Jelinek v. Astrue

In the United States Court of Appeals For the Seventh Circuit No. 10-3340 JESSICA J. JELINEK, Plaintiff-Appellant, v. M ICHAEL J. A STRUE, Commissioner of Social Security, Defendant-Appellee. Appeal from the United States District Court for the Northern District of Indiana, South Bend Division. No. 3:09-cv-368—Christopher A. Nuechterlein, Magistrate Judge. A RGUED A UGUST 2, 2011—D ECIDED N OVEMBER 7, 2011 Before W OOD , S YKES, and H AMILTON, Circuit Judges. H AMILTON, Circuit Judge. Jessica Jelinek’s mother applied for supplemental security income on her daugh- ter’s behalf shortly before Jelinek’s eighteenth birth- day. She contended that Jelinek was disabled by a combi- nation of mental impairments (including bipolar disorder) and by physical impairments resulting from a 2005 car accident. An administrative law judge found Jelinek’s 2 No. 10-3340 collective impairments severe but not disabling. On appeal, Jelinek argues that the ALJ improperly rejected the opinion of her treating psychiatrist and that this mistake led to additional errors in the ALJ’s reasoning. We reverse the judgment and remand for further pro- ceedings. I. Facts and Procedural Background Jelinek, now 23 years old, was 17 in August 2005 when her mother filed an application for supplemental security income benefits on her behalf. Her application was denied initially and on reconsideration. In February 2008 Jelinek appeared at a hearing before an ALJ, who later issued a decision confirming the denial of benefits. The Appeals Council denied review, leaving the ALJ’s deci- sion as the Commissioner’s final word. See 20 C.F.R. § 416.1481; Steele v. Barnhart, 290 F.3d 936, 940 (7th Cir. 2002). Jelinek sought review in the district court, see 42 U.S.C. § 405(g), and a magistrate judge presiding by consent under 28 U.S.C. § 636(c) upheld the decision. This appeal followed. A. The Medical Evidence Jelinek received counseling at the Bowen Center (from several medical sources) on a near-monthly basis from October 2002 through February 2008. She was assessed initially with depression and thoughts of suicide, precipi- tated in part by a history of abuse inflicted by her father. Jelinek reported a history of cutting and burning herself. No. 10-3340 3 Early treatment notes cite a “suicidal gesture” in Janu- ary 2002 (apparently including a plan to overdose by taking Aleve), but staff at the Bowen Center initially concluded that Jelinek posed a “low risk” of harm to herself or others and showed no sign of psychosis. Jelinek’s school records from this period show 71 absences during the 2002-03 school year and 79 absences during the 2003-04 school year. The diagnoses of Jelinek’s mental impairments shifted over the course of her treatment at the Bowen Center. By the time of her hearing before the ALJ, she was either diagnosed with or had a history of major depression, bipolar disorder (with manic and psychotic features), ADHD, post-traumatic stress disorder, anxiety disorder, and borderline personality disorder. Adding to that, she also was obese and suffered from chronic pain. The treatment for her mental impairments included an ever- changing combination of medications and dosages, among them the anti-depressants Zoloft, Prozac, Trazodone, and Celexa; the anti-anxiety drugs Ativan, Trazodone, and Vistaril; the stimulant Adderall; and the anti-psychotic drugs Abilify, Geodon, Risperdal, and Zyprexa. Treatment notes show some concern among Jelinek’s doctors that, at least initially, her mother was resistant to Jelinek’s medication regimen. Doctors peri- odically assigned Jelinek a global assessment of func- tioning (“GAF”) score, which is a psychiatric measure of 4 No. 10-3340 a patient’s overall level of functioning.1 The record shows that Jelinek’s GAF scores tended to fluctuate between 50 and 55, numbers on the border between “moderate” and “serious” impairment in function, with a high of 65 in April 2007 and a low of 20 during her brief hospital- ization in December 2007. In July 2005, Jelinek was in a car accident that left her with fractures to her pelvis and two vertebrae. She did not require surgery, and by October 2005 her doctors deemed those fractures to be healed and recommended physical therapy. Jelinek, though, continued to report back pain. From