Summary
finding reversible error when ALJ did not give appropriate reasons for weighing opinions from nurse practitioner
Summary of this case from Stephanie T. v. SaulOpinion
No. 14–2303.
03-26-2015
Kevin VOIGT, Plaintiff–Appellant, v. Carolyn W. COLVIN, Acting Commissioner of Social Security, Defendant–Appellee.
Dana Wayne Duncan, Attorney, Duncan Disability Law, SC, Wisconsin Rapids, WI, for Plaintiff-Appellant. David Levitt, Attorney, Social Security Administration Office of the General Counsel, Region V, Chicago, IL, Richard D. Humphrey, Attorney, Office of the United States Attorney, Madison, WI, for Defendant-Appellee.
Dana Wayne Duncan, Attorney, Duncan Disability Law, SC, Wisconsin Rapids, WI, for Plaintiff-Appellant.
David Levitt, Attorney, Social Security Administration Office of the General Counsel, Region V, Chicago, IL, Richard D. Humphrey, Attorney, Office of the United States Attorney, Madison, WI, for Defendant-Appellee.
Before POSNER, KANNE, and TINDER, Circuit Judges.
Opinion
POSNER, Circuit Judge.
The plaintiff had applied to the Social Security Commission in 2009 (when he was 40 years old) for benefits to which he claimed to be entitled by reason of being disabled from gainful employment as a result of psychiatric disorders (primarily depression and bipolar disorder ), chronic back and hip pain, and an anal fissure (cut or tear). The administrative law judge to whom his application was referred (John H. Pleuss) denied his claim on the ground that he's capable of performing unskilled sedentary work and is therefore not totally disabled. The district court, to which the applicant turned, upheld the denial of benefits, precipitating this appeal.
Voigt had been trained as a machinist, and until 2002 (the claimed onset date of his total disability) had worked intermittently as a machinist and as an assembly-line worker, jobs that the administrative law judge agreed he was no longer capable of doing, because of his physical and mental problems. Between 2001 (possibly earlier) and 2008, Voigt had taken prescription antidepressant medications such as Paxil, but he quit taking them because of their adverse side effects.
In the fall of 2009, having abandoned the antidepressant medications, he sought the help of “crisis workers” at a mental health clinic. The intake report of his visit to the clinic summarizes his confused and rather wild description of his mental state. In a subsequent visit to the clinic he told the crisis worker who interviewed him that he thought it might be good for him to be in prison, where he might (he thought—we know not on what basis) get some additional experience as a machinist and earn money that he could save. Yet he also told that same worker in a later interview that his goal was to own a restaurant, which was and is both unrealistic given his mental condition and irrelevant to improving his skills as a machinist.
He was examined at the clinic by “an advanced practice psychiatric nurse” (see APNA, “What Is an Advanced Practice Psychiatric Nurse?” www.apna.org/i4 a/pages/index.cfm?pageid=3866# 1, visited March 15, 2015, as were the other websites cited in this opinion). The nurse, Debra Day, diagnosed him as suffering from depression (no surprise), but two years later (after his eighth examination by her), she submitted a report to the Social Security Administration in which she described Voigt as bipolar (oddly she did not mention depression, though of course depression is an aspect of bipolar disorder, which used to be called “manic depression ”) and opined that his mental illnesses would cause him to miss work more than four days each month—which the vocational experts on whom the administrative law judges rely testify disqualifies a person from gainful employment. Garcia v. Colvin, 741 F.3d 758, 760 (7th Cir.2013) ; Pepper v. Colvin, 712 F.3d 351, 361 (7th Cir.2013) ; Treichler v. Commissioner of Social Security Administration, 775 F.3d 1090, 1096 (9th Cir.2014) ; Ghanim v. Colvin, 763 F.3d 1154, 1159 (9th Cir.2014). To qualify for gainful employment one must be able to work on a “sustained basis,” defined as eight hours a day five days a week, see 20 C.F.R. §§ 404.1512(a), 416.912(a) ; Rollins v. Massanari, 261 F.3d 853, 859 (9th Cir.2001) ; SSR 96–8p, “Purpose,” ¶ 1, and to be incapable of gainful employment is to be totally disabled within the meaning of the Social Security Act. 42 U.S.C. §§ 423(d)(1)(A), 1382c(a)(3)(A). To miss four workdays a month would reduce one's average workweek from five to four days, which would not constitute working on a sustained basis as defined by the Commission.
