MEMORANDUM *
Leticia De Guzman appeals the district court’s decision upholding the Social Security Administration (“SSA”) Commissioner’s denial of her application for disability insurance and supplemental social security income benefits. Because the parties are familiar with the facts of this case, we recount them here only as necessary.
De Guzman contends that the administrative law judge (“ALJ”) erred in making an adverse credibility determination and rejecting the opinions of three of her doctors. We review the district court’s order affirming the Commissioner’s denial of benefits de novo to determine whether such denial “was supported by substantial evidence and a correct application of the law.” Roberts v. Shalala, 66 F.3d 179, 182 (9th Cir.1995). “Where evidence is susceptible to more than one rational interpretation, it is the ALJ’s conclusion that must be upheld.” Burch v. Barnhart, 400 F.3d 676, 679 (9th Cir.2005).
I
De Guzman challenges the ALJ’s conclusion that she was not entirely credible because of inconsistencies between her reported activities and her alleged limitations, as well as a lack of objective medical evidence to corroborate her symptoms.
A
De Guzman reported wildly varying activity levels over the course of several months that were inconsistent with her claims of chronic, debilitating pain. On January 27, 2003, Dr. Luis Wainstein performed a disability evaluation. She told him that she had experienced “abdominal pain ever since 1995 when she underwent abdominal surgery,” as well as daily headaches that “started shortly after her abdominal surgery.” Nevertheless, she could still work as a hotel housekeeper, “walk 4-5 blocks going to downtown Seattle ... without discomfort,” “clean the bathroom and use[ ] the vacuum cleaner in the apartment,” and wash dishes. When he asked her how she was able to work despite her pain, she told him that “lately this has been getting worse.” However, Dr. Wainstein could not determine a medical explanation for the “aggravation of her symptoms.” Shortly after this visit, De *204Guzman had a chest x-ray and CT scan with unremarkable findings.
After her first disability claim was denied, her symptoms suddenly took a turn for the worse. She filed a request for reconsideration on March 26, 2003, asserting that she was “always very sick” and could not work “due to the pain I always feel in my head, stomach, and chest.” She claimed that she could not “do any house chores because of pain and fatigue” and was “homebound,” “staying] in bed all day because of pain” in her stomach, head, and chest. Three months later, in June 2003, she acknowledged in her daily activities questionnaire that since the onset of her alleged disability, she did light housework such as a “little dusting and sweeping” and used public transportation to go out with her husband.1
A claimant’s “prior statements inconsistent with [her] claim of pain ... may be properly taken into account in determining whether or not [her] claim of disabling pain should be believed.” Fair v. Bowen, 885 F.2d 597, 604 n. 5 (9th Cir.1989). It is certainly true that “[s]ymptoms may vary in their intensity, persistence, and functional effects, or may worsen or improve with time, and this may explain why the individual does not always allege the same intensity, persistence, or functional effects of his or her symptoms.” Soc. Sec. Rul. 96-7p. Hence, the ALJ may look to the case record to find “explanations for any variations in the individual’s statements about symptoms and their effects.” Id. Here there was no objective medical evidence to account for De Guzman’s fluctuating activity levels over this six-month period. While the lack of such evidence is “only one factor that the adjudicator must consider in assessing an individual’s credibility,” id., the record as a whole supports the ALJ’s conclusion that she was not entirely credible.
When she first applied for disability benefits in 2002,2 she alleged post-operative stomach pain and claimed that she “became unable to work because of [her] disabling condition on December 1, 2002.” However, she admittedly continued working as a hotel housekeeper until January 29, 2003. According to the ALJ, that De Guzman continued to work despite her “alleged continued difficulties” was “[m]ost damaging to [her] credibility.” The ALJ properly relied upon discrepancies in her 2002 application in making his adverse credibility determination. See Soc. Sec. Rul. 96-7p (noting that in assessing credibility, the ALJ “must also look at statements the individual made to SSA at each prior step of the administrative review process and in connection with any concurrent claim or, when available, prior claims for disability benefits”).
B
In making the adverse credibility determination, the ALJ also cited De Guzman’s routine and conservative treatment as well as “absent or only mild objective findings and observation of no major difficulties with functioning during the bulk of her examinations.”