February 2006 through December 2007, she visited Dr. Aashish Deshpande for pain management. Dr. Deshpande prescribed fentanyl for her pain and, by turns, Vicodin and later Percocet for “breakthrough” periods of additional pain. It was the car accident that apparently precipitated Jelinek’s application for supplemental security income benefits, and in October 2005, soon after filing, she was 1 A GAF between 41 and 50 indicates “Serious symptoms (e.g., suicidal ideation, severe obsessional rituals, frequent shop- lifting) OR any serious impairment in social, occupational, or school functioning (e.g., no friends, unable to keep a job).” In turn, a GAF between 51 and 60 reflects “Moderate symptoms (e.g., flat affect and circumstantial speech, occasional panic attacks) OR moderate difficulty in social, occupational, or school functioning (e.g., few friends, conflicts with peers or co-workers).” American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders, Text Revision (DSM-IV-TR) 34 (4th ed. 2000). No. 10-3340 5 examined by Dr. John Heroldt, a state-agency psycho- logist. He noted that Jelinek reported being “bipolar and manic,” but he opined that she did not exhibit psy- chotic symptoms. He diagnosed her with major depres- sion (with psychotic features) and borderline per- sonality disorder, and assigned her a GAF score of 50. After meeting with Jelinek, Dr. Heroldt concluded that she was unable to handle her own finances. In February 2006 a second state-agency psychologist, Dr. F. Kladder, reviewed Jelinek’s treatment records to assess her residual functional capacity. Dr. Kladder opined that Jelinek presented several symptoms of depres- sion, including anhedonia (the inability to experience pleasure), appetite and sleep disturbance, and difficulty concentrating. But he concluded that Jelinek suffered only mild restrictions in the activities of daily living and moderate difficulties in maintaining social functioning, and in concentration, persistence, and pace. He opined that Jelinek was “moderately limited” in her ability to complete a normal workday or workweek, that she had trouble in close relationships and would likely have trouble taking criticism or supervision, and that her allegations of disability were credible. He also noted that Jelinek’s medical record showed one or two episodes of decompensation of prolonged duration. See 20 C.F.R. Pt. 404, Subpt. P, App. 1, § 12.00 (defining “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, 6 No. 10-3340 or pace”); see also Larson v. Astrue, 615 F.3d 744, 750 (7th Cir. 2010) (explaining term). From August 2006 through January 2008, Dr. Snieguole Radzeviciene, a psychiatrist at the Bowen Center, over- saw Jelinek’s psychiatric treatment. His notes from an April 2007 evaluation identify Jelinek’s impairments as major depressive disorder, ADHD, anxiety disorder, PTSD, obesity, and chronic pain. Those notes also docu- ment Jelinek’s concern that she did not have insurance and could not afford therapy. In August 2007 Dr. Radzeviciene observed that Jelinek appeared “more volatile” and documented her report that a recent breakup with her boyfriend had led to an increase in worry, depression, crying, and irritability. In December 2007 Jelinek admitted herself to the Bowen Center’s inpatient unit. She reported hearing voices and said she feared that someone was trying to kill her. The on-call psychiatrist, Dr. Barbara Eichman, was unsure how long Jelinek had been “decompensating” and characterized her as “very delusional.” Dr. Eichman cited Jelinek’s breakup with her longtime boyfriend as a possible factor and opined that Jelinek was probably not taking all her medications. The doctor also expressed doubt about the efficacy of the series of med- ications (six in total, including two pain medications, two anti-depressants, and two anti-psychotics) that Jelinek was then prescribed. One month later Dr. Radzeviciene completed a mental- impairment questionnaire. He diagnosed Jelinek with bipolar disorder with manic features, ADHD, and obesity. No. 10-3340 7 Among Jelinek’s symptoms, Dr. Radzeviciene listed poor memory; sleep, mood, and emotional disturbance; dif- ficulty concentrating; oddities of thought, speech, and behavior; manic syndrome; and general anxiety. He noted that Jelinek’s mania