I. Background

Moss filed her application for benefits in October 2003. She had been working as a gas-station attendant for six years before she fractured and dislocated her right ankle and suffered a severe laceration on her right knee in a car accident. At the time of the accident, she was 43 years old, had a high-school education, and had previously worked as a housekeeper in hotels and nursing homes. After the accident, however, Moss was unable to return to her job at the gas station.

Moss underwent immediate surgery to repair her ankle and was released from the hospital three days after the accident. Two months later, the pins that had been used to repair the fracture were removed. Moss’s surgeon, Dr. Asamonja Roy, an orthopedic specialist, thought that Moss was doing well and encouraged her to begin walking and putting weight on her right foot. Moss started physical therapy and by mid-November was walking and able to bear 60 to 70% of her weight on her right foot. By December 2003 she was able to walk at least 600 feet with the use of a cane.

In the months following her accident, Moss also was treated by her family physician, Dr. Steven Norris. A month after the injury, Dr. Norris detected damage to the peripheral nervous system of her ankle as a result of the fracture. In January 2004 he reviewed x-rays of her ankle and observed a bone formation that he suspected was limiting her range of motion, which, he reported, was generally “pretty good.” Dr. Norris also noted that Moss was suffering from “[r]ight lower extremity neuropathy,” for which he prescribed Neurontin and advised her to continue with physical therapy.

Five months after the accident, in March 2004, Moss was still experiencing difficulties, so she scheduled appointments with a neurologist as well as a second orthopedic specialist. Moss met first with the neurologist, Dr. Rakesh Garg, who was referred to her by Dr. Roy. Following his examination, Dr. Garg concluded that Moss had a “slight limitation” of movement in her right ankle, but he did not detect any nerve damage.

Five days later Moss met with Dr. Steven Kodros, the orthopedic specialist, who, unlike Dr. Garg, described her range of motion as “quite limited.” Dr. Kodros noted that Moss was experiencing pain when she flexed her ankle, and that she had stiffness, tenderness, and diminished sensation in her right foot. He further observed that Moss was walking with a steppage gait pattern and using a cane. Dr. Kodros reviewed an outside CT scan and x-rays of Moss’s ankle and found “some mild residual bony deformity and some early degenerative changes.” He noted that Moss had residual symptoms “consistent with the natural history of her injury itself and more specifically related to residual posttraumatic arthrofibrosis, periarticular soft tissue adhesions, and likely the early development of some early posttraumatic arthritis.” Dr. Kodros ordered a second CT scan and several weeks later followed up with Moss about his findings. The new CT scan confirmed that the fracture itself had healed but not without leaving small bony fragments and debris in the ankle. He noted some early degenerative changes and features that suggested avascular necrosis—death of the bone tissue. Based on these findings, Dr. Kodros recommended “conservative management” that would include use of a customized “Arizona ankle-brace” and corticosteroid injections. Should these measures fail, he indicated during this March 2004 consultation, he would recommend surgery.

Moss next saw Dr. Kodros in February 2005. The exam revealed that her condition had not improved. Moss had not obtained the Arizona brace because her insurance would not cover it, so Dr. Kodros gave her a temporary brace and repeated his recommendation that she attempt conservative management before resorting to more surgery. Several weeks later, Moss also visited Dr. Norris for a routine check-up. He noted that she was wearing an ankle brace and had a limited range of motion in her right ankle and an altered gait due to pain.

Before her car accident, Moss had received medical treatment for other unrelated ailments. Medical records document treatment in early 2003 for diverticulosis, chronic constipation, hyperthyroidism, sinusitis, and fatigue. And before that, in 2000, Moss was diagnosed as suffering from migraine headaches.

The Social Security Administration denied Moss’s application a month after it was filed. A month after that, the agency also denied reconsideration. Moss requested further review, and in January 2006 an ALJ heard testimony from Moss and a vocational expert.

Moss testified that she experiences “chronic pain 24/7,” as well as stiffness, tightness, and swelling in her right ankle. She stated that she is unable to sit or stand for extended periods of time and cannot walk with full weight on her right foot, resulting in back and hip pain. The pain, she said, interferes with her sleep, requiring her to take sleeping pills. She continues to take Neurontin to alleviate the burning feeling and “pins and needles” sensation in her ankle. She also takes Tylenol Arthritis and Motrin, which provide only minimal relief but do not cause the drowsiness and constipation she experienced with the prescription pain medications she had been taking.