Day's report listed a total of 13 symptoms exhibited by Voigt of poor psychological and social functioning, ranging from paranoia to “oddities of thought, perception, speech or behavior.” She deemed him “unable to meet competitive standards” (requirements for gainful employment) of punctuality, of “sustain[ing] an ordinary routine without special supervision,” of “work [ing] in coordination with or proximity to others without being unduly distracted,” and of being able to “complete a normal workday,” “accept instructions,” “get along with coworkers,” and “deal with normal work stress.” (We omit five other requirements of gainful employment that Nurse Day deemed Voigt unable to satisfy.)
At her first examination of him, in October 2009, Day gave him a GAF score of 50. “GAF” stands for Global Assessment of Functioning, and a score of between 41 and 50 signifies serious psychiatric illness. The American Psychiatric Association has since eliminated the GAF scale from its Diagnostic and Statistical Manual of Mental Disorders as being unreliable—but this occurred after the administrative law judge issued his decision, which was in January 2012. (The length of time it's taken the case to get to us is lamentable.)
Day prescribed an antidepressant medicine called Cymbalta. Voigt reported improvement in his mental states and absence of the side effects that he'd experienced with the antidepressant medications that he had been taking previously. On the basis of that report, Day raised Voigt's GAF score to 55. That brought it into the range of “moderate symptoms” or “moderate difficulty in social, occupational, or school functioning (e.g., few friends, conflicts with peers or coworkers).” He continued making progress, and after examining him again, Day raised his GAF score to 65, signifying “mild symptoms” and “generally functioning pretty well.” During two of their sessions she gave him a score of 70. GAF scores bounce around a great deal, however, because they depend on how the patient happens to feel the day he's examined. See Punzio v. Astrue, 630 F.3d 704, 710 (7th Cir.2011) ; I.H. Monrad Aas, “Guidelines for Rating Global Assessment of Functioning (GAF),” Annals of General Psychiatry 10:2, pp. 4–5 (2011). Overall Voigt's GAF score rose from 50 at his first examination by her to 65 at the last one, consistent with her treatment notes, which state that Voigt reported that the medication was working, enabling him to control his irritation and other negative emotions better. It's not surprising that his mood improved over the course of his visits, as he got to know her better and the therapy she prescribed took effect. His GAF scores were computed anew at each visit; what they would have been had they been computed elsewhere by a practitioner whom he was being examined by for the first time is unknown. The critical question, however, was whether the medication that Day prescribed so improved his mental health as to enable him to qualify for a full-time job. She thought not.
Meanwhile he'd been visiting another clinic because of physical distress that included the anal fissure, the back and hip pain (he described the hip pain as “sharp” and “stabbing”), and hemorrhoids (possibly related to the fissure). An osteopath designated by the Social Security Commission to examine Voigt after he applied for disability benefits reported that he suffers from lower back pain and trochanteric bursitis (an inflammation near the hip that causes pain in the hip, see Cleveland Clinic, “Trochanteric Bursitis,” http:// my.clevelandclinic.org/health/diseases_conditions/hic_Bursitis/hic_ Trochanteric_Bursitis). The osteopath also observed that Voigt walked with a “slow and painful gait.” Another physician and a psychologist examined Voigt's medical records and concluded that despite his physical and psychiatric problems he was capable of performing unskilled sedentary work. Neither examined him, however.
After many months of gradual improvement, Voigt complained to Day, and to another licensed professional counselor as well, of a deterioration in his mental condition. He reported “raging in the parking lot” of the mental health clinic and becoming “very angry and upset” at a Super Bowl game. Day left Voigt's GAF score at 65, but the other counselor reduced it to 51 to 60. (The administrative law judge erred in stating that Voigt's lowest GAF score was 55.) This discrepancy in their scores is consistent with the probability noted earlier in this opinion that one's GAF score will rise as one builds up a relationship with the practitioner who is scoring him.
