BOGGS, Circuit Judge.
Paul Morris was injured in a work-related automobile accident in 1992 and, as a consequence, began receiving long-term disability benefits from the Appellee (“the Plan”) in 1993. In 2004, the Plan’s new third-party administrator, Broadspire, requested documentation of Morris’s ongoing disability. Following a series of independent examinations and Plan-sponsored file reviews, the Plan terminated Morris’s long-term disability benefits. Morris, after exhausting his internal appeals, appealed to the district court, which affirmed the Plan’s determination as being neither arbitrary nor capricious. For the reasons discussed below, we affirm the judgment of the district court.
I
Paul Morris was employed as a meter reader for American Electric Power (“AEP”) when he was injured in an on-the-job automobile accident in August 1992. Though he continued to work in the months immediately following the accident, by April 1993 his condition was such that he could not perform the walking, climbing, standing, and moving of objects that his job required. On that basis, and further claiming that he suffered from depression and post-traumatic stress disorder as a result of the accident, Morris applied for and was granted long-term disability status by AEP’s then-plan administrator, Aetna. At the same time, and with the encouragement of the Plan, he applied for and obtained Social Security disability benefits.
Under the terms of the Plan, disability determinations are two-tiered. For the first two years that a participant receives long-term disability benefits, he is considered disabled if he is unable to perform the work required of him in his “own occupation” — that is, the occupation he held immediately prior to going on disability. After that initial two-year period, in order for long-term disability benefits to continue, the participant must be unable to perform any occupation that he is qualified to perform, as long as that occupation pays at least 60% of his pre-disability base rate of pay and is consistent with the participant’s formal education, training, and work experience (the “any-oecupation standard”).
[980]*980In April 1995, Aetna informed Morris that he had met the any-occupation standard, entitling him to benefits beyond the initial two-year time frame. Under the terms of the plan, Morris was subject to provisions requiring him to provide continuing objective proof of his disability at least once each year.
Over the next ten years, Morris continued to see a neurologist, Dr. Nahid Dad-mehr, for complaints of headaches, pains in his back and legs, and right elbow pain. He also saw a psychiatrist, Dr. Maureen Stark, on several occasions for depression.
In June 2004, the Plan’s new claims administrator, Broadspire, sought updated documentation of Morris’s disability pursuant to the terms of the plan document. Broadspire asked Morris to complete a questionnaire and an “attending physician statement and evaluation of physical abilities” and to submit medical records as proof of continued disability. Morris eventually submitted those forms, which indicated that he was still limited by back and leg pain, as well as by panic attacks, depression, and irritability. The plan then requested that he complete a functional capacity evaluation (“FCE”), which resulted in a finding that Morris was “functioning in the Medium Physical Demand Category.” The FCE evaluator, Jessica lams, believed that Morris “appeared] rehabable if pain can be controlled,” and recommended “frequent change in position, allowances for self management of symptoms while working or exercising, and some type of work conditioning program” to facilitate a return to regular daily work.
The Plan also arranged for Morris to undergo an independent psychiatric examination by Dr. Don Mclntire, the results of which included a diagnosis of moderate Bipolar II Disorder and mild post-traumatic stress disorder. Morris tested within the average range for overall intellectual functioning, and did not exhibit any clinically significant behavioral impairments during the evaluation. Dr. Mcln-tire also administered an MMPI-2 personality inventory assessment, which revealed that Morris had significant difficulties with depression and anxiety and was likely to be uncomfortable being around other people.
Upon receiving the results of the FCE and Dr. Mclntire’s examination, Broad-spire forwarded them to its own reviewing physicians: Dr. Vaughn Cohan, a neurologist, and Dr. Barry Glassman, a psychiatrist. Dr. Cohan concluded that Morris’s medical file showed his neurological symptoms to be consistent with sedentary or light work, and that there was no indication that the medications Morris was taking would prevent him from working. Dr. Glassman agreed that Morris would be best placed in a low-stress, physically undemanding position, but also noted that there were no examination findings to indicate that such would be medically necessary.