*205Although it is true that De Guzman had major surgery to treat her cervical condition, she herself was “very pleased with her increasing mobility, her lack of pain, and her lack of falls,” and she now admits that “the surgery was successful.” Dr. Johnston observed that she recovered “amazingly well” from the operation, with reduced neck pain that was “mild” and “well controlled with Tylenol.” Her films showed normal spinal alignment, and her doctors were “quite happy” with her outcome.
During her post-operative follow-up visits, she denied having dizziness or light-headedness except at one visit, when she said she had experienced dizziness on and off. However, it had resolved itself by the next visit. She did not report any dizziness to her doctors subsequent to September 2004, though in May 2005 she testified at the first hearing that she still experienced dizziness brought on “[w]hen I get up and also when I clean, like cleaning the bathroom.” Though she testified both in May 2005 and December 2006 that she took daily naps, she attributed them to fatigue rather than dizziness.
She still occasionally reported having headaches associated with a bony mass on her left frontal bone, which was removed in December 2005. By January 2006, she was “doing well” and “[djenying any pain,” though her head and neck pain had apparently returned by the time of her second hearing in December 2006.
Although De Guzman reported a history of leg weakness, knee pain, balance issues, and a history of falling, her x-rays were normal, and her treating physician, Dr. Christine Johnston, noted that her “current constellation of symptoms is not necessarily consistent with any clear neurological process.” After her third fall in March 2004, De Guzman visited the emergency room and “asked [the doctor] to assess [her] for [a] walker.” The doctor complied with this request, and she was fitted with a walker at this time.
In June 2004, a month after her cervical spine surgery, Dr. Johnston noted that “[p]ostoperatively, she is doing amazingly well.... She has not had any falls since the surgery. She has continued to use a walker but is at times able to walk on her own.” By July 2004, she had “regained much of the strength in her legs and [was] able to walk well without use of a walker.” According to her physical therapist, she was “making an excellent recovery.” By August 2004, she reported that “she has been walking in Chinatown and doing her [physical therapy] exercises without difficulty.” 3 Her treating physician, Dr. So-hail Mirza, noted in October 2004 that “it is unlikely her left knee pain is directly related to her cervical problem.”
She did not report falling again until a year later, after her one-year work restriction had ended. Although De Guzman testified at her first hearing that only her left knee gave her problems, in August 2005, she reported that her “right knee gave out, causing her to fall to the ground.” Dr. Vaught reported that there were two falls that month: “[o]ne while hurrying in the house and only using [a] cane” and “[o]ne while walking outside looking up and tripped over tree trunk.” In her assessment, Dr. Vaught noted that the falls had an “unclear etiology,” but because De Guzman had fallen while using a cane, she asked her to use a walker until she could see a physical therapist for “gait *206training” and “recommendations regarding safety.”
In September 2005, a physical therapist assessed De Guzman and determined that she did “not test as a fall risk on the Berg Balance scale.” Furthermore, De Guzman’s knees displayed full muscle strength, full range of motion, and normal alignment. Hence, the physical therapist was “unsure at this time as to why she is falling because objectively she is doing quite well actually.” Because of De Guzman’s recent falls, the physical therapist thought that she “would benefit from using walker vs. cane if fatigued or outside the home” and “discussed importance of attending to environment more closely.” In October 2005, Dr. Vaught reported no falls since her previous visit; she noted that De Guzman “tried to be more careful and pay more attention when walking and th[ought] this has helped.”
“[C]onfliet[s] between a [claimant’s] ... subjective complaints and the objective medical evidence in the record” can constitute “specific and substantial reasons that undermine[ ] ... credibility.” Morgan v. Comm’r Soc. Sec. Admin., 169 F.3d 595, 600 (9th Cir.1999). Although De Guzman testified about headaches, neck pain, dizziness, and knee problems, the objective medical record did not support these subjective complaints. After successful spinal surgery, her condition improved markedly and was well controlled to the point that she needed only yearly follow-ups. Furthermore, her most recently documented falls, which occurred four years ago, resulted from inattention rather than an underlying pathology, as evidenced by the fact that her falling improved after she “tried to be more careful and pay more attention when walking.” It was therefore not unreasonable for the ALJ to conclude that the recommendation to use a cane or walker was merely a precaution based on her subjective complaints rather than a clear endorsement or corroboration of her complaints.