Day believed that Voigt's deterioration was attributable in part to his drinking, taking Vicodin (a pain-relieving drug), and smoking marijuana, and that these behaviors were retarding his progress in therapy. But while they were a big concern, she didn't think they played the primary role in his problems. (This is important, since if Voigt's deterioration were attributable solely to drug and alcohol abuse he might well be barred from obtaining benefits no matter how serious his disability. See 42 U.S.C. § 423(d)(2)(C) ; Kangail v. Barnhart, 454 F.3d 627, 628 (7th Cir.2006).)
The administrative law judge went far outside the record when he said that if Voigt were as psychologically afflicted as Day thought, he “would need to be institutionalized and/or have frequent inpatient treatment”—a medical conjecture that the administrative law judge was not competent to make, see Browning v. Colvin, 766 F.3d 702, 705 (7th Cir.2014) ; Goins v. Colvin, 764 F.3d 677, 680 (7th Cir.2014) ; Pate–Fires v. Astrue, 564 F.3d 935, 946–47 (8th Cir.2009), and that was implausible to boot. The institutionalization of the mentally ill is generally reserved for persons who are suicidal, otherwise violent, demented, or (for whatever reason) incapable of taking even elementary care of themselves. Voigt is none of these things, and so cannot be institutionalized against his will: “a State cannot constitutionally confine [against his will] without more a nondangerous individual who is capable of surviving safely in freedom by himself or with the help of willing and responsible family members or friends.” O'Connor v. Donaldson, 422 U.S. 563, 576, 95 S.Ct. 2486, 45 L.Ed.2d 396 (1975). But it doesn't follow, as the administrative law judge may have thought, that Voigt is therefore capable of gainful employment.
The judge remarked that because Voigt's “parents apparently pay his rent and expenses ..., these facts raise a question as to whether [his] continuing unemployment is actually due to medical impairments.” No, it doesn't, because if he can't work, he has to be supported by someone or by some agency. More to the point is the administrative law judge's remark that Voigt had had only sporadic employment in the decade preceding the claimed onset of his disability in 2002. He may have been a malingerer, mooching off his parents, or he may have been more psychologically disturbed than anyone realized. In any event his parents may be continuing to support him because whatever his physical and mental state in the decade ending in 2002 he may now no longer be capable of working.
The administrative law judge also contradicted himself about whether Voigt is currently a malingerer when later in his opinion he remarked that Voigt had “made statements during treatment suggesting he actually is eager to work, but is unable to find work, which suggests that the claimant is not unemployed due to disability.” Actually those statements were consistent with his wanting to lead a normal life yet being unable to land a job because he's disabled from gainful employment by a combination of physical and mental problems that a prospective employer would quickly notice. Voigt acknowledged that having been unemployed for such a long time makes it harder for him to get a job, but that's not inconsistent with his being disabled. A disabled person may want to work, may seek work, and in some cases may land work. We've noted cases in which although the claimant is not only working but also “earning a decent wage, he really is permanently disabled from engaging in gainful activity. Maybe his boss feels desperately sorry for him and is retaining him on the payroll even though he is incapable of working. That act of charity ought not be punished by denying the employee benefits and thus placing pressure on the employer to retain an unproductive employee indefinitely. Maybe a seriously disabled worker is able to work only by dint of his extraordinary determination and the extraordinary assistance extended to him by kindly fellow workers.” Jones v. Shalala, 21 F.3d 191, 192 (7th Cir.1994) (citations omitted).
The only other witness at the hearing was a vocational expert, who testified on the basis of what the administrative law judge found that Voigt could and could not do that while he could no longer work as a machinist he could do simple sedentary work such as that of an office assistant (whatever that means), security guard (very doubtful—a security guard with significant mental health problems is likely to be a danger to himself and others), an assembler, or a packager. The vocational expert said that there are 23,000 such jobs in Wisconsin (where Voigt lives), but did not explain where he'd gotten that figure. He added that anyone who would miss work two or more days per month (not just three or more) more than occasionally would not qualify for gainful employment.
