We are called upon to decide whether an alcoholic who is able to remain sober for considerable periods of time is entitled to disability benefits.

I

Tylitzki filed applications for Disability Insurance Benefits and Supplemental Security Income (“benefits”) based on alcoholism, nerves, mental problems, and depression. Benefits were denied both initially and, following a hearing, by the administrative law judge (“ALJ”). The decision of the ALJ became the final decision of the Secretary of Health and Human Services (“Secretary”) when the Appeals Council denied review. The district court affirmed the decision. This timely appeal followed. We have jurisdiction under 28 U.S.C. § 1291 .

II

We review the district court’s order upholding the Secretary’s denial of benefits de novo. Hammock v. Bowen, 879 F.2d 498, 501 (9th Cir.1989). We must affirm the Secretary’s decision if it is supported by substantial evidence and is based on proper legal standards. Substantial evidence is such relevant evidence as a reasonable mind might accept as adequate to support a conclusion. Clem v. Sullivan, 894 F.2d 328, 330 (9th Cir.1990). In determining whether substantial evidence supports a decision, we consider the record as a whole. Richardson v. Perales, 402 U.S. 389, 401 , 91 S.Ct. 1420, 1427 , 28 L.Ed.2d 842 (1971).

III

Tylitzki argues that the ALJ erred in (a) failing properly to consider certain medical evidence establishing disability, (b) finding that Tylitzki’s alcoholism was controllable, and (c) finding that Tylitzki could perform his past relevant work as a truck driver.

A

Tylitzki claims that the ALJ improperly disregarded evidence in finding that Tylitz-ki’s alcoholism was in remission rather than in a chronic state. On the contrary, our examination of the record indicates that the Secretary clearly considered Tylitzki’s alcoholism to be an ongoing condition. Furthermore, the ALJ used the terminology of examining physicians Dr. Jimenez and Dr. Barron. We cannot say that the ALJ’s characterization of Tylitzki’s alcoholism as “in remission” was either error or harmful.

Tylitzki claims that the ALJ erred by not finding non-exertional limitations. However, the ALJ clearly considered evidence of such limitations, for example, in his examination of vocational expert Dr. Jones. Furthermore, contrary to Tylitzki’s claim, the ALJ made specific findings relating to such limitations.

B

Tylitzki also contends that the ALJ erred in finding that his alcoholism was controllable.

Alcoholism is not per se disabling, but can constitute a disability if it is uncontrollable. Clem, 894 F.2d at 331 . Once a claimant establishes, as did Tylitzki, that he or she is a diagnosed alcoholic, the Secretary must adequately inquire into the claimant’s ability to control his or her drinking. The Secretary (or the ALJ whose decision the Secretary adopts) must then make specific findings regarding that ability and its disabling effect. Cooper v. Bowen, 815 F.2d 557, 560 (9th Cir.1987).

The ALJ inquired into Tylitzki’s ability to control his drinking. The ALJ specifically found that Tylitzki’s alcoholism was controllable to the extent that Tylitzki could perform his past relevant work as a truck driver. Consequently, the ALJ concluded that Tylitz-ki’s alcoholism was not a disability as defined by the Social Security Act. See Clem, 894 F.2d at 331 .

*1414 The ALJ’s conclusion that Tylitzki’s alcoholism was controllable was based upon two findings: (1) that Tylitzki had been an alcoholic for many years but had, nevertheless, been able to keep working as a truck driver, and (2) that Tylitzki was able to abstain from alcohol for long periods.

The ALJ’s analysis is somewhat troubling. The fact that Tylitzki was able to hold a job for nearly twenty-seven years prior to 1980 does not demonstrate that Tylitzki’s alcoholism was controllable in 1985, when he claims he became disabled. The Secretary cites Gregory v. Bowen, 844 F.2d 664, 666-67 (9th Cir.1988) and Ray v. Bowen, 813 F.2d 914, 917 (9th Cir.1987) for the proposition that long-standing impairments that have not previously precluded work are presumed not disabling unless there is evidence of deterioration. Those cases do not support such a presumption. In both Gregory and Ray , long-standing conditions that were not previously disabling were found not presently disabling specifically because the conditions had not changed. In the present case there is substantial evidence indicating that Tylitzki’s condition had deteriorated from the days when he was able to hold the same job for nearly twenty-seven years.

Tylitzki accepted retirement from his long-term employer in 1979. The record documents a history of about twenty hospital admissions dating from July 1978 through January 1988. There is no record of, and neither party alleges, any hospitalizations during most of Tylitzki’s productive pre-re-tirement period. The record contains additional evidence of a deteriorating condition dating from 1978. Tylitzki was fired, by the same employer, six times during 1980-83 due to alcohol abuse. Tylitzki undertook multiple rehabilitation programs during 1978-88. At the time of the March 7, 1989 hearing before the ALJ, Tylitzki had not worked for four years and had been living in a halfway house for three years.

The finding that Tylitzki had been able to keep working as a truck driver, therefore, is not supported by substantial evidence. On the other hand, there is substantial evidence in the record to support the ALJ’s alternative finding that Tylitzki was able to abstain from drinking for long periods of time.

The ALJ found that Tylitzki remained sober approximately eleven months per year. This finding was primarily based on Tylitz-ki’s own testimony. In general, such testimony from a chronic alcoholic is suspect as rationalizing his own abuse and would be insufficient to support a denial of benefits. Brown v. Heckler, 713 F.2d 441, 442-43 (9th Cir.1983). The medical evaluations in the record indicate, as well, that Tylitzki’s testimony is dubious. Indeed, the ALJ specifically found Tylitzki “to be credible only to the extent that his subjective allegations are confirmed by objective medical findings.”

