Summary
finding that the ALJ improperly considered a claimant's past ability to work without considering the claimant's testimony that his pain grew progressively worse until he could not work
Summary of this case from Dunn v. ColvinOpinion
No. 07-2206.
Argued April 22, 2008.
Decided September 8, 2008.
Appeal from the United States District Court for the Southern District of Indiana, Indianapolis Division. No. 06 C 194. David F. Hamilton, Chief Judge.
Marcie E. Goldbloom, Daley, Debofsky Bryant, for Plaintiff-Appellant.
Edward P. Studzinski, Social Security Administration, Chicago, IL, for Defendant-Appellee.
Before KENNETH F. RIPPLE, Circuit Judge, TERENCE T. EVANS, Circuit Judge, and ANN CLAIRE WILLIAMS, Circuit Judge.
ORDER
Shawn Gaylor injured his back in 1996 and has since undergone four surgical procedures and has been treated for depression and anxiety. He stopped working in January 2001, claiming that he was in too much pain to continue. Numerous doctors have treated Gaylor and have reviewed his medical records. Some of them conclude that he is unable to work, while others say that he can perform limited work. In denying Gaylor's claim for Social Security disability benefits, the ALJ found that Gaylor's back injury was not severe enough to qualify for coverage and that Gaylor's account of his pain was exaggerated and not credible. The ALJ also found that Gaylor's mental impairments were severe but did not foreclose all avenues of employment. The Appeals Council affirmed, and the district court upheld the decision of the ALJ. Because the ALJ's credibility determination and functional capacity determination were not supported by substantial evidence, however, we reverse and remand.
I. BACKGROUND A. Gaylor's claim of disability based on his back injuries
Gaylor, currently age 38, first experienced back problems in 1996 when he felt a sharp "pop" in his lower back while working for Chrysler. Orthopedic surgeon Dr. Michael Coscia has been treating Gaylor since 1998. Early on Dr. Coscia diagnosed dehydration and degeneration in Gaylor's spine as well as herniation and an annular tear. He noted that by 1999 Gaylor had already undergone two epidural steroid injections with no pain relief. Dr. Coscia referred Gaylor to Dr. John McLimore to manage Gaylor's pain. Dr. McLimore prescribed some pain medications when surgical procedures failed to relieve Gaylor's symptoms, but he also attempted to wean Gaylor off of the medication, noting that he was at risk of dependency.
Gaylor also saw Dr. Chakropani Gupta in 1999 for a lumbar discogram, which confirmed that Gaylor had significant disk degeneration with a total disruption of the nucleus and "multiple, concentric annular tears, and [an] anterior and a posterior annular tear" causing severe lower back pain and right buttock and leg pain. In late 1999, Dr. Gupta also performed an invasive procedure called intradiscal electrothermal therapy, which resulted in about a 50% improvement in Gaylor's pain. But this relief lasted only for about three months, when his pain became significantly worse. This concluded Dr. Gupta's treatment of Gaylor.
In early 2000, Dr. Coscia performed several surgical procedures on Gaylor "[d]ue to positive MRI scan findings and [a] failure of extensive nonoperative surgical therapy." Specifically, Dr. Coscia performed a hemilaminotomy, foramintomy, diskectomy, and a canal and nerve root decompression procedure. But after a month, Gaylor noted continued pain, and in mid-2001 Gaylor was again reporting severe pain. Dr. Coscia attempted dietary supplements to restore Gaylor's damaged cartilage but noted that the final alternative would be a spinal fusion. Also in 2001 Gaylor's family physician at the time, Dr. Phillip Whitfield, confirmed the danger of pain medication addiction identified by Dr. McLimore but noted that no other options were available to help Gaylor cope. Gaylor testified that he cooperated with the doctors' attempt to wean him from pain medication, but that the attempt was unsuccessful because the resultant pain was too severe and he was totally incapacitated.
By the end of 2002, a "persistent severe" pain prompted Dr. Coscia to perform a spinal fusion at two levels in Gaylor's back. Afterward, Gaylor experienced about two weeks of relief but then returned to taking large quantities of pain medication and reporting that his degree of pain was unchanged. Despite these reports, Dr. Coscia noted that the fusion had been a success and that there was no sign of any complications. He concluded that Gaylor was deconditioned to pain and needed to be weaned from his medication.