Although concluding that Voigt was not totally disabled, the administrative law judge characterized Voigt's trochanteric bursitis, fissure, depression, and bipolar disorder as severe, while discounting his back pain on two grounds. One was that “he ha[d] not taken any narcotic based pain-relieving medications.” He had, however, taken Vicodin (albeit illegally), a powerful pain reliever. The other ground was Voigt's failure to undergo “intensive treatment[,] like injections, which would be expected [to be prescribed for] a person experiencing disabling pain.” Injecting steroids or other drugs is sometimes suggested for treating back pain, see Web MD, Back Pain Health Center, “Injections for Back Pain Relief,” www.webmd. com/backpain/guide/back-pain-injection-treatments, but the administrative law judge offered no reason for thinking that it would have been appropriate for Voigt. Nor did he note the natural reluctance of a person with psychiatric problems (perhaps of any person) to take powerful pain medications, as they can have serious side effects if not carefully used. A mentally ill person is more likely to abuse opiates like Vicodin (see National Alliance on Mental Illness, “Opiates and Mental Illness,” www2.nami.org/Content/NavigationMenu/Hearts_and_ Minds/Smoking_Cessation/Opiate_Abuse_and_Mental_Illness.htm) than a healthy person. In addition, opiates can increase symptoms of bipolar disorder. See La Hacienda Treatment Center, “Opiates,” www.lahacienda. com/resources/articles/opiates/; cf. Beardsley v. Colvin, 758 F.3d 834, 840 (7th Cir.2014) ; Myles v. Astrue, 582 F.3d 672, 677–78 (7th Cir.2009) ; Craft v. Astrue, 539 F.3d 668, 679 (7th Cir.2008).
At the hearing Voigt described some of his psychological symptoms but also emphasized health problems resulting from his anal fissure, including bleeding, and poor bowel control that requires him to spend 40 consecutive minutes on the toilet about four times every month. The administrative law judge discounted these problems by noting Voigt's refusal to undergo a colonoscopy between 2006 and 2008 and a rectal examination in 2010, but failed to note that Voigt testified that he'd been told that he would have to have surgery to repair the fissure and that he was fearful of having a colonoscopy —a common fear, known as “colonoscopy jitters.” See, e.g., Riverside, “Got the Colonoscopy Jitters?” www.riversideonline.com/services/ cancer /colon/colonoscopy-jitters.cfm; cf. Beardsley v. Colvin, supra, 758 F.3d at 840 ; Craft v. Astrue, supra, 539 F.3d at 679 ; Orn v. Astrue, 495 F.3d 625, 638 (9th Cir.2007).
The administrative law judge discounted Voigt's mental problems on the ground that they had been solved by his taking Cymbalta. That was contrary to Day's report, however, which said that his mental state continued to be disturbed, and though it was exacerbated by his substance abuse she did not consider that the primary cause of his inability to relate to people in a normal way. The administrative law judge gave “very little weight” to Day's report, however, on the ground that a mere nurse is not an “acceptable medical source,” and that Day had seen Voigt on only a few occasions and had given him a pretty high GAF score, and further that as mentioned earlier “if the claimant were as limited as indicated in [Day's] opinion, the claimant would need to be institutionalized and/or have frequent inpatient treatment. These inconsistencies render the opinion less persuasive, and it is due very little weight.”
These were not adequate reasons for rejecting Day's report and therefore turning down Voigt's application for disability benefits. As for Day's not being an “acceptable medical source,” the administrative law judge failed to note that “evidence from other sources,” including nurse practitioners, may be used to “show the severity of [the applicant's] impairment(s) and how it affects [his or her] ability to work.” 20 C.F.R. §§ 404.1513(a), (d), (d)(1), 416.913(a), (d), (d)(1) ; see also SSR 06–3p. As for the number of visits, Day had examined Voigt eight times in two years and the administrative law judge did not say that the visits were too infrequent to have enabled Day to assess the trend (first up, then down) in Voigt's condition. Most questionable of all is the judge's statement that if Day's description of Voigt's symptoms were accurate he would have to be institutionalized or hospitalized (if hospitalization is what the administrative law judge meant by “inpatient treatment”). Cymbalta may have enabled Voigt to keep out of mental or other hospitals; the question is whether it enables him, despite acute difficulties that Cymbalta has not dispelled in dealing with other human beings, to obtain gainful employment.
