WILLIAM A. FLETCHER, Circuit Judge:

Leo Orn filed an application for Social Security benefits claiming that he was unable to work because of disability. Orn has been diagnosed with several disorders, including asthma, severe chronic obstructive pulmonary disease, diabetes, sleep apnea, and morbid obesity. Following a remand from the Appeals Council, the Administrative Law Judge ("ALJ") agreed with Orn that he was unable to perform his past work. But, after rejecting the opinions of Orn's treating physicians and Orn's testimony, the ALJ concluded that the government had met its burden to prove that Orn was able to perform other work that exists in the economy. The Appeals Council and district court affirmed.

We hold that the ALJ did not give "`specific, legitimate reasons . . . that are based on substantial evidence in the record'" for dismissing the opinions of Orn's two treating physicians. See Thomas v. Barnhart, 278 F.3d 947, 957 (9th Cir.2002) (quoting Magallanes v. Bowen. 881 F.2d 747, 751 (9th Cir.1989)). We also hold that the ALJ erred in discrediting Orn's testimony. The ALJ's reasons for discrediting Orn's testimony are not "clear and convincing." See Morgan v. Comm'r of the Soc. Sec. Admin., 169 F.3d 595, 599 (9th Cir.1999). When Orn's testimony and the opinions of his treating physicians are credited, Orn has established that he is disabled. See McCartey v. Massanari, 298 F.3d 1072, 1076-77 (9th Cir.2002). We remand for a calculation of benefits.

I. Background

Orn worked in the food service industry for approximately twenty years. On May 23, 2000, while working in a warehouse to prepare pallets for shipment, Orn suffered an asthma attack and collapsed. He was taken to the emergency room and was hospitalized for a day.

After his hospitalization, Orn continued to receive emergency and outpatient medical treatment for his chronic respiratory disorders. Orn has been diagnosed with asthma and severe chronic obstructive pulmonary disease. Both conditions, according to one of his treating physicians, are "progressively worsening." His clinical examinations consistently reveal wheezing and his pulmonary function tests are abnormal. He is on several medications for his respiratory diseases.

Orn was hospitalized for respiratory problems again in 2003. That hospital stay lasted a week. He was discharged with instructions to receive supplemental oxygen twenty-four hours a day. Orn continues to require continuous supplemental oxygen. The oxygen is delivered to his nose through tubing attached to a tank.

Orn has several other medical conditions. He has been diagnosed with sleep apnea. He testified that during the night he is "constant[ly] waking up because [he's] always choking" from the sleep apnea. Orn often sleeps in fifteen-to thirty-minute intervals. During the day, he is "always tired." While he was still working, Orn sometimes fell asleep and his "coworkers [had] to awaken" him.

Orn has diabetes which he treats with oral medication. He has problems with his circulation. He also suffers from chronic foot ulcers, which are open sores that develop on the tops and bottoms of both feet. Orn is morbidly obese. He is 5' 9" tall. In recent years, his weight has fluctuated between about 300 and 320 pounds.

Orn testified that his activities are limited because he is "constantly" short of breath. He can walk for about half an hour, but requires frequent breaks. He cannot sit for longer than half an hour because he develops pain. Orn's wife sometimes helps him dress and shower because he gets "short of breath." Orn does not cook or do any housework other than "make [his bed] or something like that." Orn's daily activities have been limited since he stopped working. He spends most of his time indoors because he must avoid fumes, odors, dust, and gases. His activities include reading, watching television, and coloring in coloring books.

Orn testified that after he stopped working he had gaps in his insurance coverage that affected his ability to obtain treatment for his medical conditions. One of Orn's treating physicians, Dr. Doerning, noted in his report that Orn "has had some difficulty maintaining his insurance and has been through some lapses in his continuity of care with the pulmonary and allergy specialists" due to his lack of insurance. Orn is unable to afford the medical device used to treat sleep apnea. The record indicates that because of financial difficulties, Orn lost his house and is living with relatives.

Dr. Doerning reported significant limitations in Orn's ability to work. He found that Orn's non-exertional limitations, including fatigue, are "severe enough to interfere with attention and concentration" "constantly." In 2002, Dr. Doerning completed a Multiple Impairments Questionnaire that evaluated Orn's capacity to work. He described Orn as capable of sitting for four hours on a "sustained basis" in a "competitive five day a week work environment" and of standing and walking for zero to one hour in that environment. Medical records document numerous other observations by Dr. Doerning. For example, in a physical examination performed in 2002, Dr. Doerning stated that Orn was "still disabled secondary to his respiratory problems."

On April 11, 2002, an ALJ held a hearing to adjudicate Orn's claim for disability benefits. After the hearing, the ALJ denied benefits, concluding that Orn "retains the residual functional capacity to perform sedentary work." The ALJ discredited Orn's testimony. He also rejected Dr. Doeming's evaluations of Orn's ability to sit, stand, and walk. He relied instead on the opinion of the consulting physician for the Department of Social Services, Dr. Karamlou. Dr. Karamlou had examined Orn about a year and a half earlier, on December 20, 2000. Dr. Karamlou's report contains a notation that, in his opinion, Orn was capable of standing and walking for six hours each day in a competitive work environment.

