Summary
holding that the ALJ impermissibly "played doctor" when he reached his own independent medical conclusion in finding that Plaintiff was not disabled because no doctor prescribed insulin
Summary of this case from Errica L. v. Comm'r, Soc. Sec. Admin.Opinion
No. 08-2908.
Argued April 22, 2009.
Decided September 9, 2009.
Appeal from the United States District Court for the Northern District of Illinois, Eastern Division, Charles P. Kocoras, J.
Barry A. Schultz (argued), Evanston, IL, for Plaintiff-Appellant.
Edward J. Kristof (argued), Social Security Administration, Chicago, IL, for Defendant-Appellee.
Before MANION, KANNE, and SYKES, Circuit Judges.
Carolyn Myles, who suffers from type-2 diabetes, claims that she is disabled because of symptoms of that disease, and is seeking disability insurance benefits. Her claim was rejected by an administrative law judge. Myles argues before this court that the ALJ failed to consider all of the facts in the record and made improper medical and credibility determinations. The ALJ's opinion contains multiple errors, the cumulative effect of which is to leave us without confidence that the ALJ's decision builds a "logical bridge" between the evidence and his conclusion, and so we vacate and remand for further proceedings.
Myles has had diabetes since at least 2002. Since at least July 2004, her diabetes has been uncontrolled or poorly controlled. In August 2004, Myles saw a physician, Dr. Max Goldschmidt, and at that time, she reported suffering from diarrhea, blurred vision and seeing dots in front of her eyes, frequent urination, and headaches. Myles reported to the doctor that she had not been taking one of her medications, Metformin, on a daily basis, because it gave her diarrhea. Dr. Goldschmidt instructed her not to take it if she could not do so daily, and adjusted her dosage of another medication.
Myles applied for disability insurance benefits in January 2005, claiming that she was unable to work because of her diabetes. That same month she again saw Dr. Goldschmidt. She had been out of her medication for six days and said that she was fatigued. It appears that at this time Dr. Goldschmidt re-prescribed Metformin. Dr. Goldschmidt also signed a letter saying that Myles was unable to work for an "undetermined" period, although he did not give reasons. Dr. Goldschmidt signed several similar letters in January and early February.
At another appointment that January, Myles again complained of frequent urination, and Dr. Goldschmidt discovered she had a urinary tract infection, for which he prescribed an antibiotic. Myles continued to see Dr. Goldschmidt and other doctors regularly, and in March 2005, Dr. Goldschmidt noted that he might need to consider prescribing insulin.
In 2005 a state agency physician examined Myles in relation to her application. The agency physician, Dr. Kale, noted that she had a history of poorly controlled diabetes, polyuria (passage of more than 2.5 liters of urine every 24 hours), nocturia (need to get up at night to urinate), and occasional hand and foot numbness. The examination did not reveal any neurological problems with Myles's hands or feet, and her ability to grasp and grip was not impaired.
In April 2006, Myles complained of hair loss, and Dr. Lovinger at the Lake County Health Department ordered her to stop taking Metformin and to substitute Avandia, which seems to have stopped the hair loss. At that time, Dr. Lovinger noted that Myles was not checking her blood sugar levels regularly and that her diabetes remained uncontrolled.
In August 2006, Myles reported to Dr. Lovinger that she had been suffering from fatigue and muscle weakness. Dr. Lovinger found no neurological problem. Dr. Lovinger noted that Myles may need to start insulin, but at that time he did not prescribe insulin. Instead, he left her on her prior oral medications.
In November 2006, Myles returned to Lake County, complaining of depression, for which she was prescribed Zoloft. She complained again of pain in her feet and legs that month, and, although an examination revealed no motor deficits, she was diagnosed with neuropathy. The doctor also noted that she complained of polyuria at this time. Myles complained to doctors at Lake County of tingling in her fingers the next month.
In May 2007, Dr. James Sims, a physician at Lake County, completed a Medical Assessment of Condition and Ability to Do Work Related Activities at the request of Myles's attorney. Dr. Sims, who had been treating Myles for five years, opined that Myles could stand or walk six to eight hours uninterrupted and sit six to eight hours uninterrupted on "good" days. He also opined that she could lift 25 pounds occasionally and 10 pounds frequently, and prescribed no grasping limitations. But he further opined that Myles would have trouble completing a work day and work week without interruption from her symptoms, and that she could be expected to have "good" and "bad" days.