The administrative law judge committed the further error of thinking that how one uses his time at home is compelling evidence of whether or not one is employable. See Beardsley v. Colvin, supra, 758 F.3d at 838 ; Bjornson v. Astrue, 671 F.3d 640, 647 (7th Cir.2012) ; Garrison v. Colvin, 759 F.3d 995, 1016 (9th Cir.2014). Voigt apparently spends much of his time at a computer conducting research on his medical problems and playing video games. The administrative law judge cited treatment notes, and testimony, that indicated that Voigt is “able to persist at activities like online research,” and enjoys “playing computer games—for hours.” But whether Voigt's online research or playing such games is inconsistent with his having severe back pain would depend on the precise nature of the pain, how comfortable the chair he sits on while using the computer is, and what his alternatives are—he can't spend all day standing or lying down even if sitting is painful. These issues were not explored.
Nor did the administrative law judge explain his assumption that doing limited online research or playing video games online requires the same concentration as is required for full-time employment. In Powers v. Apfel, 207 F.3d 431, 435 (7th Cir.2000), we expressed skepticism about an administrative law judge's assertion that “ability to watch television for several hours indicates a long attention span.” Playing a video game is less passive, hence more intellectually challenging, than watching television. But the administrative law judge did not consider the extent to which experience in playing video games would equip Voigt for any of the jobs that he might otherwise be incapable of performing.
And finally the administrative law judge fully credited the vocational expert's testimony. He did this even though the testimony did not explain the source of the expert's estimate of the number of jobs that Voigt could perform, and even though the inclusion of “security guard” among those jobs should have nudged the administrative law judge to explore the expert's basis for thinking that someone with Voigt's psychiatric problems would be able to perform such a job without danger to himself and others. Nurse Day had noted Voigt's “problems with irritability and [with his] sometimes explosive behavior.” Among the people he doesn't get along with are police officers—which doesn't augur well for success as a security guard.
The administrative law judge said that “the vocational expert's testimony is consistent with the information contained in the Dictionary of Occupational Titles,” but as we and others have explained, the DOT does not contain information on which to base an estimate of the number of available jobs of a particular kind. Browning v. Colvin, supra, 766 F.3d at 709 ; Herrmann v. Social Security Administration, 772 F.3d 1110, 1113–14 (7th Cir.2014) ; Brault v. Social Security Administration, Commissioner, 683 F.3d 443, 446–47 (2d Cir.2012) (per curiam); Guiton v. Colvin, 546 Fed.Appx. 137, 143–45 (4th Cir.2013) (concurring opinion); Coppernoll v. Astrue, 2009 WL 1773132, at *8 (W.D.Wis. June 23, 2009) ; Jon C. Dubin, “Overcoming Gridlock: Campbell After a Quarter–Century and Bureaucratically Rational Gap–Filling in Mass Justice Adjudication in the Social Security Administration's Disability Programs,” 62 Administrative L.Rev. 937, 964–71 (2010); Peter J. Lemoine, “Crisis of Confidence: The Inadequacies of Vocational Evidence Presented at Social Security Disability Hearings (Part II),” Social Security Forum, Sept. 2012, p. 1.
As we said in the Browning case, we “have no idea what [is] the source or accuracy of the number of jobs that vocational experts (including the one in this case, whose estimates the administrative law judge accepted without comment) claim the plaintiff could perform that exist in the plaintiff's area, the region, or the nation. There is no official source of number of jobs for each job classification in the Dictionary of Occupational Titles, and while there are unofficial estimates of jobs in some categories, the vocational experts do not in general, and the vocational expert in this case did not, indicate what those data sources are or vouch for their accuracy. And many of them estimate the number of jobs of a type the applicant for benefits can perform by the unacceptably crude method of dividing the number of jobs in some large category (which may be the only available data) by the number of job classifications in the category, even though there is no basis for assuming” that there is the same number of jobs in each narrow category. 766 F.3d at 709 (emphasis added). There is no indication that the estimate of the number of jobs that the applicant in this case could fill (if the administrative law judge's estimate of his capacity to work is correct, as it may very well not have been) is any more accurate than it was in Browning.
The judgment of the district court is reversed and the case remanded to that court with instructions to remand it to the Social Security Administration for further proceedings consistent with this opinion. We do not say that Voigt is in fact totally disabled from gainful employment, however—only that he's entitled to a more careful analysis of his claim by the Social Security Administration.