Williams, Circuit Judge.

Dana Wayne Duncan, Attorney, DUNCAN DISABILITY LAW, SC, Nekoosa, WI, for Plaintiff-Appellant.

Lu Han, Attorney, SOCIAL SECURITY ADMINISTRATION, Office of the General Counsel, Region V, for Defendant-Appellee.

Before Flaum, Manion, and Williams, Circuit Judges.

Williams, Circuit Judge.

Mark Schloesser, who suffers from a combination of physical impairments, applied for disability insurance benefits in November 2012. The Social Security Administration ("SSA") initially denied his application. After reconsideration and a hearing, an Administrative Law Judge ("ALJ") found him disabled and granted benefits in August 2014. One month later, sua sponte , the SSA Appeals Council commenced review of the ALJ's favorable decision. The Appeals Council reversed the ALJ's favorable decision, concluding that the ALJ's findings were not supported by substantial evidence and that Schloesser was not disabled as of September 30, 2011, his last insured date.

Schloesser sought review of the Appeals Council's decision in the district court. It affirmed the Appeals Council's decision, finding that it was supported by substantial evidence. Schloesser now appeals, arguing that the Appeals Council erred because: (1) it failed to apply SSR 83-20 in its determination of his onset date; (2) its findings that he did not suffer from severe impairments of cervical radiculopathy, major joint dysfunction, and history of left shoulder surgery were not supported by substantial evidence; and (3) its finding that his residual functional capacity ("RFC") did not include being off-task up to 10% of the workday or needing unscheduled breaks was not supported by substantial evidence. We affirm the denial of benefits because we find that SSR 83-20 was irrelevant to the Appeals Council's determination and that its findings regarding his impairments and RFC are supported by substantial evidence. I. BACKGROUND

Mark Schloesser worked for 23 years as a dry curer in a meat-processing factory, a position that required him to regularly lift more than 70 pounds. But, after undergoing rotator cuff surgery on his left shoulder in 2001 and then a lactimectomy (disc removal in his lower back) in 2002, Schloesser left the meat-processing factory in 2003. For approximately six years, he was self-employed in construction, until his persistent shoulder and lower back problems prevented him from being able to regularly lift more than 50 pounds as required by his construction work. In November 2012, Schloesser applied for disability insurance benefits under 42 U.S.C. § 416(i). Schloesser, who had not worked since October 2009, complained that he suffered disabling pain in his back, neck, and shoulder, which, combined with his obesity, made it impossible for him to work. He alleged a disability onset date of October 1, 2009 (later amended to January 1, 2011). The SSA initially denied his application, and so he requested a hearing before an ALJ.

A. ALJ's Favorable Decision

On August 13, 2014, based on the medical evidence submitted and Schloesser's testimony at the hearing, an ALJ found Schloesser to be disabled. In making this finding, the ALJ applied the five-step sequential evaluation process outlined in 20 C.F.R. § 404.1520(a)(4). The ALJ first determined that Schloesser had not engaged in substantial gainful activity since his alleged onset date of January 1, 2011, and, second, that Schloesser suffered from severe impairments of a history of cervical radiculopathy, degenerative disc disease, major joint dysfunction, history of left shoulder surgery, and obesity. Third, the ALJ found that Schloesser did not have an impairment that met or medically equaled any impairment in Appendix 1 to 20 C.F.R. Part 404, Subpart P, and that Schloesser could perform only light work, stand and walk for four hours, sit for six hours, occasionally climb, stoop, kneel, crouch and crawl, frequently balance, not reach overhead with either extremity, no more than frequently reach in all other directions with his non-dominant left upper extremity, occasionally flex, extend and rotate his neck, he may need unscheduled breaks in addition to regular breaks, and he may be off-task more than 10% of the workday. Fourth, the ALJ found that Schloesser was not capable of performing his past relevant work as a dry curer or construction worker based on this RFC. Finally, relying on testimony from a vocational expert, the ALJ determined that given Schloesser's age, education, work experience, and RFC, there were no jobs that existed in significant numbers in the national economy that he could perform. Based on these findings, the ALJ found Schloesser disabled since January 1, 2011, which was prior to his September 30, 2011 date last insured, and so granted his application for disability benefits.