In describing her daily activities, Moss stated that she is able to dress herself and bathe using a stool. She tries to do housework but can no longer squat or bend; she is able to make light meals and wash dishes, though standing is painful. She no longer drives or does any yard work, laundry, or grocery shopping. Moss testified that during her husband’s frequent business travels, her adult daughter, who lives with them, helps with the laundry and shopping. For exercise, Moss said, she uses her cane to take occasional walks around the block and up and down the driveway.

The ALJ asked a vocational expert (“VE”) whether there are entry-level jobs available to a 46-year-old high-school graduate who is limited to unskilled, sedentary work that requires no climbing and only minimal standing and walking. The VE responded that although Moss cannot perform her past relevant work and has no transferrable skills, he found three positions that exist in significant numbers in the national economy—cafeteria cashier, ampule sealer, and surveillance system monitor—that Moss could perform given the limitations described.

The ALJ concluded that Moss is not disabled because she can perform some sedentary jobs that exist in significant numbers in the national economy. The ALJ applied the sequential five-step analysis, see 20 C.F.R. § 404.1520, and at step one found that Moss had not been engaged in substantial gainful activity and at step two that her ankle injury constitutes a severe impairment. The ALJ acknowledged that Moss also suffers from back and abdominal pain, sinusitis, fatigue, and headaches. At step three, however, the ALJ found that Moss’s ankle injury does not meet or equal a listed impairment. Moving to step four, the ALJ found that Moss cannot perform her past relevant work as a cashier or housekeeper but even with all of her ailments still retains the residual functional capacity to perform some sedentary jobs.

In making these determinations, the ALJ declined to fully credit Moss’s complaints of pain because, the ALJ said, there had been no medical finding that she needs a cane or that she is unable to effectively ambulate. The ALJ further stated that the medical evidence and Moss’s own account of her daily activities do not corroborate her testimony concerning the intensity, duration, and limiting effects of her symptoms. The ALJ additionally discounted the medical opinions of Dr. Kodros, the orthopedic specialist. The ALJ characterized Dr. Kodros’s opinions as inconclusive and inconsistent with Dr. Garg’s findings, and surmised that some of Dr. Kodros’s medical opinions “may have been made to help the claimant in a pending legal matter since the doctor was reporting directly to her attorney.” Therefore, at step five the ALJ concluded that a significant number of sedentary jobs exist in the national economy that Moss can perform despite her limitations.

II. Analysis

Because the Appeals Council declined to review the ALJ’s decision, the ALJ’s ruling is the final decision of the Commissioner of Social Security. Getch v. Astrue, 539 F.3d 473, 480 (7th Cir. 2008). And while Moss raises several arguments about the district court’s order upholding the denial of benefits, we review the district court’s decision de novo, meaning we review the ALJ’s ruling directly. See Elder v. Astrue, 529 F.3d 408, 413 (7th Cir. 2008). Therefore, only Moss’s challenges to the ALJ’s decision are at issue here. See Skinner v. Astrue, 478 F.3d 836, 841 (7th Cir. 2007). We will uphold that decision if it is supported by substantial evidence. See 42 U.S.C. § 405(g); Skinner, 478 F.3d at 841.

Moss raises a number of arguments, primarily that the ALJ failed to give appropriate weight to the medical opinions of Dr. Kodros which in turn contributed to a flawed assessment of her credibility. She also argues that at step three the ALJ failed to conduct a legally sufficient analysis of the listings of impairments. We agree with Moss about Dr. Kodros and about the ALJ’s adverse credibility determination, and further conclude that the ALJ’s determination that Moss’s impairment does not meet or equal a listed impairment is not supported by substantial evidence.