In June 2007, an administrative law judge held a hearing on Myles's application and found that she was not disabled. Analyzing Myles's claim under the five-step analysis of 20 C.F.R. § 404.1520(a), the ALJ found Myles had not engaged in gainful employment since her onset date; that her diabetes was severe; but that it did not meet or equal any of the impairments listed in Appendix 1, Subpart P, Regulation No. 4 of the Social Security regulations. The ALJ next determined Myles's Residual Functional Capacity, and found that she had marked limitations in her capacity to work and could not continue in any of her past jobs. In assessing her RFC, the ALJ found Myles not to be credible for several reasons: Myles's claims of urinary frequency were unbelievable because, the ALJ noted, she had not complained about them to a doctor since January 2005; she had exaggerated claims of weight loss due to medication, saying she had lost 40 pounds when she really only lost 18; and the ALJ stated that Myles had not complied with her treatment, rendering her claims of severe symptoms less credible.
A vocational expert testified that based on the ALJ's hypothetical questions, Myles could still work in a bench assembly, packager, tester, clerk, or cashier position, and that there were at least 7,500 such positions available. But, the VE added, a person with occasional numbness or tingling of the hands, even as little as a sixth of the day, would not be able to perform these jobs. Further, the VE testified, a person who needed a restroom break at least once an hour would not be able to maintain employment in those jobs. The ALJ determined that Myles did not suffer from hand limitations or frequent urination, found that she could still maintain employment, and denied her claim. Myles sought review from the Appeals Council, which declined to hear the case. The district court affirmed the ALJ's decision.
On appeal, Myles points to a number of errors made by the ALJ. Together, these errors serve to undermine the ALJ's determination that she was not disabled and persuade us that a remand is necessary. The strongest of these arguments is that the ALJ did not analyze key facts in regard to her symptoms, particularly in regard to urinary frequency and hand problems. The VE made it clear that if Myles's claims of urinary frequency or tingling in her hands were true, she could not maintain employment. The ALJ rejected both claims. But regarding urinary frequency, the ALJ ignored record evidence, and regarding Myles's complaints of hand limitations, the ALJ simply did not perform any analysis that we can see.
As to urinary frequency, Myles argues that the ALJ was factually wrong when he rejected her assertions that she had to use the restroom at least once an hour. The ALJ found that there had been no complaints of urinary frequency since Myles was treated for a urinary tract infection in early 2005. Myles argues that, in fact, she complained to doctors about urinary frequency later than that, and that an ALJ may not rely on a mistake of fact to reject a claimant's testimony. The ALJ stated that although, if true, Myles's claims that she needed to use the restroom at least once an hour or several times an hour would have rendered her unemployable, "[t]his is absolutely not believable in light of the fact that the claimant has not complained of having to use the bathroom frequently to medical personnel." This was a credibility finding. We will uphold an ALJ's credibility finding if the ALJ gives specific reasons for that finding, supported by substantial evidence. Moss v. Astrue, 555 F.3d 556, 561 (7th Cir. 2009). But the ALJ may not simply ignore evidence. Lopez ex rel. Lopez v. Barnhart, 336 F.3d 535, 540 (7th Cir. 2003).
The ALJ overlooked two complaints to doctors when he asserted that Myles had not complained of urinary frequency after early 2005. First, in June 2005, Myles complained to Dr. Kale, who noted that Myles complained of "polyuria." Second, Myles complained to the Lake County Health Department again in November 2006. The government urges that Myles's complaints did not dictate a finding that urinary frequency caused limitations. But it is not that the error requires a different finding; rather, the ALJ's basis for his credibility determination on this issue is wrong, and so the ALJ must reconsider the credibility determination in light of the evidence. See Allord v. Barnhart, 455 F.3d 818, 822 (7th Cir. 2006).
Myles also argues that the ALJ did not analyze her claims of fatigue and hand limitations in his opinion, as he was required to do. See Young v. Sect'y of Health and Human Servs., 957 F.2d 386, 393 (7th Cir. 1992); see also SSR 96-8p, *7. The ALJ acknowledged these complaints, but his analysis does not articulate his reasons for rejecting them, except to say there is no objective medical evidence to support them. However, an ALJ may not discredit testimony of pain solely because there is no objective medical evidence to support it. See Villano v. Astrue, 556 F.3d 558, 562 (7th Cir. 2009). Thus, the ALJ erred in rejecting these claims as well.
Myles further argues that another basis the ALJ used to determine that she was not credible was flawed because the ALJ did not explore it as thoroughly as Social Security Rulings require. She argues that the ALJ relied upon what he said was a failure to follow her treatment in finding her not to be credible. Myles argues that the ALJ should have considered her reasons for the instances when she did not take her medication or test her blood sugar. The ALJ, in his opinion, cited "an issue of compliance," regarding three instances between August 2004 and June 2007 where Myles did not keep up with her treatment. This conduct, the ALJ noted, "fails to suggest symptoms that are particularly troublesome for the claimant."