The Appeals Council of the Social Security Administration reversed the ALJ in May of 2002. The council stated that the ALJ erred in rejecting the opinion of Orn's treating physician, in finding that Orn's "activities are inconsistent with his allegations of disabling pain," and in failing to use testimony from a vocational expert. After the decision of the Appeals Council, Orn was hospitalized for one week for respiratory problems and was discharged with instructions to receive continuous supplemental oxygen.

At the second hearing, conducted on June 17, 2003, Orn again testified to limitations imposed by his various medical conditions. He described the continuous supplemental oxygen requirement. He stated that he weighed 311 pounds, which was 13 pounds less than he had weighed six months earlier. Orn also submitted medical evidence for the time period since his previous hearing, including a Multiple Impairments Questionnaire completed in 2003 by a second treating physician, Dr. Nguyen.

Dr. Nguyen was Orn's primary physician during his 2003 hospitalization and provided his post-hospitalization follow-up care. Like Dr. Doerning, Dr. Nguyen opined that Orn's non-exertional limitations, including fatigue, would be "severe enough to interfere with attention and concentration" "constantly." Dr. Nguyen reported that Orn's asthma, chronic obstructive pulmonary disease, diabetes, and morbid obesity "severely compromise his capacity to work in a competitive manner." He assessed Orn as capable of sitting for zero to one hour per day and of standing and walking for zero to one hour per day in a five-day competitive work environment.

The ALJ again discredited Orn's testimony and the opinion of his first treating physician, Dr. Doerning. He also discredited the opinion of Orn's second treating physician, Dr. Nguyen. Despite the additional evidence presented in the second hearing, the ALJ's "conclusion regarding the claimant's physical impairments" was "unchanged." The ALJ gave a hypothetical to the vocational expert that did not include the limitations described by Orn and his two treating physicians. Based on that hypothetical, the vocational expert testified that jobs existed for the Residual Functional Capacity ("RFC") described by the ALJ. The ALJ concluded that Orn was capable of working as a surveillance system monitor, a cashier, and a ticket seller. The Appeals Council affirmed. A federal magistrate, sitting by consent, also affirmed. Orn now appeals.

II. Standard of Review

We review the district court's decision in a social security case de novo. Burch v. Barnhart, 400 F.3d 676, 679 (9th Cir.2005). The Social Security Administration's disability determination should be upheld unless it contains legal error or is not supported by substantial evidence. Stout v. Comm'r, Soc. Sec. Admin., 454 F.3d 1050, 1052 (9th Cir.2006); 42 U.S.C. §§ 405(g), 1383(c)(3). "Substantial evidence is more than a mere scintilla but less than a preponderance." Bayllss v. Barnhart, 427 F.3d 1211, 1214 n. 1 (9th Cir.2005) (internal quotation marks and citation omitted). It is "such relevant evidence as a reasonable mind might accept as adequate to support a conclusion." Burch, 400 F.3d at 679 (internal quotation marks and citation omitted). "Where evidence is susceptible to more than one rational interpretation," the ALJ's decision should be upheld. Id. "However, a reviewing court must consider the entire record as a whole and may not affirm simply by isolating a `specific quantum of supporting evidence.'" Robbins v. Soc. Sec. Admin., 466 F.3d 880, 882 (9th Cir.2006) (quoting Hammock v. Bowen, 879 F.2d 498, 501 (9th Cir.1989)). We review only the reasons provided by the ALJ in the disability determination and may not affirm the ALJ on a ground upon which he did not rely. See Connett v. Barnhart, 340 F.3d 871, 874 (9th Cir.2003).

III. Discussion

At step one of the five-step disability determination, the ALJ concluded that Orn had not engaged in substantial gainful activity since the alleged onset of his disability in 2000. See 20 C.F.R. §§ 404.1520, 416.920 (describing disability determination); Burch, 400 F.3d at 679 (same). At step two, the ALJ concluded that Orn had established that his asthma, chronic obstructive pulmonary disease, left knee problems, and obesity were all severe impairments because each interfered with his ability to perform basic work activities. See 20 C.F.R. § 404.1521; Webb v. Barnhart, 433 F.3d 683, 686 (9th Cir.2005). The ALJ acknowledged that because Orn had a severe medically determinable impairment, "all medically determinable impairments must be considered in the remaining steps of the sequential analysis." See 42 U.S.C. § 423(d)(2)(B) ("In determining whether an individual's . . . impairments are of a sufficient medical severity that such . . . impairments could be the basis of eligibility under this section, the Commissioner of Social Security shall consider the combined effect of all of the individual's impairments without regard to whether any such impairment, if considered separately, would be of such severity."); Celaya v. Halter, 332 F.3d 1177, 1181-82 (9th Cir.2003) (same). At step three, the ALJ concluded that none of Orn's impairments met or equaled a listed impairment which would require an automatic finding of disability. At step four, the ALJ concluded that Orn's impairments prevented him from performing his past work. The burden then shifted to the government to prove that Orn could perform "other work that exists in the national economy." See 20 C.F.R. § 404.1545(a)(5)(ii). Finally, at step five, the ALJ concluded that Orn's RFC allowed him to perform sedentary work. The ALJ concluded that Orn was capable of working as a surveillance system monitor, a cashier, and a ticket seller.