The Plan then obtained an employment assessment report (“EAR”), which was prepared after a vocational consultant interviewed Morris and reviewed the results of the FCE and Dr. Mclntire’s psychiatric examination, as well as the reports of Drs. Cohan and Glassman. Taking into account the “sedentary or light work” limitations indicated by Dr. Cohan’s recommendation, the EAR identified two occupations that Morris could perform: Automatic Presser and Shirt Presser. Both of these jobs were indicated to be available in Columbus, Ohio, where Morris resided, and met the necessary salary and physical requirements. As a result of the EAR’s identification of jobs that Morris could perform, Broadspire determined that Morris did not meet the any-occupation standard and [981]*981would no longer be eligible for disability payments.
Morris appealed, using the Plan’s internal appeals process. Over the course of the two appeals permitted by the plan document, Morris submitted additional materials from his treating physicians and records from a recent hospital stay incurred after he experienced “seizure-like symptoms.” The Plan obtained reports from eight more reviewing physicians, each of which found no objective data to support a finding of disability under the any-occupation standard. The Plan denied Morris’s appeals, and this litigation followed.
II
A
We “review de novo the decision of a district court granting judgment in an ERISA disability action based on an administrative record.” Glenn v. MetLife, 461 F.3d 660, 665 (6th Cir.2006), aff'd, Metro. Life Ins. Co. v. Glenn, 554 U.S. 105, 128 S.Ct. 2343, 171 L.Ed.2d 299 (2008). If the plan administrator is vested with discretion to interpret the plan, we review the administrator’s denial of benefits to determine whether that denial was arbitrary and capricious. Ibid, (citing Firestone Tire & Rubber Co. v. Bruch, 489 U.S. 101, 115, 109 S.Ct. 948, 103 L.Ed.2d 80 (1989)). Though highly deferential, this standard nevertheless requires “some review of the quality and quantity of the medical evidence and the opinions on both sides of the issues” and does not require us merely to rubber-stamp the administrator’s decision. McDonald v. Western-Southern Life Ins. Co., 347 F.3d 161, 172 (6th Cir.2003).
The ultimate question in any given disability case on “arbitrary and capricious” review is whether a plan can offer a reasoned explanation, based on the evidence, for its judgment that a claimant was not “disabled” within the plan’s terms. Elliott v. Metro. Life Ins. Co., 473 F.3d 613, 617 (6th Cir.2006). In pertinent part, the plan here at issue reads:
After the first 24 months following your date of disability, disability is defined as an illness or injury that requires the regular treatment of a duly qualified physician that may reasonably be expected to prevent you from performing the duties of any occupation for which you are reasonably qualified by your education, training and experience. If you meet the requirements of this “two-year test,” you will continue to be considered to be disabled and receive benefits up to the maximum benefit period.
Once the Plan makes a judgment as to whether a claimant has met the requirements of the “two-year test,” its determination is entitled to deference unless it is arbitrary and capricious. That is, if the Plan’s determination that an individual can perform any occupation “is the result of a deliberate, principled reasoning process and if it is supported by substantial evidence,” it will not be disturbed.