But the ALJ was required by Social Security Rulings to consider explanations for instances where Myles did not keep up with her treatment, and he did not do so. See SSR 96-7p, *7; see also Moss, 555 F.3d at 562. Inability to pay for medication or negative side effects from medication may excuse failure to pursue treatment. SSR 96-7p at *8. In one instance, the ALJ noted that Myles did not take Metformin daily. But Metformin caused Myles diarrhea and hair loss, and eventually her doctors instructed her to stop taking it. The ALJ does not explain why this was an invalid explanation. The ALJ also notes that in one instance Myles was out of medication for six days, and in another she reported not testing her blood glucose regularly. But Myles was on public aid, and at least at one point, her blood glucose test strips were not covered. Again, the ALJ failed to consider whether this was a valid explanation for these isolated incidents. We will remand an ALJ's determination that lacks adequate discussion of the issues. Villano, 556 F.3d at 562.
Myles next argues that the ALJ diminished the severity of her symptoms by drawing his own, unsupported medical inference as to her treatment. She is correct. The ALJ decided, absent any medical evidence, that Myles's condition was less serious because it was treated only with oral medication and not with insulin therapy.
The government argues that Myles failed to present this argument to the district court, and so waived it. But the argument is preserved for review. Myles argued to the district court that the ALJ "diminished the severity of Ms. Myles' condition by his statement that `the Claimant does not even take insulin.'"
The ALJ impermissibly "played doctor" and reached his own independent medical conclusion when he determined that "[t]he level of treatment received also fails to infer limitations beyond the limitations described above in this decision. The claimant does not even take insulin." See Blakes ex rel. Wolfe v. Barnhart, 331 F.3d 565, 570 (7th Cir. 2003); see also Schmidt v. Sullivan, 914 F.2d 117, 118 (7th Cir. 1990) ("Common sense can mislead; lay intuitions about medical phenomena are often wrong."). He also stated that "[t]he claimant's own doctors do not indicate significant problems at this point, as they do not even prescribe insulin medication." But no doctor gave any reason why insulin was not prescribed. The inference that it was not prescribed because Myles was not experiencing significant problems appears to be the ALJ's own inference, and is wholly unsupported by the record.
Myles further argues that the ALJ impermissibly analyzed only the evidence in her treating physician's report which supported his ultimate conclusion while ignoring the evidence that undermined it. The ALJ accepted portions of the report of her treating physician, Dr. Sims, but failed to address the report's conclusion that Myles could not engage in sustained activity because of her symptoms.
The government argues that Dr. Sims is not a treating physician, because the only evidence in the record that supports his claim that he has treated Myles is his own assertion that he has done so since 2002. But the record contains evidence, such as lab work from 2006, that shows Dr. Sims treated Myles. Furthermore, Dr. Sims is a physician at the Lake County Health Department, where Myles has been receiving treatment for years.
The ALJ failed to analyze portions of Dr. Sims's report that stated that Myles had marked limitations in her abilities "to perform at a consistent pace without an unreasonable number and length of rest periods," and "to complete a normal workday and workweek without interruptions." An ALJ may not selectively consider medical reports, especially those of treating physicians, but must consider "all relevant evidence." See Clifford v. Apfel, 227 F.3d 863, 871 (7th Cir. 2000); Books v. Chater, 91 F.3d 972, 979 (7th Cir. 1996). It is not enough for the ALJ to address mere portions of a doctor's report. Godbey v. Apfel, 238 F.3d 803, 808 (7th Cir. 2000). The ALJ attempted to dispose of Dr. Sims's report by saying that the report "contains limitations that are far less restrictive" than the ALJ's determination. The ALJ noted that Dr. Sims stated that Myles could be expected to have "good" and "bad" days, but dismissed this conclusion by stating that the doctor did not state how many bad days Myles had. He does not appear to have inquired further into how often these "bad" days might occur. Although the ALJ did find lifting and postural restrictions more significant than Dr. Sims recommended, he failed to address Dr. Sims's conclusions about interruptions in the workday altogether.
In light of the ALJ's questionable credibility findings, cursory analysis of symptoms, improper medical determination regarding medication, and selective discussion of the evidence, his determination that Myles is not disabled is not supported by substantial evidence. The record does not command a determination that Myles should be awarded benefits, but the ALJ has not adequately supported his conclusions. We VACATE and REMAND the decision of the district court. On remand, the ALJ should consider all of the evidence in the record, and, if necessary, give the parties the opportunity to expand the record so that he may build a "logical bridge" between the evidence and his conclusions.