B. Appeals Council Reversed ALJ

On September 9, 2014, the SSA's Appeals Council sent notice to Schloesser that it intended to set aside the ALJ's favorable decision and issue a finding that Schloesser was not entitled to disability insurance benefits because he was not disabled before his date last insured. Schloesser filed a response on November 12, 2014. The Appeals Council determined that the new evidence he submitted was mostly dated after his insured status had expired and the evidence that did pertain to the period prior to his date last insured was duplicative or cumulative of the evidence already considered. Following the same five-step process as the ALJ, the Appeals Council: (1) agreed with the ALJ's findings at step one; (2) partially agreed with its findings at step two, but disagreed that Schloesser had severe impairments of cervical radiculopathy, major joint dysfunction, and left shoulder surgery on or before his date last insured; (3) agreed with the ALJ's finding at step three; (4) agreed with the ALJ's finding that he could not work his prior job and partially agreed with the ALJ's determination of Schloesser's RFC, but disagreed that he could only occasionally flex, extend, and rotate his neck, may need to be off task for more than 10% and may need unscheduled breaks; and, importantly, (5) disagreed that there were no jobs that existed in significant numbers in the national economy that he could perform, based on its determination of his RFC. Therefore, it found that Schloesser was not disabled as of his date last insured.

In finding that Schloesser did not have the severe impairment of cervical radiculopathy at step two, the Appeals Council explained that Schloesser was diagnosed with only mild, not severe, cervical degeneration with radicular symptomology in December 2011, which was over two months after his date last insured. It also noted that his symptoms improved with physical therapy, and his radiculopathy resolved in February 2012. He did not report neck pain again until February 2014. Based on this evidence, the Appeals Council did not agree with the ALJ's determination that Schloesser's RFC included only occasional flexing, extending, and rotating of his neck at step four.

The Appeals Council also found that Schloesser did not suffer from the severe impairments of major joint dysfunction and left shoulder surgery on or before his date last insured. It explained that, except for records of rotator cuff surgery in 2001, there was no evidence regarding significant shoulder pain until well after his date last insured. In fact, after he was diagnosed with left shoulder subacromial impingement syndrome in February 2012 and referred to physical therapy, he did not seek subsequent treatment for significant shoulder symptoms. And even though it found that Schloesser did not suffer from any severe impairment due to his shoulder, the Appeals Council still credited and accounted for Schloesser's non-severe shoulder pain in its determination of his RFC. But, at step five, it still found that "even when considering limitations in [Schloesser's] ability to reach with the left upper extremity, the vocational expert testified that there would still be a number of jobs [he] could perform."

Next, at step four, the Appeals Council found that Schloesser's RFC did not include being off-task more than 10% of the workday or needing unscheduled breaks. The Appeals Council determined that "the limited evidence available before the date last insured does not substantially support [Schloesser's] allegations that he experienced disabling symptoms before the expiration of his insured status." It explained that Schloesser's testimony in August 2013 to the ALJ that, starting in 2010, he could not "sit much more than a half an hour before the pain [was] to a point where [he] need[ed] to lay back down" and that he spent more than 50% of his day laying down was not supported by the medical evidence from 2011. It noted that Schloesser was diagnosed with severe degenerative disc disease in February 2011 and experienced a limited range of motion in his lumbar spine, a slightly flexed posture, slightly reduced manual muscle strength of 4+/5, and a mildly antalgic gait in March 2011, but he only rated his pain as a 2 on a 10-point scale and indicated that he stopped taking his prescribed painkiller Tramadol, in favor of Tylenol and Ibuprofen to manage his lower back pain. And even though Schloesser experienced some weakness and reduced range of motion in his lower left extremity, all other muscles tested were at 5/5 and his doctors observed no tenderness in his lower back. Then, as the Appeals Council observed, in August 2011, one month before his date last insured, Schloesser reported discontinuing ibuprofen use. Finally, the Appeals Council stated that Schloesser did not report increased back pain again until May 2013, indicating that the pain had worsened "over the last year," meaning it had only increased since 2012, well after his date last insured.