Through 2003, Gaylor was still reporting "dramatic disabling" pain to Dr. Coscia, but Dr. Coscia concluded that Gaylor did not suffer from a long-term disability. In early 2003 Dr. McLimore opined that Gaylor seemed more comfortable than the degree of pain he claimed to experience.
Dr. Eric Heathers, who had treated Gaylor as his family physician in 1999 and 2000 before Dr. Whitfield, began seeing him again in 2003. Dr. Heathers confirmed Gaylor's injuries and that he experienced a limited range of motion, limited reflexes, and limited strength. By late 2003, Dr. Heathers reported that he had seen Gaylor thirty-six times since 1998 and that his symptoms had not improved despite surgery. Moreover, he found that Gaylor's pain was consistent with the injuries identified in his back. He determined that Gaylor experienced persistent pain, spasms, loss of motion, loss of strength, and loss of sensation, and that Gaylor's condition required him to change position more than once every two hours. Contrary to Dr. Coscia, Dr. Heathers determined that Gaylor's symptoms would interfere with his ability to maintain reliable attendance at work and that, at best, he could work three days a week with some limitations.
Gaylor also saw Dr. Lynette Green-Mack, who specializes in spine rehabilitation, beginning in 2003. She diagnosed him with chronic pain secondary to failed back, post lumbar laminectomy syndrome, lumbar facet syndrome or lumbosacral spondylosis, and sacroiliitis, and she prescribed medication for persistent pain. As surgery and six epidurals had failed to decrease Gaylor's pain, she attempted nerve block and joint block procedures as a possible solution and continued to attempt pain patches and additional medication as well. She found that Gaylor's symptoms were consistent with the medical evidence and that he would be prevented from maintaining reliable attendance at work. She also noted Gaylor's need to change position more than once every two hours.
Also in 2003, Dr. Coscia referred Gaylor to Dr. Terry Trammel for a second opinion on his continued back pain after fusion surgery. Gaylor reported constant, easily aggravated pain, with his leg pain exceeding his back pain, and that no treatment had given him lasting relief. Dr. Trammel diagnosed post-fusion syndrome, but found that Gaylor's fusion was "solid" and opined that Gaylor was doing well and had reached maximum medical improvement. He determined that Gaylor was not markedly disabled from his pain. In his own records, however, Dr. Trammel noted that he told Gaylor that even a successful fusion operation had less than a 50% chance of alleviating his pain for a prolonged period.
Throughout, Gaylor was also referred to other physicians for additional pain management. He also briefly attended physical therapy that caused an increase in his pain, and he testified that he discontinued treatment because the pain became intolerable.
In summary, Drs. Coscia and Trammel opined that Gaylor could work with limitations, but Drs. Heathers and Green-Mack opined that he could not work. Dr. McLimore did not make a conclusion one way or the other but was concerned about Gaylor's possible addiction to pain medication and questioned the severity of Gaylor's pain.
Gaylor testified that he needs to lie down most of the day and that he experiences uncontrollable pain once or twice a week that leaves him in bed from several hours a day to days at a time. He has a constant pain level of about 7 to 8 on a scale of 10 and cares for his children only when they require little more than simple meal preparation and supervision. Otherwise, he occasionally does light housework, drives short distances, shops for groceries, and visits family or attends church services. He described his daily activities as almost entirely dominated by resting and watching television. He claimed he can sit in reasonable comfort for only ten to twenty minutes and can stand for only five to ten minutes and that his numerous surgeries have been unsuccessful at relieving his pain.
The ALJ determined that Gaylor was not disabled because of his back injury. Though Gaylor's injury was severe, the ALJ found that it did not meet the listings, which are benchmarks for severity that create a presumption of disability when the claimant's condition equals or exceeds them. After reviewing the medical evidence on both sides, he sided with the opinions of Drs. Coscia and Trammel that Gaylor was not in as much pain as he claimed to be and found that Gaylor was capable of light or sedentary work. He also found that Gaylor's testimony about his pain was not entirely credible. The ALJ noted that Gaylor walks over a mile per day, fixes meals, goes to restaurants and doctor's appointments, and is the "primary" caregiver for his children, whom he can lift despite his injuries. He also noted that Gaylor's fusion was successful, he could rise from a squat, and his gait was normal. But the ALJ's decision that Gaylor's back pain was not disabling gave little attention to Gaylor's claims of disability based on his depression and anxiety. So the Appeals Council remanded for additional consideration of these mental elements.