had been resolved but that she remained depressed and that her mood episodes were likely to recur. He also noted that Jelinek’s depression and anxiety tended to make her chronic pain worse. He assessed Jelinek’s ability to maintain attention for two-hour segments as “fair,” and opined that she had poor or no ability to maintain regular atten- dance, to perform at a regular pace without an unreason- able number of rest periods, to deal with normal work stress, or to deal with the stress of skilled or semi-skilled work. He described Jelinek’s deficiencies of concentra- tion, persistence, or pace as “frequent” and episodes of deterioration or decompensation as “repeated.” B. The Hearing Testimony At her February 2008 hearing before the ALJ, Jelinek testified that she lived across the hall from her mother, who helped with shopping and cleaning her apartment. Jelinek recounted problems with insomnia and bipolar mania, and explained that she had missed a lot of school and ultimately dropped out of high school in her sopho- more year because of depression and anxiety. She had gotten her GED and was then enrolled in four college classes, but added that she had dropped three of four classes the previous semester because of stress and the breakup with her boyfriend. Regarding her work history, 8 No. 10-3340 Jelinek testified that she was working eight hours per week as a library assistant at her college and had previ- ously worked part-time as a motel clerk for about three months. Vocational expert Leonard Fisher also testified at the hearing. The ALJ prompted the expert to opine whether jobs existed in the local economy for light, unskilled work. The expert testified that about 10,000 jobs existed in the local four-county region for a claimant with Jelinek’s age, education, and background of part-time work. When the ALJ changed the hypothetical to seden- tary, unskilled work, the expert opined that only 600 to 900 relevant jobs existed. The expert also opined, however, that an unskilled worker with Jelinek’s profile who missed more than one day a month (other than vacation days, sick days, and holidays) would have “difficulty in sustaining competitive employment.” C. The ALJ’s Decision In June 2008 the ALJ issued a written decision concluding that Jelinek was not disabled. The ALJ found that Jelinek had engaged in no substantial gainful activity since her application was filed. Because Jelinek’s initial application was a child application filed on her behalf, the ALJ first examined her claims under the three- step “child standard” for the period before her eighteenth birthday. Under supplemental security income rules, a child is disabled if she has a “medically determinable physical or mental impairment, which results in marked and severe functional limitations” that “has lasted or can No. 10-3340 9 be expected to last for a continuous period of not less than 12 months.” See 42 U.S.C. § 1382c(a)(3)(C)(i). This assessment requires a three-step analysis set forth in 20 C.F.R. § 416.924(a). First, if the child is engaged in substantial gainful activity, the ALJ will deny the claim. Second, if the child does not have a severe medical impairment or combination of impairments, then she is not disabled. Third, the child’s impairments must meet, medically equal, or functionally equal any of the listings contained in 20 C.F.R. Pt. 404, Subpt. P, App. 1. The ALJ found that Jelinek had the severe impairments of ADHD, bipolar disorder, personality disorder, and PTSD, and a history of pelvic, cervical, and thoracic fractures. The ALJ also cited Jelinek’s poor school perfor- mance and excessive absences. The ALJ found that none of these impairments, alone or in combination, met, medically equaled, or functionally equaled any listing before Jelinek reached age 18.2 He found that Jelinek had (1) “no limitation” in acquiring and using information, attending and completing tasks, and caring for herself; and (2) “less than marked limitation” in interacting or 2 To determine if an impairment is “functionally equivalent” to a listing, an ALJ analyzes its severity in six “domains”: (1) acquiring and using information; (2) attending and com- pleting tasks; (3) interacting and relating with others; (4) moving about and manipulating objects; (5) caring for oneself; and (6) health and physical well-being. 