C
The ALJ also referred to De Guzman’s lack of cooperation at examinations as a further reason to disbelieve her subjective complaints. However, the only episode in which she was not fully cooperative was her consultative examination with Dr. Timothy Popanz, in which she “appeared irritable and agitated.” She appeared to have been cooperative during the dozens of other examinations she received from the many doctors she visited. While the ALJ’s reliance on this one episode is not a clear and convincing reason to reject her testimony, it was harmless error in light of the other specific and cogent reasons the ALJ cited in support of his adverse credibility determination. See Carmickle v. Comm’r Soc. See. Admin., 533 F.3d 1155, 1162 (9th Cir.2008).
D
Because the ALJ pointed to “specific, clear and convincing reasons” for disbelieving De Guzman, substantial evidence supports his adverse credibility determination. Smolen v. Chater, 80 F.3d 1273, 1281 (9th Cir.1996).
II
De Guzman contends that the ALJ did not give sufficient reasons for discrediting the opinions of treating physician Dr. So-hail Mirza, examining psychologist Timothy Popanz, Ph.D., and treating physician Amy Vaught.
A
On October 8, 2004, Ms. De Guzman apparently requested and received a form letter from Dr. Mirza (addressed “To *207Whom It May Concern”) stating that “[t]he patient ... [m]ay not return to work/school” for one year from the date of surgery “and [has] permanent limits on bending, lifting, [and] carrying.” The ALJ rejected this opinion on grounds that “prohibiting the' claimant from working for a year after her cervical surgery grossly exaggerates a reasonable recovery time for such a surgery.” 4
In Thomas v. Barnhart, 278 F.3d 947, 957 (9th Cir.2002), we clarified that when evaluating conflicting medical opinions, the “ALJ need not accept the opinion of any physician, including a treating physician, if that opinion is brief, conclusory, and inadequately supported by clinical findings.” Indeed, the ALJ refused to accept Dr. Mirza’s brief, conclusory, and nonspecific statement that Ms. De Guzman has “permanent limits,” which conflicted with his June 2004 report that Ms. De Guzman was “progressing well,” along with Dr. Vaught’s November 2004 report that her cervical pain was merely “mild.” These specific, legitimate reasons for discounting Dr. Mirza’s restrictions constituted substantial evidence.
De Guzman contends that the ALJ was obligated to investigate further what Dr. Mirza meant by “permanent limits.” At the first hearing in 2005, the ALJ asked De Guzman’s counsel to “ask [Dr. Mirza] to be more specific about ‘permanent limits.’ ” De Guzman’s counsel agreed, and the ALJ granted her thirty days to obtain the information, at which time she could “make any additional comments.” Although counsel sent Dr. Mirza a physical capacities evaluation form to fill out, for some reason he did not comply with her request, and counsel did not follow up after Dr. Mirza left Harborview Medical Center.5 The ALJ satisfied his duty to develop the record when he “voiced his concern to Appellant and her counsel, requested an additional inquiry into the basis for Dr. [Mirza’s] opinions and explained that he would keep the record open so that it could be supplemented by the responses ----” Tidwell v. Apfel, 161 F.3d 599, 602 (9th Cir.1998).
B
On July 14, 2003, Dr. Popanz conducted a one-time “diagnostic interview and mental status exam” in which he concluded that De Guzman suffered from depression, dysthymic disorder, and panic disorder with agoraphobia.
The ALJ discounted Dr. Popanz’s opinion because it “provided little detail regarding specific diagnostic criteria to support [her] diagnoses.” The ALJ noted that Dr. Popanz’s conclusions were “based on a subjective report of symptoms and contrast sharply with the other psychological evidence in the record.” In particular, the ALJ pointed to inconsistencies in De Guzman’s reports of depression prior to the examination. Her husband reported an onset date of 1995, yet she was able to work for many years thereafter at a substantial gainful level. Three weeks before her visit, on June 26, 2003, De Guzman claimed that her depression had been in remission for three years until the previ*208ous month; however, on June 30, 2003, she claimed that she “has had depression for the past several months” and was “currently applying for disability based on her depression.”6 As Dr. Popanz’s diagnosis “was premised on [De Guzman’s] own subjective complaints, which the ALJ had already properly discounted” based on inconsistencies, the ALJ proffered specific and legitimate reasons for discounting the diagnosis.