Nevertheless, we cannot find error in the ALJ’s decision to accept Tylitzki’s testimony on this point because it is indeed corroborated by the evidence in the record. Tylitzki was hospitalized for alcohol abuse, on average, far more often than one month out of each year since the onset of such hospitalizations in 1978. However, Tylitzki apparently was hospitalized only about once a year from the claimed onset of his disability on June 7, 1985 through the March 7, 1989 hearing before the ALJ. Therefore, there is substantial evidence in the record to support the ALJ’s finding that, “especially living in the environment where he is now,” Tylitzki could control his drinking for considerable periods of time. 1 Cf. Cooper, 815 F.2d at 561 (“No evidence was presented-that Cooper could control his alcohol consumption. There was no evidence, for example, that Cooper had been able to stop his drinking in the past.”); see also Arroyo v. HHS, 932 F.2d 82, 87 (1st Cir.1991) (“We emphasize that even though *1415 alcoholism, by definition, imports a certain lack of control, evidence that a claimant has been diagnosed a chronic alcoholic is not sufficient to establish that the claimant has lost the ability to control his consumption of alcohol”).

C

Tylitzki also argues that the ALJ erred in finding that he could perform his past relevant work as a truck driver. Tylitzki claims that, at best, he could “piece together a series of jobs, interrupted for weeks at a time by his alcohol drinking,” and that this would not constitute substantial gainful activity.

We must not lose sight of the fact that the ultimate burden remains on Tylitzki to show that he cannot perform his previous occupation. Clem, 894 F.2d at 330 . Moreover, to meet his burden of showing that he is disabled, he must show that his impairment — his alcoholism — precluded his past relevant work for any period of time which has lasted or could reasonably be expected to last for twelve continuous months. Id. at 331 . Even uncontrolled drinking does not necessarily constitute a disability. Id. “An uncontrollable addiction to alcohol may not prevent the person so addicted from undertaking gainful activity.” Cooper, 815 F.2d at 561 (Sneed, J., dissenting); Clem, 894 F.2d at 332 ; see also Arroyo, 932 F.2d at 87-88 .

In this case, in concluding that Tyl-itzki was able to perform his past relevant work, the ALJ relied on the testimony of Dr. Galpher, a medical advisor, and Dr. Jones, a vocational expert. Dr. Galpher testified regarding Tylitzki’s residual functional capacity for such work. Dr. Jones testified regarding Tylitzki’s general ability to perform such work.

Dr. Jones’ testimony was based on a hypothetical posed by the ALJ. Dr. Jones was to assume that certain of Dr. Galpher’s conclusions relating to Tylitzki’s functional capacity were true. Dr. Jones was further to assume that Tylitzki could abstain from alcohol for eleven months out of each year. Dr. Jones testified that, under those conditions, Tylitzki could work for eleven months, would thereafter probably be fired during his drunken period, and then could find another job of eleven months’ duration. Dr. Jones also testified that acquiring and losing a series of jobs over time due to episodic alcohol abuse was typical of chronic alcoholics.

We conclude that such a series of jobs would constitute substantial gainful activity when they are of significant duration as under the terms of the ALJ’s hypothetical. See Arroyo, 932 F.2d at 88 . We also conclude that the ALJ’s hypothetical was reasonable in view of the frequency of hospitalizations since 1985. 2 We further conclude, as we must, that there was substantial evidence supporting the ALJ’s finding that Tylitzki could perform his past relevant work during the relevant period of time.

IV

While it may be chilling to conceive of any alcoholic behind the wheel “of one of the big ones,” Dissent, at 7831, that is not the question before us. In fact, nothing we decide today will affect whether Tylitzki gets behind the wheel of a truck, or any other vehicle. The only question before us is whether substantial evidence supported the ALJ’s conclusion that Tylitzki could find substantial gain *1416 ful employment in his past relevant work during the period in question. Based on the record before us, we must answer that question in the affirmative. Substantial evidence supported the ALJ’s conclusion that Tylitzki remained sober for considerable periods of time. Substantial evidence, including the expert testimony of Dr. Galpher and Dr. Jones, supported the ALJ’s conclusion that Tylitzki could perform a series of jobs in his relevant line of work during much of the relevant period. Therefore, the ALJ did not err in denying Tylitzki’s application for disability benefits.

AFFIRMED.

The per curiam opinion represents only the views of the majority, Judges O’Scannlain and Kleinfeld.

. Contrary to the dissent's assertion, the ALJ did not disregard the medical assessments made by Dr. Jimenez and Dr. Barron. Indeed, their findings support the ALJ's determination that Tylitzki remained sober for considerable periods of time. For example, Dr. Jimenez described Tylitzki's alcohol abuse as "currently in remission” and his "disability” as "temporaiy.” Similarly, Dr. Barron diagnosed Tylitzki's alcohol abuse as "in remission,” "intermittent,” and “sporadic.” Those findings are consistent with the medical records, Tylitzki's own testimony, and the statement in his brief that he had months of sobriety.

. The dissent argues that the ALJ's hypothetical presented a false picture of Tylitzki's impairment by suggesting that he "would stay sober for eleven months, be drunk one month, and then, like clockwork, be sober again for eleven months.” Dissent, at 1416. We respectfully suggest that the hypothetical accurately reflects the facts of Tylitzki’s situation. See Gamer v. HHS, 815 F.2d 1275, 1279-80 (9th Cir.1987). He concedes in his brief that his "alcoholism [is] episodic in nature,” and the medical records demonstrate that his binges occurred, on average, one month out of every twelve. Moreover, there is no indication that the vocational expert took the unrealistic view of the nature of Tylitzki’s alcoholism suggested by the dissent. Dr. Jones' testimony reveals the understanding that