The ALJ concluded that Orn was not disabled within the meaning of Social Security regulations by disregarding the opinions of his two treating physicians and the testimony of Orn at his two hearings. We agree with Orn that the ALJ improperly disregarded this evidence.

A. Opinions of Treating Physicians

Orn argues that the ALJ improperly disregarded the opinions of his treating physicians, Drs. Doerning and Nguyen. Both Drs. Doerning and Nguyen offered diagnoses of Orn's medical conditions, prognoses for his conditions, and assessments of his ability to work. The record contains numerous reports and forms completed by Orn's two treating physicians from 1999 through 2003. However, the ALJ chose to rely on the opinion of a consulting physician, Dr. Karamlou, who performed one physical examination of Orn in 2000. Based on that examination, Dr. Karamlou opined that Orn could stand and walk for six hours in an eight-hour workday.

By rule, the Social Security Administration favors the opinion of a treating physician over non-treating physicians. See 20 C.F.R. § 404.1527. If a treating physician's opinion is "well-supported by medically acceptable clinical and laboratory diagnostic techniques and is not inconsistent with the other substantial evidence in [the] case record, [it will be given] controlling weight." Id. § 404.1527(d)(2). If a treating physician's opinion is not given "controlling weight" because it is not "well-supported" or because it is inconsistent with other substantial evidence in the record, the Administration considers specified factors in determining the weight it will be given. Those factors include the "[l]ength of the treatment relationship and the frequency of examination" by the treating physician; and the "nature and extent of the treatment relationship" between the patient and the treating physician. Id. § 404.1527(d)(2)(i)-(ii). Generally, the opinions of examining physicians are afforded more weight than those of non-examining physicians, and the opinions of examining non-treating physicians are afforded less weight than those of treating physicians. Id. § 404.1527(d)(1)-(2). Additional factors relevant to evaluating any medical opinion, not limited to the opinion of the treating physician, include the amount of relevant evidence that supports the opinion and the quality of the explanation provided; the consistency of the medical opinion with the record as a whole; the specialty of the physician providing the opinion; and "[o]ther factors" such as the degree of understanding a physician has of the Administration's "disability programs and their evidentiary requirements" and the degree of his or her familiarity with other information in the case record. Id. § 404.1527(d)(3)-(6).

The Administration has explained § 404.1527 in Social Security Ruling 96-2p. That ruling provides, in relevant part:

[A] finding that a treating source medical opinion is not well-supported by medically acceptable clinical and laboratory diagnostic techniques or is inconsistent with the other substantial evidence in the case record means only that the opinion is not entitled to "controlling weight," not that the opinion should be rejected. Treating source medical opinions are still entitled to deference and must be weighed using all of the factors provided in 20 C.F.R. 404.1527. . . . In many cases, a treating source's medical opinion will be entitled to the greatest weight and should be adopted, even if it does not meet the test for controlling weight.

S.S.R. 96-2p at 4 (Cum. Ed.1996), available at 61 Fed.Reg. 34,490, 34,491 (July 2, 1996).

In turn, we have explained:

The opinions of treating doctors should be given more weight than the opinions of doctors who do not treat the claimant. Lester [v. Chater, 81 F.3d 821, 830 (9th Cir.1995) (as amended).] Where the treating doctor's opinion is not contradicted by another doctor, it may be rejected only for "clear and convincing" reasons supported by substantial evidence in the record. Id. (internal quotation marks omitted). Even if the treating doctor's opinion is contradicted by another doctor, the ALJ may not reject this opinion without providing "specific and legitimate reasons" supported by substantial evidence in the record. Id. at 830, quoting Murray v. Heckler, 722 F.2d 499, 502 (9th Cir.1983). This can be done by setting out a detailed and thorough summary of the facts and conflicting clinical evidence, stating his interpretation thereof, and making findings. Magallanes [v. Bowen, 881 F.2d 747, 751 (9th Cir.1989).] The ALJ must do more than offer his conclusions. He must set forth his own interpretations and explain why they, rather than the doctors', are correct. Embrey v. Bowen, 849 F.2d 418, 421-22 (9th Cir.1988).

