALJ’s rejection of physician’s check-box form where it was contradicted by evidence in the record).

Here, there is a long record of treatment by Dr. Rhoades that supports his notations on the form. In arguing that “the ALJ also premised his weighing of Dr. Rhoades’s assessment on the fact that his treatment notes reflected “waxing and waning symptoms’ depending on particular situational stressors,” the Commissioner distorts the ALJ’s reasoning. The ALJ simply stated that “the claimant’s psychological symptoms wax and wane based on situational stressors” without tying this observation to Dr. Rhoades’s treatment notes.

More importantly, symptoms that “wax and wane” are not inconsistent with a diagnosis of recurrent, major depression. “A person who has a chronic disease, whether physical or psychiatric, and is under continuous treatment for it with heavy drugs, is likely to have better days and worse days.” Bauer v. Astrue, 532 F.3d 606, 609 (7th Cir.2008). No doctor concluded that Larson’s symptoms were just a response to situational stressors as opposed to evidence of depression. The ALJ’s conclusion to the contrary thus finds no support in the record. See Hopgood ex rel. L.G. v. Astrue, 578 F.3d 696, 702 (7th Cir.2009).

Even if the ALJ had articulated good reasons for rejecting Dr. Rhoades’s opinion, it still would have been necessary to determine what weight his opinion was due under the applicable regulations. See 20 C.F.R. § 404.1527(d)(2). An ALJ must consider the length, nature, and extent of the treatment relationship; frequency of examination; the physician’s specialty; the types of tests performed; and the consistency and support for the physician’s opinion. Moss v. Astrue, 555 F.3d 556, 561 (7th Cir.2009); Elder v. Astrue, 529 F.3d 408, 415 (7th Cir.2008); Hofslien v. Barnhart, 439 F.3d 375, 377 (7th Cir.2006).

These factors support Dr. Rhoades: he had treated Larson for several years on a monthly basis; he is a psychiatrist, not a psychologist; and his opinion is consistent with the evidence in the record. Apart from the ALJ’s unhelpful statement that Dr. Rhoades’s opinion was entitled to “some weight,” the ALJ said nothing regarding this required checklist of factors. See 20 C.F.R. § 404.1527(d)(2); Moss, 555 F.3d at 561; Bauer, 532 F.3d at 608-09. Thus, the ALJ erred by failing to give controlling weight to Dr.

Rhoades’s opinion about the limitations on Larson’s social functioning and her experience with episodes of decompensation. Once we give his opinion the proper weight, the record shows that Larson’s condition meets the standards of the Listing, and thus that the ALJ should have found her disabled at Step 3. Ill Our conclusion is reinforced by the problems we see in the approach the ALJ took in his assessment of Larson’s credibility. Normally, we give an ALJ’s credibility determinations special deference because the ALJ is in the best position to see and hear the witness. See Eichstadt v.

Astrue, 534 F.3d 663, 667-68 (7th Cir.2008). But it is nevertheless possible to upset a credibility finding if, after examining the ALJ’s reasons for discrediting testimony, we conclude that the finding is patently wrong. See Schaaf, 602 F.3d at 875; Prochaska v. Barnhart, 454 F.3d 731, 738 (7th Cir.2006); Sims v. Barnhart, 442 F.3d 536, 538 (7th Cir.2006). Here, the ALJ rejected Larson’s testimony about the severity of her symptoms.

The ALJ suggested that Larson must have overstated the effects of her impairments because she had developed relationships with two friends whom she visited often, she had “held down” a series of part-time jobs, and she had accommodated her fear of the public by going to the grocery store at night. But Larson’s ability to maintain a small number of close friendships does not undermine her testimony that she is afraid of going out in public. And the ALJ’s assertion that Larson has succeeded in holding down a series of part-time jobs stretches the evidence beyond the breaking point.

There is a significant difference between being able to work a few hours a week and having the capacity to work full time. See SSR 96-8p, 1996 WL 374184, at *1; Diaz v. Prudential Ins. Co. of Am., 499 F.3d 640, 648 (7th Cir.2007); Carradine v. Barnhart, 360 F.3d 751, 755 (7th Cir.2004). Larson was able to work for Calliss part-time only because he was a friend who tolerated frequent breaks and absences that an ordinary employer would have found unacceptable. This does not contradict her claim of disability. See Henderson v. Barnhart, 349 F.3d 434, 435 (7th Cir.2003).

In fact, it is not accurate to say that Larson “held down” (meaning kept) these part-time jobs. She testified without contradiction that she was fired from her job at Head Start because of her nervous breakdown and the medications she was taking, and she quit her job at the gas station because of the stress in dealing with unfamiliar customers. Last, the ALJ’s conclusion that Larson accommodated her fear of going out in public does not discredit her testimony that she has a fear of going out in public and gives in to that fear regularly.

Nothing in the record supported the ALJ’s inference that there were ways in which Larson’s condition could be treated or managed. See Myles v. Astrue, 582 F.3d 672, 677-78 (7th Cir.2009). The ALJ’s reasons for his adverse credibility ruling find no support, on close examination, and for that reason, the credibility determination cannot stand. IY For these reasons, we find that Larson is entitled to an award of benefits. The ALJ’s decision is Reversed and the case is Remanded for entry of an order consistent with this opinion.