20 C.F.R. § 416.926a(b)(1). To functionally equal a listing, the ALJ must find an “ex- treme” limitation in one category or a “marked” limitation in two categories. 20 C.F.R. § 416.926a(a), (e)(2)(i). 10 No. 10-3340 relating to others, moving about and manipulating objects, and health and physical well-being. He reasoned that her symptoms were “reasonably controlled with medication compliance” but noted that she sometimes needed prompting to take her medication. On this basis, the ALJ concluded that Jelinek had not been disabled before her eighteenth birthday. The ALJ then reviewed Jelinek’s application under the five-step “adult standard” for the period after she turned 18.3 He found that Jelinek had developed no new impairments and found again, without additional analysis, that Jelinek had no impairment or combination of impairments that met or medically equaled a listing. The ALJ acknowledged the restrictions in concentration, persistence, and pace suggested by Dr. Radzeviciene’s January 2008 opinion, as well as the doctor’s opinion that she had no effective ability to maintain regular attendance, perform at a consistent pace, or deal with normal work stresses. But without explicit analysis, the ALJ offset those concerns with a statement that Jelinek “functioned well and passed the GED test, attended college and obtained good grades, and worked selling Avon and part-time at a hotel.” The ALJ opined that Jelinek retained the residual functional capacity to 3 If a child claimant turns 18 after filing a disability application but before a final decision, the ALJ uses the rules governing child applications for the period before the claimant turned 18. For the period starting the day the claimant turns 18, the ALJ uses the disability rules for adults who file new claims. 20 C.F.R. §§ 416.920, 416.924(f). No. 10-3340 11 engage in light and sedentary unskilled work. In making this determination, he concluded that Jelinek’s statements about the intensity, persistence, and limiting effects of her symptoms were not “entirely credible” because progress notes often showed her to be doing well. He reiterated that she had obtained her GED, and he cited Dr. Radzeviciene’s April 2007 examination as support for his conclusion that medication non-com- pliance was often the culprit behind Jelinek’s decom- pensation. Then, citing the vocational expert’s testimony, the ALJ concluded that jobs existed for Jelinek in signifi- cant numbers in the national economy. He concluded that Jelinek was not disabled from her eighteenth birthday to the date of the decision. II. Discussion On appeal Jelinek challenges several aspects of the ALJ’s decision. Chiefly she contends that the ALJ improp- erly discounted Dr. Radzeviciene’s assessment of her mental impairments. This mistake led the ALJ in turn, Jelinek argues, (1) to evaluate improperly whether her mental impairments met the listing for “affective disor- ders”; (2) to overstate her residual functional capacity when questioning the vocational expert; and (3) to err in finding her not credible. On judicial review, a court will uphold the Commissioner’s decision if the ALJ applied the correct legal standards and supported his decision with substantial evidence. 42 U.S.C. § 405(g); Castile v. Astrue, 617 F.3d 923, 926 (7th Cir. 2010); Terry v. Astrue, 580 F.3d 471, 475 (7th Cir. 2009). A decision 12 No. 10-3340 denying benefits need not discuss every piece of evidence, but when an ALJ fails to support her conclusions ade- quately, remand is appropriate. Villano v. Astrue, 556 F.3d 558, 562 (7th Cir. 2009). We limit our review to the reasons articulated by the ALJ in the written decision. See SEC v. Chenery, 318 U.S. 80, 93-94 (1943); Spiva v. Astrue, 628 F.3d 346, 353 (7th Cir. 2010); Larson, 615 F.3d at 749. We begin with Jelinek’s contention that the ALJ failed to adhere to the “treating physician rule.” A treating physician’s opinion that is consistent with the record is generally entitled to “controlling weight.” 20 C.F.R. § 404.1527(d)(2); Schaaf v. Astrue, 602 F.3d 869, 875 (7th Cir. 2010). An ALJ who chooses to reject a treating physi- cian’s opinion must provide a sound explanation for the rejection. 