In this case, we join the majority of other circuits in concluding that the medical improvement standard, as created in 42 U.S.C. 423(f) and defined by 20 C.F.R. 404.1594(b)(1), applies in "closed period" cases in which a disability claimant is found to have been disabled for a finite period of time.1 Exercising jurisdiction pursuant to 28 U.S.C. 1291, we affirm in part, reverse in part, and remand for further proceedings consistent with this opinion.

* Appellant is a disability claimant who filed an application for disability benefits in April 1993, which was approved for a closed period of disability spanning from December 8, 1991, through December 31, 1992. A request for reconsideration was denied on July 19, 1993. Following this denial, claimant requested a hearing before an ALJ. The ALJ agreed with the earlier determination that claimant was only eligible for benefits for the closed period noted above because he was not disabled after December 31, 1992. The Appeals Council denied review, and claimant filed a complaint in federal district court. The district court affirmed the commissioner's decision.

We review the commissioner's decision only to determine whether substantial evidence supports that decision and whether the applicable legal standards were applied correctly. See Glenn v. Shalala, 21 F.3d 983, 984 (10th Cir. 1994). "We examine the record as a whole, including whatever in the record fairly detracts from the weight of the [commissioner's] decision and, on that basis, determine if the substantiality of the evidence test has been met." Casias v. Secretary of Health & Human Servs., 933 F.2d 799, 800-01 (10th Cir. 1991).

On appeal, claimant asserts that the district court erred in holding that the medical improvement standard does not apply to closed period cases; that the ALJ's determination of his residual functional capacity was flawed; and that a proper hypothetical question was not asked of the vocational expert. We address each claim in turn.

* After a claimant has been receiving disability benefits for some period, the Social Security Administration is required to review his case periodically to determine whether there has been any medical improvement in the claimant's condition and whether that improvement affects his ability to work. See 20 C.F.R. 404.1594. If the benefit recipient's condition has improved, his eligibility to receive those benefits may terminate. See Jones v. Shalala, 10 F.3d 522, 524 (7th Cir. 1993).

The following standard of review informs the decision to terminate benefits:

A recipient of benefits . . . may be determined not to be entitled to such benefits on the basis of a finding that the physical or mental impairment on the basis of which such benefits are provided has ceased, does not exist, or is not disabling only if such a finding is supported by--

(1) substantial evidence which demonstrates that--

(A) there has been any medical improvement in the individual's impairment or combination of impairments (other than medical improvement which is not related to the individual's ability to work), and

(B) the individual is now able to engage in substantial gainful activity . . . .

42 U.S.C. 423(f). The Social Security Administration's regulations define medical improvement as

any decrease in the medical severity of [the] impairment(s) which was present at the time of the most recent favorable medical decision that [the claimant was] disabled or continued to be disabled. A determination that there has been a decrease in medical severity must be based on changes (improvements) in the symptoms, signs and/or laboratory findings associated with [the] impairment(s).

20 C.F.R. 404.1594(b)(1). The medical improvement standard clearly applies when a disability award has become final and the commissioner brings an action to terminate those benefits. See Glenn, 21 F.3d at 984-85. The issue here, however, is whether the medical improvement standard applies in closed period cases, such as this one.2

We begin with a brief review of the statutory and regulatory development of the medical improvement standard. After a period in which the Social Security Administration had abandoned the medical evidence standard in favor of a current evidence standard, Congress enacted the Social Security Disability Benefits Reform Act of 1984 ("Reform Act"). See generally 93 A.L.R. Fed. 161 (1989). The Reform Act adopted the medical improvement standard, which now appears at 42 U.S.C. 423(f), and defined it in this way:

[T]he term 'action relating to medical improvement' means an action raising the issue of whether an individual who has had his entitlement to benefits . . . based on disability terminated (or period of disability ended) should not have had such entitlement terminated (or period of disability ended) without consideration of whether there has been medical improvement in the condition of such individual . . . since the time of a prior determination that the individual was under a disability.

See 42 U.S.C. 423 note (1984 Acts) (Pub. L. No. 98-460 Sec. 2(d)(6)).