Based on these findings, the Appeals Council found Schloesser's RFC to be: ability to perform a range of light work, but only occasionally climb, stoop, kneel, crouch, and crawl, frequently balance, frequent but not constant reaching in all other directions with the nondominant left upper extremity, and no overhead reaching bilaterally. It then considered the vocational expert's testimony to the ALJ that someone of Schloesser's age, education, and work experience with this RFC, with the added limitations of: only light lifting and carrying, standing and walking only four hours of the work day and sitting up to six, and being off-task up to 10% of the workday while at the workstation in addition to normal breaks, was able to work as a packer, receptionist, and office clerk. The Appeals Council therefore found that Schloesser had not been disabled as of his date last insured.

II. ANALYSIS

Schloesser sought judicial review of the final decision by the Appeals Council by filing suit in the district court, pursuant to 42 U.S.C. § 405(g). The district court affirmed the Appeals Council decision, prompting this appeal. We review the district court's decision de novo . Where the Appeals Council reverses an ALJ's grant of disability benefits, we limit our review to determining whether the "Appeals Council's decision is supported by substantial evidence on the record as a whole." Bauzo v. Bowen , 803 F.2d 917, 919 (7th Cir. 1986) (noting that an ALJ's conflicting findings are part of record as a whole and are considered in determining whether Appeals Council's decision is supported by substantial evidence); Elder v. Astrue , 529 F.3d 408, 413 (7th Cir. 2008). Substantial evidence is "such relevant evidence as a reasonable mind might accept as adequate to support a conclusion." Kepple v. Massanari , 268 F.3d 513, 516 (7th Cir. 2001) (internal citation and quotation marks omitted). Unless patently wrong, we defer to the credibility findings made by the Appeals Council. Engstrand v. Colvin , 788 F.3d 655, 660 (7th Cir. 2015) (citing Curvin v. Colvin , 778 F.3d 645, 651 (7th Cir. 2015) ); see Haynes v. Barnhart , 416 F.3d 621, 626 (7th Cir. 2005) (we may not decide facts anew or make independent credibility determinations). As long as the Appeals Council identified supporting evidence in the record and built a "logical bridge" from that evidence to its conclusion, we must affirm. Id . This is true even if reasonable minds could differ about the ultimate disability finding. Id .

Though Schloesser devotes a considerable amount of his appellate argument to showing he is currently disabled, the Appeals Council rejected Schloesser's claim because it determined he was not disabled as of his date last insured. Therefore, in our review of the Appeals Council decision, we consider only its finding that he was not disabled as of his date last insured, not Schloesser's argument on appeal that he is currently disabled. This is because he was no longer eligible for disability benefits after his date last insured, as he had exhausted his earned "quarters of coverage" since he had not worked for several years. Parker v. Astrue , 597 F.3d 920, 924 (7th Cir. 2010, as amended on reh'g in part May 12, 2010); see 42 U.S.C. § 423(c) ; 20 C.F.R. § 404.140.

A. SSR 83-20 Irrelevant to Determination of Onset Date

Schloesser argues that the Appeals Council violated SSR 83–20 by not consulting a medical expert in order to determine whether the onset date of his disability occurred before his date last insured. However, SSR 83–20 only addresses the situation in which a finding is made "that an individual is disabled as of an application date and the question arises as to whether the disability arose at an earlier time." Scheck v. Barnhart , 357 F.3d 697, 701 (7th Cir. 2004). As in Scheck , the Appeals Council did not find Schloesser disabled and so was not required to find an onset date. Instead, the Appeals Council determined directly whether Schloesser was totally disabled by September 30, 2011, considering all relevant evidence, including evidence regarding his present condition, to determine the progression of his degenerative impairments. Parker , 597 F.3d at 925. Therefore, SSR 83–20 is inapplicable.