20 C.F.R. § 404.1527(d)(2); Campbell v. Astrue, 627 F.3d 299, 306 (7th Cir. 2010); Schmidt v. Astrue, 496 F.3d 833, 842 (7th Cir. 2007). Jelinek argues, and we agree, that the ALJ did not explain satisfactorily in the written decision his rejection of Dr. Radzeviciene’s opinion. When an ALJ decides to favor another medical professional’s opinion over that of a treating physician, the ALJ must provide an account of what value the treating physician’s opinion merits. See Scott v. Astrue, 647 F.3d 734, 740 (7th Cir. 2011). The ALJ’s deci- sion did not meet these requirements for rejecting Dr. Radzeviciene’s opinion. The ALJ’s decision does not allow us to conclude that he weighed the merits of Dr. Radzeviciene’s opinion, let alone engaged in the careful analysis required by the regulations and case law. Dr. Radzeviciene’s assess- No. 10-3340 13 ment, as reflected in the questionnaire he completed in January 2008, addressed Jelinek’s symptoms and her residual functional capacity and thus was highly relevant to several parts of the ALJ’s analysis. But in a section devoted to determining whether Jelinek’s mental impairments met or medically equaled a listed impair- ment under the adult five-step analysis, the ALJ limited his comments to the following: It is noted that . . . Dr. Radzeviciene opined in January 2008 that the claimant has fair (seriously limited) ability to maintain attention for two-hour periods, sustain an ordinary routine without special supervision; complete a normal workday and work- week; work in coordination with others without being unduly distracted; understand, remember, and carry out detailed instructions; set realistic goals and make plans independently of others; and use public transportation; poor/no ability to maintain regular attendance and be punctual; perform at a consistent pace without an unreasonable number and length of rest periods; and deal with normal work stresses. However, the undersigned points out that the claimant functioned well and passed the GED test, attended college and obtained good grades, and worked selling Avon and part-time at a hotel. That’s it. Though the ALJ mentioned Dr. Radzeviciene’s opinion once more in passing, the judge never linked the activities he cited with an assessment of Dr. Radzeviciene’s opinion or explained whether that opinion supported a finding that Jelinek’s impairments met a listing. 14 No. 10-3340 The Commissioner does not defend the denial of benefits by asserting that the ALJ’s discussion of Dr. Radzeviciene’s opinion was adequate. Instead, in his brief and at oral argument, the Commissioner has asserted that the opinions of Dr. Kladder and Dr. Heroldt, the two state-agency psychologists, explained the ALJ’s rejection of Dr. Radzeviciene’s opinion. But the Commis- sioner’s say-so is not enough. We have made clear that what matters are the reasons articulated by the ALJ. Spiva, 628 F.3d at 353; Larson, 615 F.3d at 749. Although the ALJ cited the opinions of both psychologists, he did not use either opinion to support his decision to reject Dr. Radzeviciene’s opinion. See 20 C.F.R. § 404.1527(d)(2); Larson, 615 F.3d at 751; Moss v. Astrue, 555 F.3d 556, 561 (7th Cir. 2009). In any event, the ALJ would be hard-pressed to justify casting aside Dr. Radzeviciene’s opinion in favor of these earlier state-agency opinions. See Scott, 647 F.3d at 739-40. By 2008, the state-agency opinions were two years old. Dr. Radzeviciene’s opinion, on the other hand, was the most recent professional word on Jelinek’s mental im- pairments, by a treating psychiatrist who had seen her repeatedly over a two-year period with full access to her complete medical record to that point. No other medical opinion available to the ALJ provided a similarly comprehensive picture of Jelinek’s overall mental health at the time of the hearing. Neither the opinion of Dr. Heroldt nor the opinion of Dr. Kladder fully supported the ALJ’s wholesale rejection of Dr. Radzeviciene’s opinion. While these psychologists No. 10-3340 15 did not think that Jelinek’s residual functional capacity was as limited as Dr. Radzeviciene did, Dr. Heroldt scored Jelinek’s GAF at 50, which corresponds to a serious impairment of functioning, and Dr. Kladder concluded, contrary to the ALJ’s decision, that Jelinek’s claim of disability was credible. Even if we could follow the ALJ’s reasoning from