Reddick v. Chater, 157 F.3d 715, 725 (9th Cir.1998); accord Thomas, 278 F.3d at 957; Lester, 81 F.3d at 830-31.

When an examining physician relies on the same clinical findings as a treating physician, but differs only in his or her conclusions, the conclusions of the examining physician are not "substantial evidence." As we explained in Murray, "In this case, . . . the findings of the non-treating physician were the same as those of the treating physician. It was his conclusions that differed. . . . If the ALJ wishes to disregard the opinion of the treating physician, he or she must make findings setting forth specific, legitimate reasons for doing so that are based on substantial evidence in the record." 722 F.2d at 501-02 (emphases in original). By contrast, when an examining physician provides "independent clinical findings that differ from the findings of the treating physician," such findings are "substantial evidence." Miller v. Heckler, 770 F.2d 845, 849 (9th Cir.1985); accord Andrews v. Shalala, 53 F.3d 1035, 1041 (9th Cir.1995); Magallanes, 881 F.2d at 751; Allen v. Heckler, 749 F.2d 577, 579 (9th Cir.1985) (as amended). Independent clinical findings can be either (1) diagnoses that differ from those offered by another physician and that are supported by substantial evidence, see Allen, 749 F.2d at 579, or (2) findings based on objective medical tests that the treating physician has not herself considered, see Andrews, 53 F.3d at 1041.

If there is "substantial evidence" in the record contradicting the opinion of the treating physician, the opinion of the treating physician is no longer entitled to "controlling weight." 20 C.F.R. § 404.1527(d)(2). In that event, the ALJ is instructed by § 404.1527(d)(2) to consider the factors listed in § 404.1527(d)(2)-(6) in determining what weight to accord the opinion of the treating physician. Even when contradicted by an opinion of an examining physician that constitutes substantial evidence, the treating physician's opinion is "still entitled to deference." S.S.R. 96-2p at 4, 61 Fed.Reg. at 34,491. "In many cases, a treating source's medical opinion will be entitled to the greatest weight and should be adopted, even if it does not meet the test for controlling weight." Id. As we stated in Reddick,"Even if the treating doctor's opinion is contradicted by another doctor, the ALJ may not reject this opinion without providing `specific and legitimate reasons' supported by substantial evidence in the record." 157 F.3d at 725 (quoting Murray, 722 F.2d at 502).

1. Treating Physicians' Opinions are Entitled to Weight

The Commissioner argues that Dr. Karamlou's opinion constitutes per se substantial evidence to support the ALJ's disregard of the opinions of Drs. Doerning and Nguyen. We disagree. Dr. Karamlou's opinion that Orn could stand and walk for six hours did not rely on "independent findings." In addition, the criteria established by § 404.1527 indicate that the opinions of Drs. Doerning and Nguyen are entitled to weight in Orn's disability determination.

Dr. Karamlou, like Drs. Doerning and Nguyen, performed a physical examination of Orn. Dr. Karamlou agreed with the diagnoses provided by Orn's treating physicians and offered no alternative diagnosis. Dr. Karamlou's opinion did not rest on results from objective clinical tests that Drs. Doerning and Nguyen had not considered. Dr. Karamlou's findings "were the same as those of the treating physician[s]. It was his conclusions that differed." Murray, 722 F.2d at 501 (emphasis in original). Therefore, his conclusion concerning Orn's ability to stand or walk based on that examination was not an "independent finding," and his opinion does not alone constitute substantial evidence to support the rejection of Orn's treating physicians' opinions. See Reddick, 157 F.3d at 725; see also Robbins, 466 F.3d at 882 (prohibiting affirmance in Social Security disability cases "simply by isolating a `specific quantum of supporting evidence'" (citation omitted)).

A second, independent reason precludes the ALJ from disregarding the opinions of the treating physicians in this case. Even if Dr. Karamlou's opinion were "substantial evidence," § 404.1527 still requires deference to the treating physicians' opinions. 20 C.F.R. § 404.1527; see S.S.R. 96-2p at 1 ("A finding that a treating source's medical opinion is not entitled to controlling weight does not mean that the opinion is rejected."), 61 Fed.Reg. at 34,490; see also Green-Younger v. Barnhart, 335 F.3d 99, 106 (2d Cir.2003); McGoffin v. Barnhart, 288 F.3d 1248, 1252 (10th Cir.2002). As discussed above, § 404.1527 lists several factors that increase the weight afforded to Orn's treating physicians' opinions in this case.

For example, the treating relationship of both physicians provides a "unique perspective" on Orn's condition. See 20 C.F.R. § 404.1527(d)(2). In addition, the nature and extent of the physicians' relationships with Orn adds significant weight to their opinions. Id. § 404.1527(d)(2)(i)-(ii). Dr. Doerning was Orn's treating physician between 1999 through 2002, and his reports cover that entire period. Dr. Nguyen was Orn's primary physician during his 2003 hospitalization and provided his post-hospitalization follow-up treatment. His questionnaire was the most recent report in the record. It was also the only report describing the effects of two significant medical events: Orn's latest hospitalization and his reliance on continuous supplemental oxygen.