B. Gaylor's claim of disability based on his mental impairments
As early as 1999, Gaylor's doctors treating his back symptoms noted he suffered from depression. He was also treated by several specialists. In 2000, psychologist W. Shipley, a non-examining doctor, noted that Gaylor suffered from non-severe depression, dysthymia, and panic attacks, resulting in slight limitations of daily activities, social functioning, concentration, persistence, and pace.
Psychiatrist Carl Madsen treated Gaylor in 2000 and confirmed Gaylor's long history of depression accompanied by difficulty concentrating and unpredictable sadness. He diagnosed dysthymic disorder, panic attacks with agoraphobia, and major depression. He continued to identify concentration difficulties for Gaylor throughout 2000.
Psychiatrist David Kennedy began treating Gaylor and meeting with him on a regular basis in 2002. He also diagnosed anxiety, panic attacks, and depression, as well as difficulty concentrating and feelings of worthlessness. By 2004, Dr. Kennedy concluded that Gaylor was unable to sustain full-time employment because of his severe depression and anxiety disorder. He identified marked restrictions in daily living, maintaining social functioning, communicating, and concentrating, among other areas. These symptoms, he concluded, would interfere with Gaylor's ability to maintain reliable attendance at work.
In 2004, psychologist Henry Martin evaluated Gaylor via one round of testing on behalf of the Social Security Administration ("SSA"). He confirmed the diagnoses of anxiety, panic attacks, dysthymic disorder, isolation, and depression, and he noted that Gaylor suffered from a mixed personality disorder with feelings of inadequacy, severe psychosocial stressors, and failed back syndrome. He concluded, however, that Gaylor could work in an environment with limited social and public demands. Dr. Kennedy responded to this diagnosis in 2005 by letter, stating that Dr. Martin's conclusions significantly understated Gaylor's impairments in work, family relations, judgment, and mood.
Lastly, in 2005 psychologist Edward Czarnecki reviewed Dr. Martin's findings as well as Dr. Kennedy's letter and found that Dr. Martin had underestimated the severity of Gaylor's depression. He noted that Dr. Martin's conclusions were not consistent with his own observations, and that Dr. Martin's test results necessitated a diagnosis of "major" depressive disorder with a "very prominent" anxiety component and an "unusually intense" and "very serious" component of social withdrawal, inhibition, and isolation. He determined that "Dr. Martin's conclusion that the claimant would be able to work in a setting with limited social and public demands is not supported by the clinical findings or the diagnoses he himself indicates." Rather, he concluded that Gaylor's limited concentration and his inability to apply his cognitive skills to social situations outside the familiar home environment corroborated Dr. Kennedy's findings. He said Gaylor lacked the capacity to sustain even simple work activity and could not behave in a socially acceptable or consistent manner.
In his second decision, after the remand by the Appeals Council to consider Gaylor's mental impairments, the ALJ determined that Gaylor was not disabled on account of those impairments. Though the ALJ gave Gaylor "the benefit of the doubt" and found that his impairments of anxiety and depression were severe, the ALJ found that those impairments did not preclude employment. In reaching this conclusion, the ALJ relied on the opinion of Dr. Martin over Drs. Kennedy and Czarnecki, and he pointed to Gaylor's daily lifestyle: he often arrived at his doctors' appointments with a neat appearance and an intact memory, actively takes care of his two sons, drives, goes grocery shopping, goes to church, is interested in summer vacations with his family and upgrades for his computer, reads magazines, does housework, has friends, and worked from 1998 until 2001 even though he was already having panic attacks. He also found that Gaylor has a capacity to improve with treatment. Thus, the ALJ determined that the evaluation of Gaylor's treating psychiatrist, Dr. Kennedy, was inconsistent with the consultative examiner and Gaylor's daily functioning.
Finally, the ALJ reiterated his earlier conclusion from his first decision that Gaylor could "lift and carry 20 lbs. occasionally and 10 lbs. frequently; sit, stand and walk for six of eight, provided the work allows the individual to alternate into a sitting or standing position at his or her option for periods of one to two minutes per hour . . . and repetitive work with no more than superficial interaction with the general public, co-workers, or supervisors," among other limitations.