The issue of whether claimants involved in closed period cases could have their cases remanded for consideration under the medical improvement standard was not clearly resolved in the Reform Act, however, and the circuits that have considered the issue have disagreed. The Eighth Circuit has held that a closed period case was not an "action relating to medical improvement" for purposes of the Reform Act, and read the remand provisions of the Reform Act to apply only to cases "'of a prior determination that the individual was under a disability.'" Camp v. Heckler, 780 F.2d 721, 721-22 (8th Cir. 1986) (quoting 42 U.S.C. 423 note (1984 Acts) (Pub. L. No. 98-460 Sec. 2(d)(6))). "Prior determination," the court held, "is more naturally read as referring to a previous decision in favor of disability, followed by the claimant's receipt of benefits, further followed by a new proceeding resulting in cessation or termination on the ground of medical improvement." Id. at 722.3 In a later case, the Eighth Circuit followed Camp in refusing to apply the medical improvement test to a closed period case. See Ness v. Sullivan, 904 F.2d 432, 435 n.4 (8th Cir. 1990). More recently, however, the circuit applied the medical improvement standard to a closed period case without discussing or citing Camp. See Burress v. Apfel, 141 F.3d 875, 878-880 (8th Cir. 1998). Thus, the law in the Eighth Circuit regarding this issue seems presently unsettled.

Other circuits have rejected the Eighth Circuit's approach in Camp. The Eleventh Circuit, for example, explicitly criticized Camp for ignoring the legislative history of the Reform Act and "the broad remedial policies underlying the Disability Amendments," in which it discerned that Congress intended to reach closed period claimants. Pickett, 833 F.2d at 292 & 293 n.4. See also Jones v. Shalala, 10 F.3d 522, 523-24 (7th Cir. 1993) (applying medical improvement standard in its review of closed period case); Chrupcala v. Heckler, 829 F.2d 1269, 1274 (3d Cir. 1987) ("Fairness would certainly seem to require an adequate showing of medical improvement whenever an ALJ determines that disability should be limited to a specified period."); cf. Bowling v. Shalala, 36 F.3d 431, 435 (5th Cir. 1994) (applying traditional five-step sequential analysis to claimant's evidence in a closed period case). We are persuaded by these other circuits that applying the medical improvement standard to cases involving a closed period of disability is consistent with the language and legislative purpose in the Reform Act.4

In the case before us, the magistrate judge whose order the district court adopted assumed that the medical improvement standard did not apply in closed period cases. Furthermore, the magistrate found that even if the medical improvement did apply, the ALJ's findings support the conclusion that claimant experienced medical improvement and was able to work after that improvement. Having concluded that the medical improvement standard does apply, we must now determine whether the ALJ, who referred to the standard in his decision, applied it correctly in claimant's case. As explained below, we conclude that although the Administrative Law Judge ("ALJ") at least implicitly applied the proper standard in awarding appellant-claimant Dwayne Shepherd only a closed period of disability benefits for impairments sustained as the result of a motorcycle accident, the record lacks substantial evidence to support the ALJ's conclusion that claimant's disability ceased by December 31, 1992. Instead, the record demonstrates that claimant had achieved medical improvement and could perform substantial work by July 7, 1993.

In order to determine whether disability continues or ends, the commissioner must determine "if there has been any medical improvement in [a claimant's] impairment(s) and, if so, whether this medical improvement is related to [the claimant's] ability to work." 20 C.F.R. 404.1594(a).5 The regulations contain definitions and examples of when medical improvement is related to the ability to do work, and when it is not so related. See id. at 404.1594(b)(2) and (b)(3).

To apply the medical improvement test, the ALJ must first compare the medical severity of the current impairment(s) to the severity of the impairment(s) which was present at the time of the most recent favorable medical decision finding the claimant disabled. See id. at 404.1594(b)(7). Then, in order to determine that medical improvement is related to ability to work, the ALJ must reassess a claimant's residual functional capacity (RFC) based on the current severity of the impairment(s) which was present at claimant's last favorable medical decision. See id. at 404.1594(c)(2). The ALJ must then compare the new RFC with the RFC before the putative medical improvements. The ALJ may find medical improvement related to an ability to do work only if an increase in the current RFC is based on objective medical evidence. See id.

In this case, the ALJ was asked to review a determination, made at an earlier stage in the administrative process, that claimant was only disabled from December 8, 1991, through December 31, 1992. The ALJ agreed that claimant was disabled by December 8, 1991, because of a nonunion of a fracture of the tibia, an impairment meeting the listing at 20 C.F.R., Pt. 404, Subpt P., App. 1 1.11. The ALJ also found, however, that claimant was no longer disabled after December 31, 1992. As support for this conclusion, the ALJ cited the opinions of consultative examiners who found limitation primarily in the lower extremities and none of consequence in the upper extremities. See Appellant's App. at 17. These examinations, however, occurred in 1993 and 1994.