The "[s]upportability" of Orn's treating physicians' opinions adds further weight. See id. § 404.1527(d)(3). The primary function of medical records is to promote communication and recordkeeping for health care personnel — not to provide evidence for disability determinations. We therefore do not require that a medical condition be mentioned in every report to conclude that a physician's opinion is supported by the record. When viewed in its entirety, the record provides ample support for the opinions of Drs. Doerning and Nguyen. The record contains numerous reports from Orn's health care providers, as well as results from medical tests and laboratory findings, that support the questionnaires completed by Drs. Doerning and Nguyen.

Finally, the consistency of Orn's treating physicians' reports merits additional weight. See id. § 404.1527(d)(4). Consistency does not require similarity in findings over time despite a claimant's evolving medical status. Rather, as required by the applicable regulation, the opinions of Drs. Doerning and Nguyen were consistent "with the record as a whole." See id. The physicians offered opinions that were substantiated by the contemporaneous medical tests and Orn's medical condition. The gradual decrease in Orn's physical capacity, as illustrated by the evaluations of his treating physicians, is supported by the record. As Dr. Doerning stated, Orn's condition was "progressively worsening."

2. No "Specific, Legitimate Reasons" Supported by Substantial Evidence

The two reasons provided by the ALJ for rejecting the opinions of Orn's treating physicians are insufficient. The reasons are not "specific, legitimate reasons" that are supported by "substantial evidence." See Thomas, 278 F.3d at 957. In fact, the record contradicts them.

The ALJ's first reason for rejecting both treating physicians' opinions of Orn's functional capacity was that those physicians had "fail[ed] to indicate what claimant could do despite his limitations." The record shows the opposite. Dr. Doerning's multiple impairments questionnaire, completed on April 5, 2002, indicates that Orn could sit for four hours a day and could stand and walk for one hour per day. It also reports that Orn had no limitations in reaching, handling, fingering, or lifting, and that he could occasionally lift and carry up to ten pounds. Similarly, Dr. Nguyen's questionnaire dated one year later reported no limitations in Orn's ability in fine manipulations with fingers and hands or in reaching. It described Orn as able to sit or stand for one hour in a competitive work environment. See Smolen v. Chater, 80 F.3d 1273, 1288 (9th Cir.1996) (stating that ALJ has a duty to "conduct an appropriate inquiry" if the ALJ determines that it is necessary to know the basis of the treating physician's opinion).

The ALJ's second reason for rejecting the treating physicians' opinions was lack of objective support. After describing Dr. Doerning's assessment of Orn's limitations, the ALJ stated that "there is no objective evidence of decreased range of motion or neurological deficits to support such severe limitations in standing and walking." The ALJ's statement is correct as far as it goes, but it does not warrant the rejection of Dr. Doerning's opinion. Dr. Doerning never claimed that "decreased range of motion" or "neurological deficits" caused Orn's limitations in standing and walking. Rather, Dr. Doerning indicated that Orn's ability to work was limited by his respiratory conditions.

Dr. Doerning diagnosed Orn with "asthma and severe obstructive pulmonary disease" and reported that Orn's "symptoms and functional limitations [were] reasonably consistent with" the impairments described in his evaluation. The record provides voluminous support for Dr. Doerning's opinion that Orn's respiratory diseases adversely affected his ability to work. Orn stopped working after suffering an asthma attack. His medical records contain physical examinations which reveal wheezing and other respiratory abnormalities. He has abnormal pulmonary functional tests. Orn takes medications for his respiratory diseases and, since 2003, requires continuous supplemental oxygen. He has been hospitalized twice in four years for respiratory problems. The ALJ's reason for rejecting Dr. Doerning's opinion — that the record did not contain evidence of "decreased range of motion" or "neurological deficits" — is not "legitimate" because it is not responsive to Dr. Doerning's opinion based on Orn's respiratory problems. Compare Magallanes, 881 F.2d at 751-52 (upholding rejection of treating physician's opinion that claimant was "disabled" when it was contradicted by the opinions of four other physicians, EMG studies, and other medical tests).

Similarly, the ALJ stated that there was "no clinical evidence" to support the sitting and standing limitations assessed by Dr. Nguyen. The ALJ then described a host of other conditions for which Orn had not been diagnosed, including "disc herniation, stenosis or nerve root compression." Missing from the ALJ's list is any mention of Orn's respiratory disorders, obesity, and diabetes, which Dr. Nguyen cited as causing Orn's limitations and which are amply supported in the record.

An ALJ may not exclude a physician's testimony for a lack of objective evidence of impairments not referenced by the physician. Rather, an ALJ must evaluate the physician's assessment using the grounds on which it is based. The ALJ thus erred in rejecting the opinions of Drs. Doerning and Nguyen. See Andrews, 53 F.3d at 1043.