Moss is correct that the ALJ failed to appropriately consider the medical opinions of Dr. Kodros, one of her treating physicians. A treating physician’s opinion about the nature and severity of the claimant’s impairment is normally given controlling weight so long as it is “well-supported by medically acceptable clinical and laboratory diagnostic techniques” and is consistent with substantial evidence in the record. 20 C.F.R. § 404.1527(d)(2); Bauer v. Astrue, 532 F.3d 606, 608 (7th Cir. 2008). Dr. Kodros’s opinions support Moss’s complaints of pain and physical limitation given his findings that her ankle exhibited bony defects, bony fragments and debris, early degenerative changes, and residual symptoms from her ankle fracture “specifically related to residual posttraumatic arthrofibrosis, periarticular soft tissue adhesions, and likely the early development of some early posttraumatic arthritis.”

The ALJ, however, discounted the opinions of Dr. Kodros based on speculation that Moss was referred to him by her attorney and that his findings “may have been made to help the claimant in a pending legal matter since the doctor was reporting directly to her attorney.” And the ALJ altogether failed to address whether Dr. Kodros’s medical opinions are supported by medically acceptable clinical and laboratory diagnostic techniques. See Bauer, 532 F.3d at 608. An ALJ’s conjecture is never a permitted basis for ignoring a treating physician’s views, see Gudgel v. Barnhart, 345 F.3d 467, 470 (7th Cir. 2003); Rohan v. Chater, 98 F.3d 966, 971 (7th Cir. 1999), and it is further questionable whether the ALJ’s basis for discounting Dr. Kodros’s medical opinion would be legitimate, even if not speculative, see Reddick v. Chater, 157 F.3d 715, 726 (9th Cir. 1998) (“[T]he mere fact that a medical report is provided at the request of counsel or, more broadly, the purpose for which an opinion is provided, is not a legitimate basis for evaluating the reliability of the report.”).

Additionally, the ALJ failed to determine the weight to be accorded Dr. Kodros’s opinion in accordance with Social Security Administration regulations. See 20 C.F.R. § 404.1527(d)(2). If an ALJ does not give a treating physician’s opinion controlling weight, the regulations require the ALJ to consider the length, nature, and extent of the treatment relationship, frequency of examination, the physician’s specialty, the types of tests performed, and the consistency and supportability of the physician’s opinion. Id.; Bauer, 532 F.3d at 608; Books v. Chater, 91 F.3d 972, 979 (7th Cir. 1996). The ALJ apparently thought that Dr. Kodros’s assessment of Moss’s medical condition conflicted with Dr. Garg’s opinion and that Dr. Garg’s view should carry the day. But the choice to accept one physician’s opinions but not the other’s was made by the ALJ without any consideration of the factors outlined in the regulations, such as the differing specialties of the two doctors, the additional diagnostic testing conducted by Dr. Kodros, or the consistency of Dr. Kodros’s findings over the course of a year. Moreover, the perceived conflict between the two medical opinions is illusory. Dr. Garg, a neurologist, was looking only for nerve damage, while Dr. Kodros, an orthopedic specialist, evaluated entirely different aspects of Moss’s injury after reviewing x-rays and two CT scans, and after examining Moss twice over the course of a year. The fact that Dr. Garg did not find a neurological cause for Moss’s condition does not undercut the opinions of Dr. Kodros, whose specialty is the function of the musculoskeletal system, extremities, spine, and associated structures. See STEDMAN’S MEDICAL DICTIONARY, 1277 (27th ed. 2000). Therefore, in discounting the medical opinions of Dr. Kodros, the ALJ failed to apply the correct legal standard and further failed to support that decision with substantial evidence.

Moss is also correct that the ALJ’s failure to give appropriate weight to Dr. Kodros’s testimony resulted in a flawed determination that her complaints of pain are not credible. We will uphold an ALJ’s credibility determination if the ALJ gave specific reasons for the finding that are supported by substantial evidence. Arnold v. Barnhart, 473 F.3d 816, 823 (7th Cir. 2007). But as the regulations state, an ALJ cannot disregard subjective complaints of disabling pain just because a determinable basis for pain of that intensity does not stand out in the medical record. See S.S.R. 96-7p; Johnson v. Barnhart, 449 F.3d 804, 806 (7th Cir. 2006). If the medical record does not corroborate the level of pain reported by the claimant, the ALJ must develop the record and seek information about the severity of the pain and its effects on the applicant. Clifford

v. Apfel, 227 F.3d 863, 871-72 (7th Cir. 2000). In this case, however, the ALJ simply marginalized Dr. Kodros’s opinions without a sound explanation and then went on to conclude that neither Moss’s own testimony nor the remaining medical evidence supported her subjective complaints of pain. The ALJ specifically noted the lack of medical findings addressing Moss’s ability to ambulate or her need for a cane. Not only did the ALJ fail to seek an explanation for this lack of medical evidence, but, more importantly, the perceived gaps in the medical evidence vanish when Dr. Kodros’s assessment of Moss’s condition is properly considered.