the brief references to Jelinek’s college coursework and her limited employment, the ALJ’s decision did not build a logical bridge between those activities and his conclu- sion that she had not met a listing after her eighteenth birthday. See Clifford v. Apfel, 227 F.3d 863, 872 (7th Cir. 2000). An ALJ may consider a claimant’s daily activi- ties when assessing credibility, see Arnold v. Barnhart, 473 F.3d 816, 823 (7th Cir. 2007), but ALJs must explain perceived inconsistencies between a claimant’s activities and the medical evidence. Stewart v. Astrue, 561 F.3d 679, 684 (7th Cir. 2009); Carradine v. Barnhart, 360 F.3d 751, 755 (7th Cir. 2004); Zurawski v. Halter, 245 F.3d 881, 887 (7th Cir. 2001). The ALJ did not do so, and we are hard- pressed to understand how Jelinek’s brief, part-time employment supports a conclusion that she was able to work a full-time job, week in and week out, given her limitations. See Stewart v. Astrue, 561 F.3d 679, 684 (7th Cir. 2009); Diaz v. Prudential Ins. Co. of Am., 499 F.3d 640, 648 (7th Cir. 2007); Zurawski, 245 F.3d at 887. The ALJ did not ask the critical questions about Jelinek’s actual work hours or absentee rates in the jobs she held, and no medical provider or consultant opined that Jelinek could hold down a full-time position. Rather, the record suggests that Jelinek has experienced longstanding prob- 16 No. 10-3340 lem with absences. The activities the ALJ mentioned reflected only her willingness and ability to stay engaged in commendable but limited endeavors part- time or at her own pace. The remainder of the ALJ’s decision shows that the failure to weigh Dr. Radzeviciene’s opinion appropri- ately led to additional errors. If that opinion were fully credited, it supports both a finding that Jelinek met Listing 12.04 before the hearing date and that her residual functional capacity was consistent with a finding of disability. Dr. Radzeviciene noted that Jelinek exhibited several symptoms of depression, including appetite and sleep disturbance, psychomotor agitation or retarda- tion, decreased energy, and difficulty concentrating or thinking. Dr. Radzeviciene’s notes also reflect that Jelinek was bipolar. A finding that she suffered from these ill- nesses would satisfy Listing 12.04’s diagnostic “A criteria.” See 20 C.F.R. Pt. 404, Subpt. P, App. 1 § 12.04(A); Larson, 615 F.3d at 747-48. Dr. Radzeviciene also opined that Jelinek was experiencing “frequent” deficiencies of con- centration, persistence, or pace and “repeated” episodes of deterioration or decompensation in work or work- like settings, together enough to meet Listing 12.04’s diagnostic “B criteria.” See 20 C.F.R. Pt. 404, Subpt. P, App. 1 § 12.04(B); Larson, 615 F.3d at 747-48; Craft v. Astrue, 539 F.3d 668, 674-75 (7th Cir. 2008). These two findings — that Jelinek satisfied both the A and B criteria of Listing 12.04 — would together compel a finding that Jelinek was disabled. See 20 C.F.R. Pt. 404, Subpt. P, App. 1 § 12.00. But the ALJ’s opinion did not mention any of the symptoms cited in Dr. Radzeviciene’s 2008 opinion. In fact, though the quoted text appears in the section of No. 10-3340 17 the ALJ’s opinion devoted to whether Jelinek “met or medically equaled” a listed impairment after the age of eighteen, the section devoted to Dr. Radzeviciene’s opinion cited only the portion devoted to Jelinek’s residual functional capacity, not her symptoms or diag- noses. As noted, Jelinek also contends on appeal that the ALJ’s failure to analyze Dr. Radzeviciene’s opinion led the judge to omit key limitations when presenting his hypothetical questions to vocational expert Fisher at Jelinek’s hearing. As a result, Jelinek argues, the expert could not give the ALJ an accurate picture of the jobs available to her in the national economy. At the very least, Jelinek contends, the ALJ should have included hypothetical questions addressing issues of “concentra- tion, persistence, or pace” consistent with the medical opinions that he credited, including Dr. Kladder’s. Again, we agree. We have stated repeatedly that ALJs must provide vocational experts with a complete picture of a claimant’s residual functional capacity, and vocational experts must consider “deficiencies of concentration, persistence, and pace.” O’Conner-Spinner v. Astrue, 627 