B. Orn's Testimony

Orn also argues that the ALJ erred in discrediting his testimony. Orn testified that his impairments included fatigue, shortness of breath, difficulty with concentration, and the frequent need for medical treatment. The ALJ excluded those impairments in determining Orn's RFC. The ALJ concluded that Orn's testimony describing his "fatigue, shortness of breath and dysfunction" was "not totally credible."

An ALJ is not "required to believe every allegation of disabling pain" or other non-exertional impairment. See Fair v. Bowen, 885 F.2d 597, 603 (9th Cir.1989). However, to discredit a claimant's testimony when a medical impairment has been established, the ALJ must provide "`specific, cogent reasons for the disbelief.'" Morgan, 169 F.3d at 599 (quoting Lester, 81 F.3d at 834). The ALJ must "cit[e] the reasons why the [claimant's] testimony is unpersuasive." Id. Where, as here, the ALJ did not find "affirmative evidence" that the claimant was a malingerer, those "reasons for rejecting the claimant's testimony must be clear and convincing." Id.

Social Security Administration rulings specify the proper bases for rejection of a claimant's testimony. See S.S.R. 02-1p (Cum. Ed.2002), available at Policy Interpretation Ruling Titles II and XVI: Evaluation of Obesity, 67 Fed. Reg. 57,859-02 (Sept. 12, 2002); S.S.R. 96-7p (Cum. Ed.1996), available at 61 Fed.Reg. 34,483-01 (July 2, 1996). An ALJ's decision to reject a claimant's testimony cannot be supported by reasons that do not comport with the agency's rules. See 67 Fed.Reg. at 57860 ("Although Social Security Rulings do not have the same force and effect as the statute or regulations, they are binding on all components of the Social Security Administration, . . . and are to be relied upon as precedents in adjudicating cases."); see Daniels v. Apfel, 154 F.3d 1129, 1131 (10th Cir.1998) (concluding that ALJ's decision at step three of the disability determination was contrary to agency regulations and rulings and therefore warranted remand). Factors that an ALJ may consider in weighing a claimant's credibility include reputation for truthfulness, inconsistencies in testimony or between testimony and conduct, daily activities, and "unexplained, or inadequately explained, failure to seek treatment or follow a prescribed course of treatment." Fair, 885 F.2d at 603; see also Thomas, 278 F.3d at 958-59.

The ALJ gave four reasons for rejecting Orn's testimony. Each of these reasons is improper under Social Security Rulings, our case law, or both.

1. Failure to Lose Weight and Credibility

First, the ALJ rejected Orn's testimony because he had failed to lose weight. The ALJ wrote:

[A]lthough the claimant testified his treating physician has not advised him to diet, the medical record indicates he was given an 1800 calorie diabetic diet to follow. (Exhibit 8F, page 14). However, the claimant's weight remains in excess of 300 pounds. It may be that some of claimant's restrictions and symptoms are aggravated by excess weight. There is no indication that the claimant has maintained a weight loss program to reduce his weight and promote better health, which detracts from his credibility. And is non compliance with his doctor's instructions.

In using Orn's failure to lose weight as a basis for denying benefits, and in concluding that this failure "detracts from his credibility," the ALJ both ignored a Social Security Ruling and misapplied our case law.

a. Failure to Lose Weight

In September 2002, the Secretary issued a new ruling specifically addressing obesity. In relevant part, the ruling provides:

Treatment for obesity is often unsuccessful. Even if treatment results in weight loss at first, weight is often regained, despite the efforts of the individual to maintain the loss.

* * *

Obesity is a disease that requires treatment, although in most people the effect of treatment is limited. . . .

A common misconception is that the goal of treatment is to reduce weight to a "normal" level. Actually, the goal of realistic medical treatment for obesity is only to reduce weight by a reasonable amount that will improve health and quality of life. People with extreme obesity, even with treatment, will generally continue to have obesity. Despite short-term progress, most treatments for obesity do not have a high success rate.

* * *

Before failure to follow prescribed treatment for obesity can become an issue in a case, we must first find that the individual is disabled because of obesity or a combination of obesity and another impairment(s). Our regulations at 20 CFR 404.1530 and 416.930 provide that, in order to get benefits, an individual must follow treatment prescribed by his or her physician if the treatment can restore the ability to work, unless the individual has an acceptable reason for failing to follow the prescribed treatment. We will rarely use "failure to follow prescribed treatment" for obesity to deny or cease benefits.

* * *

When a treating source has prescribed treatment for obesity, the treatment must clearly be expected to improve the impairment to the extent that the person will not be disabled. . . . [T]he goals of treatment for obesity are generally modest, and treatment is often ineffective. Therefore, we will not find failure to follow prescribed treatment unless there is clear evidence that treatment would be successful.

