Darrell E. Jenkins appeals the denial of his application for social-security disability benefits. He argues that the Administrative Law Judge (ALJ) based the denial on insubstantial evidence and improperly evaluated his allegations of subjective pain. The District Court upheld the denial. We reverse.

Mr. Jenkins is a 55-year-old man with a high-school education. From 1978 to 1995, he ran a dairy farm. He has no work experience outside farming. Mr. Jenkins claims that he became disabled on January 1, 1995, because of severe pain and weakness in his upper extremities. On March 13, 1995, he filed an application for disability benefits. His claim was denied initially and on reconsideration. He filed a request for a hearing, which was held on August 8, 1996.

The ALJ decided that Mr. Jenkins was not entitled to disability benefits. The ALJ found that Mr. Jenkins had a severe impairment but not one that qualified him for benefits under the impairments listed in the regulations. The ALJ found that Mr. Jenkins was unable to perform his past relevant work. The ALJ decided, however, that given his residual functional capacity, Mr. Jenkins would be able to do light and sedentary work. The ALJ did not believe Mr. Jenkins's testimony that his severe pain limited his capacity to do such work. The ALJ found Mr. Jenkins's testimony inconsistent, and credited the assessment of a non-treating physician who determined that Mr. Jenkins had a greater capacity to lift and grasp than he claimed. The ALJ believed this assessment because he found it consistent with the reports of Mr. Jenkins's treating physicians. Based on this assessment and the testimony of a vocational expert, the ALJ concluded that Mr. Jenkins was not entitled to disability benefits. Mr. Jenkins appealed the ALJ's decision.

The Appeals Council considered additional medical evidence not before the ALJ. After the hearing, on November 18, 1996, Mr. Jenkins had begun regular treatments with Dr. Janet Schwartzenberg, a rheumatologist. In deciding Mr. Jenkins's request for review, the Appeals Council considered a Medical Source Statement from Dr. Schwartzenberg and her reports from these post-hearing treatments. The Appeals Council, however, determined that this evidence did not provide a basis for altering the ALJ's decision. The Appeals Council denied Mr. Jenkins's request for review.

We review the decision of the ALJ to determine whether his findings are supported by substantial evidence on the record as a whole. Pfitzner v. Apfel, 169 F.3d 566, 568 (8th Cir. 1999). In addition to evidence before the ALJ, the record includes evidence that was submitted after the hearing and considered by the Appeals Council in denying review. Riley v. Shalala, 18 F.3d 619, 622 (8th Cir. 1994). Evaluating such evidence requires us to determine how the ALJ would have weighed the newly submitted evidence if it had been presented at the original hearing. Id. In this case, the new evidence considered by the Appeals Council provides substantial support for Mr. Jenkins's arguments on appeal: first, that the ALJ's residual-functional-capacity finding was not supported by substantial evidence, and second, that the ALJ incorrectly evaluated the claims of disabling pain.

The medical reports submitted after the hearing are evidence that Mr. Jenkins has less residual functional capacity than the ALJ concluded. Adopting the assessment of a non-treating physician, the ALJ concluded that Mr. Jenkins:

"can lift and/or carry 10 pounds frequently and 20 pounds occasionally; can stand and/or walk about six hours . . . can sit about six hours . . . has limited ability to push and/or pull; can never climb ladders, ropes, or scaffolding; and has limited overhead reaching" (R. at 36).

By contrast, Dr. Schwartzenberg, a treating physician, concluded that Mr. Jenkins could lift or carry only five pounds frequently (not ten), that he could carry or lift no more than five pounds even occasionally (not twenty), and that he could stand or walk just five hours (not six) and not more than one hour continuously (R. at 175-76). In summarizing Mr. Jenkins's impaired capacities, Dr. Schwartzenberg stated that the problem with his shoulders "severely limits his ability to do anything" (R. at 177).

Because Dr. Schwartzenberg's assessment conflicts with the assessment of the non-treating physician, we must decide how the ALJ would have weighed them relatively. Though the non-treating physician had never seen Mr. Jenkins, the ALJ credited his assessment because it was consistent with the general diagnoses and observations of the treating physicians (R. at 35). The treating physicians' diagnoses and observations, however, did not include specific assessments of Mr. Jenkins's capacity to lift, carry, sit, and stand.1 The non-treating physician's specific judgements of Mr. Jenkins's capacities were inferences from other physicians' much more general findings. By contrast, Dr. Schwartzenberg saw and treated Mr. Jenkins four times before making her assessment. She is the only treating physician who has assessed and quantified the claimant's capacity to lift, carry, sit, and stand. Because her assessment was not available at the hearing, the ALJ had no choice but to rely on the assessment of a non-treating physician.