II. ANALYSIS
We will uphold an ALJ's decision that is supported by substantial evidence, that is, evidence that a reasonable person would accept as adequate to support the conclusion. See Schmidt v. Astrue, 496 F.3d 833, 841-42 (7th Cir. 2007); Murphy v. Astrue, 496 F.3d 630, 633 (7th Cir. 2007). Here, the ALJ identified the proper standard for disability and engaged in the correct five-part test for analyzing disability claims. See 42 U.S.C. § 423(d)(1)-(2); 20 C.F.R. § 404.1520(a)(4); Briscoe ex rel. Taylor v. Barnhart, 425 F.3d 345, 351-52 (7th Cir. 2005). But he failed in his responsibility to build a logical bridge between the evidence and his conclusions — he may not disregard evidence in support of disability without explaining why it is eclipsed by contrary evidence. Murphy, 496 F.3d at 634-35; see also Berger v. Astrue, 516 F.3d 539, 544 (7th Cir. 2008); Haynes v. Barnhart, 416 F.3d 621, 626 (7th Cir. 2005). He further made an unsupported determination that Gaylor was not credible, and he did not provide the logic or evidence supporting his finding of Gaylor's functional capacity. When an ALJ's decision is not supported by substantial evidence, a remand for further proceedings is the proper remedy (unless the record compels a decision for Gaylor, which it does not plainly do). Briscoe, 425 F.3d at 355.
A. The ALJ did not build a logical bridge to his conclusion that Gaylor's back injuries are not disabling
The substantial record in this case contains evidence that both supports and opposes a finding of disability. Essentially, Drs. Coscia and Trammel determined that Gaylor was able to work; Drs. Heathers and Green-Mack determined that he was not; and Dr. McLimore opined that Gaylor was not in as much pain as he claimed to be. As a result, if the ALJ had made a reasoned choice between the disparate medical findings, it would be beyond our review because of the deference afforded the ALJ's decisions. It is the ALJ's responsibility to weigh conflicting evidence and make a determination on disability. Dixon v. Massanari, 270 F.3d 1171, 1177 (7th Cir. 2001); Clifford v. Apfel, 227 F.3d 863, 870 (7th Cir. 2000). But here the ALJ did not break the tie fairly. Cf. Schmidt, 496 F.3d at 842-43 (claimant with back injury, leg pain, spinal fusion, and depression, who had reached maximum medical improvement regarding her back, was found ineligible by ALJ who properly discounted medical testimony in her favor based on the testimony's own inherent contradictions).
The ALJ's reasons for rejecting the medical opinions supporting disability are not persuasive. Pointing to the records of the doctors he ultimately credits, the ALJ emphasized the success of Gaylor's spinal fusion operation as evidence that he was not disabled. But he did not explain why the success of the operation meant that Gaylor's pain definitely would have ceased, especially when Dr. Trammel told Gaylor that, even if successful, the operation had less than a 50% chance of alleviating his pain for a prolonged period. The ALJ went on to discredit Drs. Green-Mack and Heathers because Dr. Green-Mack had not treated Gaylor before 2003 and Dr. Heathers had not treated him between 2000 and 2003, though he saw Gaylor before and after. The ALJ did not explain why the time frame of treatment justifies a lack of confidence in these doctors' findings. Their assessments are recent, and there is no suggestion that they needed additional time to render a reliable diagnosis. It is possible that Gaylor's condition worsened after some of the other doctors' examinations and is more accurately captured by these recent evaluations, a possibility not considered by the ALJ. See Clifford, 227 F.3d at 871.
The ALJ also addressed the possibility that Gaylor's pain might be less severe than he reported based on Drs. Coscia's and McLimore's assessments that he was in danger of dependency on medication and was deconditioned to pain. But Drs. Heathers and Green-Mack found that Gaylor's described pain was consistent with the medical evidence, and Gaylor's own testimony went uncontradicted that even with medication he is forced to lie down much or all of the day and can only stand for up to ten minutes and sit for up to twenty minutes.
The ALJ also stressed that Gaylor claimed to walk over a mile throughout each day. But walking was part of his prescribed therapy and was also necessary for daily life; the ALJ could not hold it against Gaylor for attempting to follow doctors' instructions to walk, especially when walking, according to Gaylor, still involved pain. See Gentle v. Barnhart, 430 F.3d 865, 867 (7th Cir. 2006); Carmdine v. Barnhart, 360 F.3d 751, 755 (7th Cir. 2004). The ALJ emphasized that Gaylor could squat and walk on heels and toes but never explained why these abilities make him fit for work. Gaylor's doctors examined his capacity in this regard, and half of those that gave an opinion on the matter found that these abilities did not support employability. The ALJ also noted that Gaylor's condition began in 1998 but he worked until 2001 and reasoned that since Gaylor's pain supposedly remained unchanged, his ability to work then implies an ability to work now. But Gaylor claimed to have quit work when the pain became so bad that he could not continue, and that pain was unabated by later medical procedures. Gaylor's testimony that the surgeries, epidurals, and other procedures did not reduce his increased pain contradicts the ALJ's assumption that he now experiences the same level of pain that he did between 1998 and 2001.