S.S.R. 02-1p at 2, 8-9, 67 Fed.Reg. at 57,861-64.

At the time in question Orn had not been found disabled, so Social Security Ruling 02-1p precludes the ALJ from considering the effect of any failure to follow treatment for obesity. Id. at 9. In addition, there is no evidence either that Orn was directed to lose weight as part of a prescribed treatment, or "clear evidence that treatment would be successful." Id. The only evidence of a diet as prescribed treatment was a single sheet of paper from Orn's hospital records entitled "Patient Discharge Instructions," cited by the ALJ as Exhibit 8F. There are five sections to the sheet: "Doctor," "Diet," "Activity," "Medications," and "Special Instructions." Under "Doctor," Dr. Nguyen is named. Under "Diet," "1800 ADA" is written, and the "Yes" blank following "Diet Copy to Patient" is checked. Under "Activity," the "As tolerated" blank is checked. Under "Medications," "continue all medications prior to admissions" is written, followed by four separate medications and their dosages. Under "Special Instructions," "Oxygen Nasal Cannula at 2 liters/min" is written. The sheet is signed by Registered Nurse "M. Williams," and by Orn. The entries on the sheet appear to be in the hand-writing of Williams.

This single sheet with the notation "1800 ADA" can hardly be described as a prescribed treatment. "Prescribed treatment" is a term of art. According to the Social Security Administration's 2002 ruling on obesity, "The treatment must be prescribed by a treating source, . . . not simply recommended. A treating source's statement that an individual `should' lose weight or has `been advised' to get more exercise is not prescribed treatment." Id. at 9. Orn's own testimony exemplifies the distinction. At his second hearing he testified that he had not been "prescribed" a diet, saying instead that he remembered the diet being described to him in the hospital.

Finally, even if the patient discharge instruction sheet were a prescribed treatment, a finding of a "failure to follow prescribed treatment" would be inappropriate. There is nothing in the record suggesting that there was any chance of such a prescription succeeding in eliminating or ameliorating Orn's obesity, let alone "clear evidence" that the treatment would be successful. See id.

b. Credibility

The ALJ concluded that Orn's failure to follow the 1800 calorie-per-day diet indicated on the "Patient Discharge Instructions" "detract[ed] from his credibility." In so concluding, the ALJ misapplied our case law. In some circumstances, a failure to seek treatment or a failure to follow a prescribed treatment is properly used as evidence supporting a conclusion that the claimant is not credible in describing his or her symptoms. The most common such circumstance is a claimant's complaint about disabling pain that cannot be objectively ascertained.

Our case law is clear that if a claimant complains about disabling pain but fails to seek treatment, or fails to follow prescribed treatment, for the pain, an ALJ may use such failure as a basis for finding the complaint unjustified or exaggerated. See, e.g., Fair, 885 F.2d at 603. In the case of a complaint of pain, such failure may be probative of credibility, because a person's normal reaction is to seek relief from pain, and because modern medicine is often successful in providing some relief. But in the case of impairments where the stimulus to seek relief is less pronounced, and where medical treatment is very unlikely to be successful, the approach to credibility makes little sense. This second case is probably best exemplified by a claimant whose obesity adversely affects his or her health and activities. See S.S.R. 02-1p at 9 (defining "prescribed treatment" narrowly and stating that failure to follow treatment for obesity will "rarely" affect disability determinations). Thus, the failure to follow treatment for obesity tells us little or nothing about a claimant's credibility. In the case before us, there is no reason to conclude from Orn's failure to adhere to an 1800 calorie-per-day diet that he is not telling the truth about his medical problems that are exacerbated by his obesity.

2. "Gaps" in Medical Treatment

Second, the ALJ stated that "large gaps in the claimant's medical treatment . . . suggest that [Orn's] symptoms were not especially troublesome." Orn's failure to receive medical treatment during the period that he had no medical insurance cannot support an adverse credibility finding. We have held that an "unexplained, or inadequately explained, failure to seek treatment" may be the basis for an adverse credibility finding unless one of a "number of good reasons for not doing so" applies. Fair, 885 F.2d at 603. But, "[d]isability benefits may not be denied because of the claimant's failure to obtain treatment he cannot obtain for lack of funds." Gamble v. Chater, 68 F.3d 319, 321 (9th Cir.1995).

Here, Orn explained in his testimony that he would have liked to have seen his doctors more often, but was unable to do so because he "can't afford it." The ALJ did not suggest that Orn's proffered reason was "not believable." Id.; see also S.S.R. 96-7p at 7-8 (stating that an "adjudicator must not draw any inferences about an individual's symptoms and their functional effects from a failure to seek or pursue regular medical treatment without first considering any explanations that the individual may provide, or other information in the case record, that may explain infrequent or irregular medical visits or failure to seek medical treatment" including inability to pay, whether "[t]he individual's daily activities may be structured so as to minimize symptoms to a tolerable level or eliminate them entirely," and whether medication may relieve symptoms).