F.3d 614, 619 (7th Cir. 2010); see Stewart, 561 F.3d at 684; Young v. Barnhart, 362 F.3d 995, 1004 (7th Cir. 2004); Kasarsky v. Barnhart, 335 F.3d 539, 544 (7th Cir. 2003); Steele, 290 F.3d at 942. And though the hypothetical questions posed by an ALJ to a vocational expert must include only the physical and mental limitations the judge deems credible, Schmidt, 496 F.3d at 846, the ALJ did not do that in this case. 18 No. 10-3340 The ALJ limited his questioning of the expert to “seden- tary” and “light” unskilled work. But “sedentary” and “light” both describe a claimant’s ability to exert herself physically over a workday or workweek. See 20 C.F.R. §§ 404.1567, 416.967(b); Social Security Ruling (“SSR”) 83- 10, 1983 WL 31251, at *5-6 (1983); Haynes v. Barnhart, 416 F.3d 621, 627 n.2 (7th Cir. 2005). Similarly, “unskilled work” is defined by regulation as “work which needs little or no judgment to do simple duties that can be learned on the job in a short period of time.” 20 C.F.R. § 404.1568(a). None of these terms addressed the impact of the mental limitations reflected in Dr. Radzeviciene’s opinion, which (as reflected in the above quotation from the ALJ’s decision) limited Jelinek’s ability to main- tain regular work attendance, to carry out instructions, and to deal with the stresses of full-time employment. What’s more, Dr. Kladder’s earlier opinion from 2006, which the ALJ largely credited, concluded that Jelinek was not a malingerer and that she suffered at least “mod- erate limitations” in her abilities to concentrate, to com- plete a normal workday or workweek, and to respond appropriately to criticism from supervisors. At the least, the ALJ was required to pose hypothetical questions to the vocational expert consistent with Dr. Kladder’s opinion (and with those of the other physicians on whose opinions he relied) to give the expert a complete picture of Jelinek’s residual functional capacity. These are reasons enough to remand the matter to the agency for further review. Nevertheless, we briefly address as well Jelinek’s final argument — that the ALJ No. 10-3340 19 failed to conduct a proper credibility analysis — to point out a few additional flaws that should be avoided on remand. Foremost among these is the ALJ’s repeated reference to Jelinek’s “medication non-compliance” as a reason for finding her not credible. The ALJ ap- parently concluded that Jelinek’s symptoms would have remained under control but for an unwillingness to take her medications as directed. But we have often observed that bipolar disorder, one of Jelinek’s chief impairments, is by nature episodic and admits to regular fluctuations even under proper treatment. ALJs assessing claimants with bipolar disorder must consider possible alternative explanations before concluding that non-compliance with medication supports an adverse credibility inference. See Punzio v. Astrue, 630 F.3d 704, 710 (7th Cir. 2010); Larson, 615 F.3d at 751; Bauer v. Astrue, 532 F.3d 606, 609 (7th Cir. 2008); Kangail v. Barnhart, 454 F.3d 627, 630-31 (7th Cir. 2006). Here, the record shows a litany of changes to Jelinek’s medications over the years, as well as concerns by her doctors over side effects, ineffective drugs, costs, insur- ance issues, and compliance issues due to both mental illness and the potential lack of family and social sup- port. Early notes from Jelinek’s visits to the Bowen Center expressed concern that her mother might not have fully supported her medication regimen. And at various times Jelinek advised doctors that she did not have insurance and was concerned about paying for therapy — a cost con- cern that could have limited her access to her several prescribed medications. Treatment notes did not al- 20 No. 10-3340 ways reflect why medication changes were made, but several notes, including notes from as late as Decem- ber 2007, show that doctors were concerned that Jelinek’s medications were not optimally treating her symptoms and that she would sometimes run out of her medica- tions. These concerns must be addressed as part of any consideration of Jelinek’s failure to comply with prescribed medication. We R EVERSE the district court’s judgment and R EMAND to the Social Security Administration for further pro- ceedings consistent with this opinion. 11-7-11