The ALJ also looked to Gaylor's daily activities as evidence of his ability to work. But it is a deficient analysis to assume that a claimant's ability to care for personal needs and the needs of his or her children is synonymous with an ability to be gainfully employed. See Gentle, 430 F.3d at 867 (remanding case in part because ALJ improperly equated work in the labor market with household work, including child-care, performed by claimant who suffered from spinal disk disease and was in pain and required aid from others); Carmdine, 360 F.3d at 755-56 (ALJ improperly presumed that claimant suffering back problems could work just because she could occasionally drive, shop, and do housework); Clifford, 227 F.3d at 872. Compare Schmidt, 496 F.3d at 843-44 (ALJ properly found claimant not credible because her daily activities were substantial, such as attending college classes, and her disability was not corroborated by medical evidence and was inconsistent with her refusal of pain management). The ALJ failed to say why Gaylor's relatively meager activities disputed the evidence in favor of disability. See Clifford, 227 F.3d at 870. Moreover, the ALJ did not support his characterization of Gaylor as the "primary" caregiver for his children. The hearing testimony shows that Gaylor had a minimal, supervisory role when he was home with his children and that he required help from family to care for them. Rather than being able to lift his children, as noted by the ALJ, Gaylor's youngest child stayed in daycare until he could walk because Gaylor could not lift the child.
The ALJ had a responsibility to confront the evidence in Gaylor's favor and explain why it is not persuasive. Indoranto v. Barnhart, 374 F.3d 470, 474 (7th Cir. 2004) (claimant with similar back difficulties explained a similar need to lie down and take hot baths that was not confronted by the ALJ, warranting reversal). He did not adequately do so.
B. The ALJ did not build a logical bridge to his conclusion that Gaylor's mental impairments are not disabling
The ALJ committed an even more severe omission in his conclusion that Gaylor is not disabled by his mental impairments. Rather than failing to explain his choice between competing medical conclusions as above, the ALJ discredited the manifest weight of the evidence with only minimal explanation. Dr. Kennedy was Gaylor's treating psychiatrist for several years, and he determined that Gaylor could not work. As a treating physician, his opinion is normally granted deference because of his close knowledge of Gaylor's condition, though this deference can dissipate depending upon the circumstances of the case. See Schmidt, 496 F.3d at 842; Hofslien v. Barnhart, 439 F.3d 375, 376-77 (7th Cir. 2006). He was contradicted by Dr. Martin, who ran a battery of tests for the SSA. But to the degree that there was a tie between the two opinions, it was broken by Dr. Czarnecki, the consulting psychologist who agreed with Dr. Kennedy and found the conclusions of Dr. Martin to be insupportable.
The weight of the evidence favored a finding of disability, and the ALJ was again required to confront that evidence and explain why it was unpersuasive. The ALJ relied on Dr. Martin's findings primarily because he judged that Gaylor's daily activities were not consistent with the other doctors' assessments. But Dr. Kennedy evaluated Gaylor's lifestyle in his almost monthly appointments, and Dr. Martin met with Gaylor only once. As addressed above, the elements of daily life and necessities do not automatically equate with an ability to work. It's not that daily activities are never relevant (if a claimant with a supposed bad back was hanging drywall as a side business, or if a supposed depressed and socially adverse client was the head of her child's PTA and taught music lessons, the contradictions might be obvious); rather the ALJ simply must show his work. He must show why the activities contradict the claim. Here, the ALJ did not explain why activities such as visiting family, shopping, and going to church contradict the conclusion of mental impairment by doctors examining the same lifestyle.
C. The ALJ's failure to support his conclusions on the medical evidence undermines his credibility finding
The ALJ's failure to adequately confront and distinguish the evidence in favor of disability from the evidence opposing that finding also infected his determination of Gaylor's credibility. Though we review credibility findings deferentially, we have more freedom to overturn a credibility finding based on objective and reviewable factors, rather than subjective ones. Clifford, 227 F.3d at 872. The ALJ was required to explain adequately the adverse credibility finding, see Ribaudo v. Barnhart, 458 F.3d 580, 584-85 (7th Cir. 2006), and articulate why he did not believe Gaylor's account of his pain and its severity, Briscoe, 425 F.3d at 354; Golembiewski v. Barnhart, 322 F.3d 912, 915-16 (7th Cir. 2003).