If he had possessed the new evidence, however, we think the ALJ would have adopted Dr. Schwartzenberg's assessment of Mr. Jenkins's residual functional capacity rather than the non-treating physician's assessment. A treating physician's opinion should not ordinarily be disregarded and is entitled to substantial weight. Ghant v. Bowen, 930 F.2d 633, 639 (8th Cir. 1991). By contrast, "[t]he opinion of a consulting physician who examines a claimant once or not at all does not generally constitute substantial evidence." Kelley v. Callahan, 133 F.3d 583, 589 (8th Cir. 1998). There is no other evidence in the record to support the ALJ's residual-functional-capacity finding besides the non-treating physician's assessment. This assessment alone cannot be considered substantial evidence in the face of the conflicting assessment of a treating physician. Henderson v. Sullivan, 930 F.2d 19, 21 (8th Cir. 1991). Moreover, Dr. Schwartzenberg's reports undermine the ALJ's sole stated reason for crediting the non-treating physician's assessment, its consistency with the findings of the treating physicians (R. at 35). If the ALJ had possessed Dr. Schwartzenberg's assessment, he could not have found this consistency.

Accordingly, we hold that the residual-functional-capacity assessment adopted by the ALJ was not supported by substantial evidence. Since the vocational expert's testimony was based upon this assessment, we also hold that his testimony was not substantial evidence that Mr. Jenkins could perform other substantial gainful activity. Ness v. Sullivan, 904 F.2d 432, 436 (8th Cir. 1990).

In addition, the new evidence supports the credibility of Mr. Jenkins's testimony that he suffers from disabling pain. The ALJ adopted the finding of the non-treating physician that Mr. Jenkins's complaints of disabling pain were "inconsistent" with objective medical findings (R. at 35). The ALJ considered the absence of objective medical support for Mr. Jenkins's claims of pain an "important factor" in discrediting his testimony (ibid.). The non-treating physician's conclusion, however, is again in conflict with the assessment of a treating physician, Dr. Schwartzenberg.

Dr. Schwartzenberg observed "severe spasm of the musculature . . . secondary to pain" when she moved the claimant's shoulders beyond 15 degrees (R. at 187), "muscle spasm and twitching that was obviously unintentional in nature" after movement of the triceps (ibid.), and "tenderness and muscle spasm in the thoracic area in the paraspinal musculature" (R. at 182). In light of these personal experiences with the patient, her examinations, and her physical findings, Dr. Schwartzenberg concluded that Mr. Jenkins's complaints of "constant neck, shoulders, arms and back pain, worse [with] any attempt at movement or touch" were credible (R. at 179). She found it credible that he would experience "pain from shoulders radiating down arms into hands" that would cause him to "frequently drop things of any weight . . .," and that he would experience constant fatigue and need to lie down unpredictably three to four times daily for at least 45 minutes to cope with these symptoms.

Dr. Schwartzenberg's opinion conflicts with the assessment of the non-treating physician that Mr. Jenkins's complaints were inconsistent with his objective medical condition. Again, we must determine how the ALJ would have weighed the assessments relatively. For the same reasons discussed above, we hold that if the ALJ had possessed the new evidence, he would have adopted the assessment of the treating physician, Dr. Schwartzenberg, over the assessment of the non-treating physician.

We also hold that the ALJ would have credited Mr. Jenkins's testimony of disabling pain. Without the support of the non-treating physician's assessment, the ALJ's remaining reasons for discrediting that testimony are insubstantial. By contrast, Dr. Schwartzenberg's reports and other indicia of credibility strongly support Mr. Jenkins's testimony of disabling pain.

The ALJ gave only two reasons for discrediting Mr. Jenkins's testimony besides the absence of objective medical evidence. First, the ALJ found that Mr. Jenkins did not seek medical treatment between May, 1995, and August, 1996. Mr. Jenkins, however, did see a chiropractor during that time (R. at 35). Moreover, Mr. Jenkins had been told on his last visit to a physician that his condition would likely improve if he continued his physician's prescribed regimen of pain medication, therapeutic exercise, and rest, advice which Mr. Jenkins apparently did follow during this time (R. at 73, 150). Even if we characterize Mr. Jenkins's actions as a failure to seek medical treatment, we cannot agree that his reliance on a chiropractor and his physician's prescribed course of therapy were inconsistent with severe pain. If following a doctor's advice and seeking additional help are not probative of severe pain, then they are, at least, not inconsistent with it. In addition, the claimant did later, of course, seek medical treatment from Dr. Schwartzenberg.