B. Appeals Council's Finding Regarding Neck and Shoulder Pain is Substantially Supported

Schloesser contends that the Appeals Council's finding that he did not suffer from severe impairments of cervical radiculopathy, major joint dysfunction, and left shoulder surgery on or before his date last insured is not supported by substantial evidence. Schloesser believes that his reports of neck and shoulder pain to his doctors shortly after his date last insured provide the necessary proof to show he suffered from these impairments and that they were severe. According to Schloesser, failure to credit these reports in his favor means that the Appeals Council's decision is not supported by substantial evidence. But Schloesser misinterprets the Appeals Council's finding. Contrary to his assertion that the finding assumes that his pain only surfaced after his date last insured, the Appeals Council did not disbelieve that he experienced some pain prior to his date last insured. Instead, it only found that the neck and shoulder impairments he was diagnosed with months later were not severe enough to disable him to the extent he alleged as of his date last insured.

The Appeals Council's discussion of the record clearly builds a logical bridge between the relevant evidence and its determination that Schloesser did not suffer from severe impairments of cervical radiculopathy, major joint dysfunction, and history of left shoulder surgery. The record shows that Schloesser visited four different doctors for a total of six visits between his alleged onset date and his date last insured. Yet, he never discussed neck or shoulder pain at any of these visits. Instead, the visits primarily concerned his lower back pain, including a diagnosis of degenerative disc disease in his lower spine. Further, as noted by the Appeals Council, Schloesser told Dr. David N. Crowther that he had discontinued use of ibuprofen (which he had been using to manage his back pain) on August 17, 2011, just one month before his date last insured. It was not until December 2011 that he first complained to any physician about his neck and shoulder pain.

Even then, an MRI of Schloesser's neck from December 2011 revealed no evidence of disc herniation at any level and only mild cervical degeneration with radicular symptomology. Yet, each time he visited Dr. Crowther to complain of neck and shoulder pain in December 2011, he also discussed his plan to apply for disability at length and asked Dr. Crowther to support him in his disability claim. But, Dr. Crowther, who observed that Schloesser was in no distress, explained in his visitation notes that he was "unable to provide a blanket unsupported statement stating [Schloesser] is disabled." Dr. Crowther advised Schloesser to attend physical therapy for his neck and shoulder pain. And, in January 2012, Schloesser's records reflect that he was not taking any pain medications and his primary reason for visiting Dr. Crowther was again to provide a letter attesting to his disability. Further, as noted by the Appeals Council, Schloesser's radicular symptomology resolved in February 2012. While we agree with Schloesser that degeneration is a progressive disease, we fail to see how a diagnosis of mild degeneration, which improved within a few short months, supports his assertion that he suffered from a severe neck impairment several months prior. And there was no testimony at the hearing before the ALJ asserting that Schloesser's neck pain was debilitating on or before his date last insured.

As concerns his left shoulder: though he was diagnosed with subacromial impingement syndrome in February 2012, Dr. Crowther noted that Schloesser demonstrated a full range of motion in all four limbs (including his left shoulder). Dr. Todd J. Duelleman noted in February 2012 that Schloesser had a strength of 4/5 in his left shoulder, had a decreased range of motion, and had increased pain with mid-range movements. Schloesser was instructed to attend seven physical therapy sessions. However, physical therapy reports from March 2012 show that Schloesser cancelled his final visit. At his sixth visit, the physical therapist noted that he had experienced some pain resolution (though he still suffered from some increased pain from lifting activities). And despite Schloesser's assertion that that the Appeals Council ignored his progressively degenerative left shoulder impairment, it clearly did not. The Appeals Council did not disbelieve that Schloesser experienced pain. It only found that his shoulder pain was not severe; and it still accounted for his left shoulder pain in determining that his RFC limited him to no overhead reaching bilaterally and frequent but not constant reaching in all other directions with his nondominant left upper extremity. See Mitze v. Colvin, 782 F.3d 879, 881 (7th Cir. 2015) (noting that ALJ had not denied that claimant was in pain but instead "didn't believe that the pain was severe enough to disable her to the extent she claimed").