The ALJ points to no evidence tied to December 31, 1992, to support his conclusion that claimant's disability ended as of that date. There is no evidence in the record of medical improvement of the fractured tibia, substantiated by changes in signs, symptoms, or laboratory findings, which took place by December 31, 1992, sufficient to support a conclusion that disability had ceased. Nor is there an evidentiary basis upon which the ALJ could have compared the severity of claimant's leg injury on December 31, 1992, with the same condition as it existed on December 8, 1991. Similarly, no evidence allows a comparison of claimant's RFC on those two dates. Although the ALJ correctly recognized that the medical improvement test applied in this closed period case, insufficient evidence supports his conclusion that such improvement had occurred by December 31, 1992.

The same cannot be said, however, of claimant's condition in July 1993. By that date, a consultative examination revealed normal upper extremities except for slight stiffness in the right elbow, normal range of motion of the joints of the lower extremities except for the right ankle, normal peripheral pulses, no ankle swelling, joints not tender to palpitation, full range of motion of the knees (albeit accompanied by crackling noises), some rotation of the knees, and evident deformities of the legs from the fractured tibias but no tenderness. See Appellant's App. Vol. II at 243-44. Muscle strength in the hands, forearms, arms, and legs were normal and equal bilaterally. See id. at 244. Claimant could walk at a normal pace and with good stability, although he required a leg brace on his right leg. See id. Claimant expressed an interest in retraining to become a motorcycle mechanic, and the consultative physician believed he could be motivated to seek alternative occupation and training. See id. Furthermore, Dr. Luther Woodcock had concluded that by July 1993, claimant had the residual functional capacity to perform sedentary work because he could occasionally lift ten pounds, frequently lift five to ten pounds, stand at least two hours in an eight-hour workday, sit about six hours in an eight-hour workday and had unlimited ability to push and/or pull. See id. at 95-96; see also 20 C.F.R. 404.1567(a) (describing the requirements for sedentary work).

Thus, there was a period between December 1992 and July 1993 during which claimant may have been eligible for benefits. Because the record is unclear, however, regarding what portion of this period saw claimant incarcerated and thus ineligible for benefits, we must remand to the district court for further development of this factual issue.

In his second issue on appeal, claimant takes issue with the ALJ's conclusion that he can do some light work and all sedentary work. Claimant first takes exception to the ALJ's credibility determination. Claimant's argument to the district court was that the ALJ's credibility findings fail to comply with the mandate of Kepler v. Chater, 68 F.3d 387 (10th Cir. 1995), in that they do not link claimant's credibility with specific evidence. We disagree.

The ALJ thoroughly explained his credibility determination, citing inconsistencies between claimant's testimony regarding his inability to stand because of pain and reports given to the consultative examiners, evidence that he had been doing some mechanic work and odd jobs around the house, the fact that he takes only aspirin and had not seen a doctor since March 1993, and the fact that his physical therapy sessions were canceled because he failed to show up for them. This identification of specific evidence complies with the requirements of Kepler.6

Finally, claimant contends that the ALJ failed to ask the proper hypothetical of the vocational expert. The ALJ asked the vocational expert to consider a person "of the same age, education, sex, background, training and experience of this claimant, and I first want you to assume that the only limitation is no prolonged time on the feet or prolonged use of the feet or legs." Appellant's App. Vol. II at 61. Claimant argues that the ALJ should also have included his impairments with walking, sitting, stamina, severe pain, cognitive skills, concentration, memory, and the need to rest during the day.

The record does contain evidence that claimant would have problems walking long distances. We conclude, however, that the hypothetical question described above included a walking dimension and that the ALJ adequately explained his refusal to credit claimant's complaints of severe pain. The other alleged impairments find no evidence in the record other than claimant's testimony which, by itself, is insufficient to establish the existence of an impairment. See Musgrave v. Sullivan, 966 F.2d 1371, 1376 (10th Cir. 1992). The ALJ is only required to ask hypotheticals encompassing impairments that find support in the record. See Jordan v. Heckler, 835 F.2d 1314, 1317 (10th Cir. 1987).

As noted above, the judgment is AFFIRMED in part and REVERSED in part, and the case is REMANDED for further proceedings.