Flaten v. Secretary of Health Human Services, 44 F.3d 1453 (9th Cir.1995), cited by the Commissioner, is inapposite in these circumstances. In that case we upheld an adverse credibility determination for failure to seek treatment despite Flaten's alleged inability to pay. Id. at 1464. We determined that rejection of Flaten's back pain testimony was appropriate because during the time Flaten alleged she was unable to afford treatment she had received other medical care and had failed to mention her back pain. Id. By contrast, Orn did not see physicians during the time he could not afford medical treatment and, unlike Flaten, Orn sought treatment for his symptoms when he was able to go to the doctor.

3. Daily Activities

Third, the ALJ rejected Orn's testimony because his activities of "read[ing], watch[ing] television and color[ing] in coloring books" "indicate that he is more functional than alleged." "This court has repeatedly asserted that the mere fact that a plaintiff has carried on certain daily activities . . . does not in any way detract from her credibility as to her overall disability." Vertigan v. Halter, 260 F.3d 1044, 1050 (9th Cir.2001). Neither of the two grounds for using daily activities to form the basis of an adverse credibility determination are present in Orn's case. First, as he described them, Orn's activities do not contradict his other testimony. See Fair, 885 F.2d at 603. Second, Orn's activities do not meet the threshold for transferable work skills, the second ground for using daily activities in credibility determinations. Id.

In Fair, we wrote that daily activities may be grounds for an adverse credibility finding "if a claimant is able to spend a substantial part of his day engaged in pursuits involving the performance of physical functions that are transferable to a work setting." 885 F.2d at 603 (emphasis omitted) (emphasis omitted); see also Burch, 400 F.3d at 681 (stating that adverse credibility finding based on activities may be proper "if a claimant engages in numerous daily activities involving skills that could be transferred to the workplace"). Here, there is neither evidence to support that Orn's activities were "transferable" to a work setting nor proof that Orn spent a "substantial" part of his day engaged in transferable skills. See Fair, 885 F.2d at 603. The ALJ must make "specific findings relating to [the daily] activities" and their transferability to conclude that a claimant's daily activities warrant an adverse credibility determination. Burch, 400 F.3d at 681.

We agree with Orn that reading, watching television, and coloring in coloring books are activities that are so undemanding that they cannot be said to bear a meaningful relationship to the activities of the workplace. See Fair, 885 F.2d at 603 ("The Social Security Act does not require that claimants be utterly incapacitated to be eligible for benefits, and many home activities are not easily transferable to what may be the more grueling environment of the workplace, where it might be impossible to periodically rest or take medication." (citations omitted)). In oral argument before this court, the Commissioner confirmed that a surveillance system monitor is a security position that requires sustained concentration and attention, as well as the ability to act immediately in emergencies. The ALJ's conclusion that because Orn sometimes reads, watches television, and colors in coloring books he has "transferable" skills to be a surveillance system monitor is, to put it charitably, not supported by substantial evidence.

4. ALJ's Personal Observations

Finally, the ALJ rejected Orn's testimony, including his claims of fatigue and difficulty concentrating, because Orn "was able to testify in a responsive manner without any noticeable problems with memory or thought content." The ALJ's observations of a claimant's functioning may not form the sole basis for discrediting a person's testimony. See S.S.R. 96-7p at 8 ("[T]he adjudicator is not free to accept or reject the individual's complaints solely on the basis of . . . personal observations."), available at 61 Fed.Reg. at 34,488. Instead, an ALJ's personal observations may be used only in "the overall evaluation of the credibility of the individual's statements." Id. Because the ALJ's other reasons for rejecting Orn's testimony fail, the ALJ's personal observations standing alone cannot support the adverse credibility finding.

C. Remand for Calculation of Benefits

When an "ALJ's reasons for rejecting the claimant's testimony are legally insufficient and it is clear from the record that the ALJ would be required to determine the claimant disabled if he had credited the claimant's testimony," we remand for a calculation of benefits. Connett, 340 F.3d at 876; accord McCartey, 298 F.3d at 1076-77. Such is the case here. The Commissioner conceded at oral argument that if the opinions of Orn's treating physicians and Orn's testimony are credited, Orn has established that he is disabled. We reach this same conclusion, even without relying on the Commissioner's concession. Because a remand for an award of benefits is appropriate, we do not reach Orn's alternative argument that remand is warranted under Celaya, 332 F.3d at 1184 (remanding where ALJ failed to consider claimant's obesity in steps four and five of the disability determination).

Conclusion

The ALJ's reasons for disregarding the opinions of Orn's two treating physicians and for disregarding Orn's testimony are legally insufficient. Once these opinions and this testimony are credited, we are obliged to conclude that Orn is disabled. We therefore reverse the decision of the district court and remand with instructions to remand for calculation of benefits.

REVERSED and REMANDED.