There are two other troubling features about the ALJ’s assessment of Moss’s credibility. First, the ALJ’s recitation of the administrative record is misleading or inaccurate on several significant points. With respect to Moss’s treatment history, the ALJ represents that her doctors had not recommended further surgery and that she had gone a relatively long period of time without seeing her family doctor. In fact, however, Dr. Kodros said that another surgery was a real possibility if Moss did not respond to his recommended course of conservative treatment. And while infrequent treatment or failure to follow a treatment plan can support an adverse credibility finding, we have emphasized that “the ALJ ‘must not draw any inferences’ about a claimant’s condition from this failure unless the ALJ has explored the claimant’s explanations as to the lack of medical care.” Craft v. Astrue, 539 F.3d 668, 679 (7th Cir. 2008) (quoting

S.S.R. 96-7p). The ALJ failed to question Moss about this gap in treatment or the evidence showing that she was unable to obtain the customized ankle brace because of problems with her insurance coverage. Moreover, the ALJ states that Moss is able to live independently while her husband is out of town, but Moss testified that her adult daughter, who lives with Moss, helps her while her husband is away. Similarly, the ALJ incorrectly states that Moss refuses to take prescription pain medications because of information she read in a book, but in fact Moss testified that she has been avoiding prescription pain medications because they cause side effects, including drowsiness and constipation.

The second aspect that gives us pause about the ALJ’s credibility assessment is the undue weight placed on Moss’s household activities in assessing her ability to work outside the home. An ALJ cannot disregard a claimant’s limitations in performing household activities. See Craft, 539 F.3d at 680; Mendez v. Barnhart, 439 F.3d 360, 362 (7th Cir. 2006). The ALJ here ignored Moss’s numerous qualifications regarding her daily activities: while washing dishes she shifts her weight to the left but still experiences pain; when she last went to the store she had to use the cart for support and was unable to stay long; and when she last tried to drive the family vehicle more than a year before the evidentiary hearing, she had difficulty pushing the pedals because of a lack of control or feeling in her foot.

Finally, the ALJ’s determination that Moss’s impairment does not meet a listed impairment is not supported by substantial evidence. In evaluating Listing 1.03, the ALJ found that Moss had failed to establish an inability to ambulate effectively, one of the necessary elements of that listing. The regulations state that “ineffective ambulation” is “defined generally” as requiring the use of a hand-held assistive device that limits the functioning of both upper extremities. See 20 C.F.R. pt. 404P, app. 1, § 1.00(B)(2)(a). But the regulations further provide a nonexhaustive list of examples of ineffective ambulation, such as the inability to walk without the use of a walker or two crutches or two canes; the inability to walk a block at a reasonable pace on rough or uneven surfaces; the inability to carry out routine ambulatory activities, like shopping and banking; and the inability to climb a few steps at a reasonable pace with the use of a single handrail. Id.

Here, the ALJ concluded that Moss had failed to establish her inability to effectively ambulate because Moss uses just one cane and because according to the ALJ, the medical evidence does not point to ineffective ambulation and Moss herself testified that she is “able to live independently” and occasionally walks around the block. As previously noted, however, the ALJ’s determinations regarding the medical evidence and Moss’s credibility are not supported by substantial evidence.

Consequently, the ALJ failed to adequately consider whether Moss in fact meets the listing based on the provided examples such as an inability to walk a block at a reasonable pace on rough or uneven surfaces, or the inability to carry out routine activities, like shopping and banking. See 20 C.F.R. pt. 404P, app. 1, § 1.00(B)(2)(a).

III. Conclusion

Accordingly, we VACATE the judgment of the district court and REMAND with instructions to remand the case to the agency for further proceedings.

2-5-09