Finally, Schloesser argues that, even if his 2011 and 2012 medical records do not provide the necessary support for his claim that he suffered from such severe impairments on or before September 30, 2011, his 2013 medical records do. The record does not contain any medical records from March 2012 until Schloesser began seeing Dr. Claire Natividad in January 2013. Even assuming these records from two years after his date last insured are relevant, Schloesser still did not complain of shoulder or neck pain and much of the focus of his visits remained his lower back pain.Because the Appeals Council cited to relevant evidence adequate to support its conclusion that Schloesser did not suffer from severe impairments of cervical radiculopathy, major joint dysfunction, and left shoulder surgery on or before his date last insured, we find this determination to be supported by substantial evidence.

C. RFC Determination Supported by Substantial Evidence

In determining his RFC, the ALJ credited Schloesser's subjective reports of the effect of his pain to determine that he may be off-task more than 10% of the workday and may require unscheduled breaks. This determination was the deciding factor in the ALJ's favorable finding that Schloesser was disabled. But the Appeals Council found that Schloesser's subjective complaints of the effect of his pain and medications on his ability to sustain work were not supported by the record. In making this adverse credibility finding, it explained that "[m]ost significantly, the record documents infrequent treatment before the expiration of the date last insured [and] [t]he medical records that do exist ... document largely unremarkable findings and some improvement of [Schloesser's] condition." So, the Appeals Council found that his RFC does not include being off-task more than 10% or needing unscheduled breaks. Schloesser contends that this resulted in a reversible error for three reasons: (1) his lack of insurance contributed to his infrequent treatment, which the Appeals Council failed to consider; (2) his August 2013 testimony before the ALJ regarding his need to lay down every half-an-hour should have been credited by the Appeals Council because it was credited by the ALJ; and (3) the Appeals Council failed to account for his non-severe impairment of anxiety, which, although unstated by the ALJ, was implicitly the reason the ALJ included the limitations of being more than 10% off-task and needing unscheduled breaks.

When the Appeals Council rejects an ALJ's credibility findings, "it should do so expressly and state its reasons for doing so." Bauzo , 803 F.2d at 922. The reasons given must be sufficiently specific and supported by the record. Engstrand , 788 F.3d at 660 (citing Curvin , 778 F.3d at 651 ). Here, the Appeals Council provided two specific reasons for its adverse credibility finding: (1) Schloesser's infrequent treatment before his date last insured; and (2) the largely unremarkable findings and improvement of his condition in the medical records that do exist. We begin with the Appeals Council's first reason. Schloesser contends that his lack of insurance was the primary reason for his infrequent treatment. Inability to afford treatment is one reason that may "provide insight into the [claimant's] credibility [,]" Craft v. Astrue , 539 F.3d 668, 679 (7th Cir. 2008), and failure to consider a claimant's reasons for not seeking treatment is erroneous. Thomas v. Colvin , 826 F.3d 953, 961 (7th Cir. 2016). However, we are unpersuaded by Schloesser's contention that the Appeals Council failed to consider this proffered reason. Despite Schloesser's assertion that "[t]he record is replete with references to [his] lack of insurance during the relevant period," he cites only to his self-completed Disability Report, which states:

The reason I do not have doctor records for several years, is because I went without insurance after getting divorced, and could not afford medical treatment. We moved to a completely new area on October 15th 2009, and had to go without insurance again for some time, until it was made available thru my current wifes employer. In February of 2011 my lower back became so severely aggravated I went to Doctor Tang at the Aspirus clinic, which of course was the first time seeing any doctor near the new town in which we lived.... We no longer go to the Aspirus clinics due to higher percentage of insurance coverage in the Riverview clinics, which is associated with where my wife works.