Here, the ALJ based his determination of Gaylor's credibility on the contradictions between Gaylor's testimony and the medical evidence opposing a finding of disability. But without the foundational determination discussed above, that the opinions of the doctors opposing disability are in fact more reliable than the opinions of the doctors supporting disability, this comparison cannot stand. It is only after the ALJ gives proper treatment to the diverging opinions of the doctors that those opinions can rightly be compared to Gaylor's testimony. The ALJ's credibility determination is "intertwined with the same gaps in the record and reasoning" that call into question the ALJ's treatment of Drs. Heathers and Green-Mack and thus must be reexamined in light of a properly reasoned treatment of the medical evidence. See Murphy, 496 F.3d at 635.
Furthermore, the ALJ's willingness to accept that Gaylor exaggerated his pain or was addicted to his medication relied on only half the evidence and did not distinguish the medical evidence corroborating his pain. This is an especially important inquiry in the assessment of credibility, given Gaylor's testimony that his cooperation with attempts to wean him off the medications left him with debilitating pain — essentially that he had little option but to continue the medication. This lack of options was echoed by the opinion of Dr. Whitfield. Once Gaylor produced evidence of his underlying impairment, his testimony of the resultant pain could not be undermined just because its cause is not medically certain. See Carradine, 360 F.3d at 753-54. Gaylor had undisputable back problems, attested to by many surgeries and medical records. Just because his spinal fusion was a successful surgery, the ALJ could not automatically conclude that his testimony about his lasting pain was false. The procedure may have been executed perfectly and yet failed to relieve the pain. It is unlikely Gaylor would have endured so many surgeries and treatments for pain that did not exist, just to increase the credibility of his social security claim. See Vertigan v. Halter, 260 F.3d 1044, 1050 (9th Cir. 2001); Cox v. Apfel, 160 F.3d 1203, 1207 (8th Cir. 1998). Moreover, the ALJ again did not articulate why Gaylor's testimony was contradicted by his daily activities. Because the record contained support for his claims, the ALJ had to show why his activities hinder his claims when they might have been completed out of necessity and in pain. See Zurawski v. Halter, 245 F.3d 881, 887-88 (7th Cir. 2001); Clifford, 227 F.3d at 871. The ALJ based his credibility finding on errors in reasoning, rather than on Gaylor's demeanor, which we have more power to review and which in this case warrants a remand. See Carradine, 360 F.3d at 753-54.
D. The ALJ did not explain his functional limitations determination and did not consider Gaylor's impairments in the aggregate
The ALJ also erred in his assessment of Gaylor's residual functional capacity. He hypothetically asked the vocational expert if a person who could stand and walk for six hours out of an eight-hour day, with a one to two minute break to change positions every two hours, was an employable individual. But he never explained why this description applies to Gaylor. The evidence at the hearings, with corroborating medical opinions, was that Gaylor had to lie down for hours a day and sometimes days at a time and that he could stand for up to ten minutes and sit for up to twenty minutes. This evidence seems plainly contrary to the ALJ's hypothetical question of the vocational expert, which included much longer periods of standing and walking and was the basis of the functional limitations the ALJ ultimately imposed. He did not build a logical bridge detailing how he reached this conclusion. When an ALJ fails to explain the basis in the record for his proposed functional limitations and asks hypothetical questions of vocational experts that are flawed and do not fully capture the claimant's limitations, then those errors undermine the entire analysis of what work a claimant can perform. Young v. Barnhart, 362 F.3d 995, 1002-05 (7th Cir. 2004). Such an omission, in itself, is sufficient to warrant reversal. See Briscoe, 425 F.3d at 352.
Lastly, the ALJ was required to consider Gaylor's back problems and mental impairments together and evaluate their aggregate effect on his ability to work. See Golembiewski, 322 F.3d at 918; Clifford, 227 F.3d at 873. His opinion lacks an evaluation of whether Gaylor's back pain is exacerbated by his depression, or the reverse, and whether that might create a more severe limitation on Gaylor than the impairments considered independently.
III. CONCLUSION
Accordingly, we REVERSE and REMAND to the ALJ for further proceedings consistent with this order.