The ALJ also found that Mr. Jenkins's use of over-the-counter pain medication at the time of the hearing was not suggestive of disabling pain. On this record, however, this fact is inconclusive. Mr. Jenkins originally took prescription pain medication. His physician later prescribed over-the-counter medication when the original prescriptions proved ineffective (R. at 160). The physician's withdrawal of prescription pain relief was based on his stated belief that it was not working, rather than a belief that Mr. Jenkins didn't need effective pain relief. If the physician's decision had been based, for example, on a disbelief of Mr. Jenkins's claims of pain or upon the physician's success in treating him, then it would be suggestive of an absence of severe pain. In this case, however, the physician's decision signaled only the ineffectiveness of the prescription, not the absence of pain.

Mr. Jenkins's claims of disabling pain are also supported by his history of past hard work and now limited daily activities. Under Polaski v. Heckler, 739 F.2d 1320 (8th Cir. 1984), in order properly to evaluate subjective claims of pain, the ALJ is required to take into account, inter alia, the claimant's work history and limitations of his daily activities. In this case, these factors support Mr. Jenkins's claims.

Before his impairment developed, Mr. Jenkins was a dairy farmer. From his high school graduation until 1993, he worked twelve- to fourteen-hour days filled with heavy physical labor: milking cows, shoveling feed, putting up hay, and repairing fences (R. at 60). In the evenings, he did all the yard work at home (R. at 74). Even in 1993, when he had lost much of his upper-body strength and flexibility, Mr. Jenkins continued farming with his wife's help (R. at 158). Despite these efforts, today, Mr. Jenkins's farm lies fallow. He and his wife testified that his days are spent alone, watching television, sometimes sitting and sometimes walking around their lawn, occasionally visiting his mother (R. at 32, 65-69, 75-6). Mr. Jenkins has given up his only hobbies: guitar playing, hunting, and fishing (R. at 66, 68-69). They testified that he requires help to dress himself, comb his hair, bathe, open building doors, or make a meal (R. at 32, 63-65, 67-69, 74-76.) His current limited activities support his claims of disabling pain, especially in light of his past hard-working lifestyle.

If the ALJ had possessed the evidence considered by the Appeals Council, we conclude that he would have approved Mr. Jenkins's claims for disability benefits. Any other conclusion would have lacked the support of substantial evidence on the record as a whole. The ALJ would have had to reject the residual-functional-capacity assessment of the non-treating physician, and would have had to abandon his conclusion that Mr. Jenkins's claims of disabling pain were inconsistent with objective medical evidence. In addition, the ALJ would have had no good reason to discredit Mr. Jenkins's claims of disabling pain, supported as they are by the assessment of his latest treating physician, his past hard work, and his now limited daily activities.

Accordingly, the judgment of the District Court is reversed, and the cause remanded with instructions to remand to the Commissioner for calculation and award of benefits.

It is so ordered.

LOKEN, Circuit Judge, dissenting.

In my view, timing is the critical issue in this case. Darrell Jenkins filed his application for disability benefits in March 1995, alleging a disability onset date of January 1, 1995. The hearing was held in August 1996, and the ALJ issued his adverse decision in late September. Mr. Jenkins first saw Dr. Janet Schwartzenberg in November 1996. Thus, she was not his treating physician during the time in question. Her reports, which were submitted to the Commissioner's Appeals Council in March, April, and May of 1997, did not state whether, in her opinion, the physical impairments, complaints of pain, and work limitations that she observed existed at their current level of severity at the alleged onset date, or on the date of Mr. Jenkins's hearing. Thus, I agree with the district court that Dr. Schwartzenberg's reports did not undermine the substantial evidence in the administrative record supporting the ALJ's adverse determination. It may well be that Mr. Jenkins was disabled when Dr. Schwartzenberg treated him. But that is grounds for a new disability application, not for reversing a decision by the Commissioner that is supported by substantial evidence on the administrative record as a whole. For these reasons, I respectfully dissent.