R. at 191 (errors in original). But this statement does not indicate that he lacked insurance during the relevant time period—from his alleged onset date (January 1, 2011) to his date last insured (September 30, 2011). At best, it shows that he may have lacked insurance at various times before 2011. During the nine months between his alleged onset date and his date last insured, Schloesser visited doctors a total of six times; yet, in the three months from December 2011 to February 2012, Schloesser visited doctors nine times, including several visits during which he discussed his plans to file for disability and asked Dr. Crowther to provide him with a letter attesting to his disability. Schloesser has provided no evidence to show that he lacked insurance during the time period between his alleged onset date and date last insured that may explain why he visited the doctor significantly more over a shorter period of time after his date last insured. There is also no indication that the Appeals Council did not consider the statements in his disability report in reaching its determination, as it indicated that it had reviewed the entire record before coming to a decision. And Schloesser has been and continues to be represented by counsel, so we assume he "is making his strongest case for benefits." Glenn v. Sec'y of Health and Human Servs. , 814 F.2d 387, 391 (7th Cir. 1987).

Next, the Appeals Council was not required to find in Schloesser's favor just because the ALJ did, even if the ALJ's decision is supported by substantial evidence. The question we consider on review is not whether an alternative finding may also be supported by substantial evidence. Rather, we ask whether the final agency finding—here the Appeals Council's finding—is supported by substantial evidence. See Scheck v. Barnhart , 357 F.3d at 699 ("[T]he [Appeals Council's] decision, if supported by substantial evidence, will be upheld even if an alternative position is also supported by substantial evidence."). Further, disregarding Schloesser's subjective testimony where it contradicts with contemporaneous reports he made to his physicians and their independent observations is permissible. Schmidt v. Astrue , 496 F.3d 833, 843-44 (7th Cir. 2007). And, here, though Schloesser testified that, starting in 2010, he must lay down every 20-30 minutes, his 2011 medical records show otherwise.

In February and March 2011, his doctors noted that Schloesser experienced no tenderness in his lower back, was not using his prescribed painkillers, had stated his symptoms had improved, and rated his lower back pain at 2 on a 10-point scale. And, again, in August 2011, Schloesser reported that he discontinued the use of ibuprofen and felt well. In fact, in January 2012, Dr. Luke J. Budleski noted that Schloesser's gait pattern was normal and Dr. Crowther noted that he was not taking any pain medications. It was not until May 2013, 20 months after his date last insured, that Dr. Natividad observed that Schloesser could only sit and stand/walk less than 2 hours total in an 8-hour working day, and that he would need to take more than 5 unscheduled breaks of approximately 30 minutes per day. Contrary to Schloesser's assertion that his documented disability from May 2013 clearly indicates that he must have also been disabled in September 2011, Dr. Natividad specifically states that the earliest time that these limitations would have applied was May 31, 2013 and noted that she "cannot comment on dates earlier than this." A reasonable mind could certainly accept the medical evidence (both the evidence from 2011 and from 2013) to conclude that the progression of Schloesser's degenerative pain was such that he would not have required being off-task more than 10% of the workday or needing unscheduled breaks until May 2013, 20 months after his date last insured. Therefore, the Appeals Council's second specific reason—that the medical records documented largely unremarkable findings and improvement of his condition—is supported by the record.

In a final attempt to assert that his RFC required such limitations, Schloesser argues that the ALJ's findings that he had nonsevere impairments of affective and anxiety disorders were the implied reasons for inclusion of the 10% off-task and unscheduled breaks limitations. However, not only did the ALJ never state this, but she specifically found that "[t]he limitation to additional unscheduled breaks and being off task more than 10% of the workday are supported by the claimant's generally credible statements regarding his pain ." But as we explain above, the reasons for the Appeals Council's adverse credibility finding are specific and supported by the record. And, again, just because the ALJ found that Schloesser suffered from nonsevere impairments of affective and anxiety disorders, the Appeals Council was not required to agree with this finding. Therefore, the Appeals Council's finding that his RFC did not include being off-task more than 10% or needing unscheduled breaks, is substantially supported by the record.

III. CONCLUSION

The judgment of the district court is AFFIRMED .