Instead of crediting Dr. Popanz’s opinion, the ALJ credited the Disability Determination Services (“DDS”) psychiatric reviewer’s assessment of De Guzman’s medical history. The DDS reviewer deemed De Guzman’s depression “Not Severe” based on “significant credibility problems” and “significant inconsistencies” between Dr. Popanz’s report and other medical evidence in the file. For example, although she had alleged severe depression when applying for disability benefits in 1996, she returned to full-time work for several years. Moreover, Dr. Popanz’s report failed to describe “full symptoms” for De Guzman’s alleged disorders, and her previous records made “no mention of anxiety, panic attacks, or agoraphobia.” He therefore concluded that the “primary care findings regarding depression are not particularly impressive, and certainly are not consistent with the degree of ... restrictions alleged.”
“In order to discount the opinion of an examining physician in favor of the opinion of a nonexamining medical advisor, the ALJ must set forth specific, legitimate reasons that are supported by substantial evidence in the record.” Van Nguyen v. Chater, 100 F.3d 1462, 1466 (9th Cir.1996). Here the DDS reviewer identified several legitimate reasons for discounting Dr. Po-panz’s diagnosis, including conflicts with the medical record and inconsistent statements. The ALJ properly gave greater weight to the DDS reviewer’s assessment, which was based on De Guzman’s cumulative medical records, rather than Dr. Po-panz’s diagnosis, which was based on a single interview. See Soc. Sec. Rui. 96-6p.
C
In May 2005, Dr. Vaught completed a physical capacities evaluation form that stated that De Guzman could sit for thirty minutes at a time for a total of two hours per day; stand for forty-five minutes at a time for a total of two hours per day; and walk for five minutes per day. According to Dr. Vaught, De Guzman could never do any lifting, carrying, pushing, pulling, bending, squatting, kneeling, crawling, or climbing ladders because “all activities cause pain in [her] neck at [her] prior surgical site.”
The ALJ refused to accord any weight to the form because it did not “indicate any measuring of effort or give[] a description of what activities or movements were actually done ... to determine the capacity found therein.” Moreover, the “limitations are not supported by the medical evidence of record and are out of proportion to any findings within [her] treatment notes.” De Guzman’s only response *209is to demand that the ALJ “recontact [Dr. Vaught] to determine the basis for her opinion.” The ALJ is under no such obligation and is free to reject “check-off reports that d[o] not contain any explanation of the bases of their conclusions.” Crane v. Shalala, 76 F.3d 251, 253 (9th Cir.1996).
Ill
For the foregoing reasons, the district court’s decision affirming the Commissioner’s denial of De Guzman’s application is
AFFIRMED.
This disposition is not appropriate for publication and is not precedent except as provided by 9th Cir. R. 36-3.
. De Guzman and her lay witnesses reported that her husband did most of the household chores, including all the shopping and cooking. Yet, the ALJ noted that her husband had been on disability for a long time. It is unclear whether the ALJ's incredulity can be ascribed to her claim of disability or his, so we do not count this potential inconsistency against her.
. De Guzman had unsuccessfully applied for disability benefits back in 1997 but represented on her 2002 application that "no previous application has been filed with the Social Security Administration by or for” her.
. These exercises included “tandem walking and standing,” "forward lunges,” and “wall squats.”
. The ALJ inadvertently conflated the names of two of De Guzman’s treating physicians, Dr. Mirza and Dr. McCarthy.
. The ALJ also cited Dr. Mirza’s refusal to fill out the physical capacities evaluation as evidence that he was not comfortable supporting De Guzman’s disability claim. However, it is unclear from the record why he did not fill out the paperwork, as he was not even present at the visit. Although we agree that this inference is not a clear and convincing reason to discredit Dr. Mirza’s opinion, it was harmless error in light of the other reasons the ALJ cited. See Carmickle, 533 F.3d at 1162.
. At that visit, she requested and was prescribed Paxil. The ALJ noted that other than this prescription, “the claimant has sought no formal mental health treatment” “[A]n unexplained, or inadequately explained, failure to seek treatment or follow a prescribed course of treatment” can cast doubt on the sincerity of a claimant's subjective complaints. Fair, 885 F.2d at 603. However, in the case of a mental health disorder, failure to seek treatment may be an unfortunate result of the disorder. See Van Nguyen v. Chater, 100 F.3d 1462, 1465 (9th Cir.1996). The ALJ improperly relied upon this factor in discounting Dr. Popanz’s opinion, but this error was "inconsequential to the ultimate nondisability determination.” Stout v. Comm’r Soc. Sec., 454 F.3d 1050, 1055 (9th Cir.2006).