B
Notwithstanding the deference afforded a plan administrator’s decisions under arbitrary-and-capricious review, courts must evaluate potential conflicts of interest and consider them as factors in determining whether the decision to deny benefits was arbitrary and capricious. Gismondi v. United Techs. Corp., 408 F.3d 295, 298 (6th Cir.2005). For ERISA purposes, a conflict of interest is present when the same entity both funds the plan and evaluates claims for benefits thereunder. Glenn, 128 S.Ct. at 2348; DeLisle v. Sun Life Assur. Co. of Canada, 558 F.3d 440, 445 (6th Cir.2009). At the root of this kind of conflict is the fact that “every dollar provided in benefits [982]*982is a dollar spent by ... the employer; and every dollar saved ... is a dollar in [the employer’s] pocket.” Glenn, 128 S.Ct. at 2348 (quoting Bruch v. Firestone Tire & Rubber Co., 828 F.2d 134, 144 (3d Cir. 1987)). In such circumstances, an employer’s immediate financial interest creates an inherent conflict with its fiduciary duty under ERISA, which requires that plan administrators discharge their duties solely in the interests of the participants and beneficiaries of the plan. Id. at 2350. Application of this “heightened” form of arbitrary-and-capricious review is shaped by the circumstances of the inherent conflict of interest. Borda v. Hardy, Lewis, Pollard & Page, P.C., 138 F.3d 1062, 1069 (6th Cir.1998).
Morris has conceded that the plan administrator in this case is vested with sufficient interpretive discretion to qualify for review under the arbitrary-and-capricious standard. He does contend, however, that a conflict of interest exists as to Broad-spire, the plan administrator, in that Broadspire’s advertising materials contain language indicating that its mission is to help its clients “contain costs” and “have a positive impact on [their] employees and [them] bottom line.” In Morris’s view, such promises align Broadspire’s interests with those of the Plan enough to conclude that the two ought to be treated interchangeably for conflict-of-interest analysis, and therefore it is functionally true that the same entity both makes eligibility determinations under the plan and pays benefits to those found eligible.
We do not appear to have ever held that a conflict of interest inherently exists in such a situation, and we will not do so now. In Glenn, the Supreme Court made it clear that the conflicted entity need not be an employer, but could instead be (for example) an insurance company — as long as it was the same entity that performed both the benefits-determination and the benefits-payout functions. Though the Court acknowledged that insurance companies were not necessarily affected to the same extent by the prospect of paying out claims they determined to be valid (because they could recoup a fraction of their costs by imposing a per-claim service charge), they nevertheless suffered an immediate financial loss. The difference, in other words, was one of degree.
Here, the difference is one of kind. Morris is, in effect, arguing that the third-party administrator has implemented a long-term strategy to carry out its duties as plan administrator for the benefit of the employer, rather than for the benefit of the plan participants and beneficiaries— presumably on the belief that, in so doing, they will be able to attract new and continued business, thus eventually lining them own pockets. Such an accusation is critically distinct from the inherent conflict of interest present when a decision maker will benefit or suffer as a direct consequence of her decisions. Because Broad-spire’s theoretical advantage from making benefit denials would be neither immediate nor guaranteed, it cannot be said to be “inherent” in the way that a conflict is when the same entity is responsible for both benefits eligibility determinations and benefits payments. Even if we were to assume, therefore, that an independent claims administrator could be so motivated by market forces as to effectively be an alter-ego of those funding the plan, it would be necessary to demonstrate that the administrator had, in fact, acted on those motivations before a conflict could be said to exist.
There is no such evidence in our case. The appellant relies on generic promotional language contained on Broadspire’s website. Such language is of a kind that one would very much expect to see, given [983]*983that the entity responsible for paying benefits is generally the one that selects a third-party administrator. Advertisements emphasizing cost savings to that entity are neither surprising nor especially probative of conflict. Further, the language cited by the appellant is hardly damning: promising to “contain costs” and “have a positive impact on your employees and your bottom line” is a far cry from even implying that the administrator will evaluate otherwise-viable claims harshly. Indeed, Broadspire’s advertisements could simply refer to the benefits of economies of scale, or to the savings to a company from using expert claims administrators to quickly pay legitimate claims while weeding out fraudulent ones.1
Morris argues that a conflict in cases like ours is inherent, because no self-funded plan would turn over administration of its claims -without an expectation that the administrator will act in its interest. This simply proves too much. Taken as true, this argument would necessitate a finding of conflict every time a third-party administrator was used. Because it is generally true that the failure to use a third-party administrator — i.e., when the same entity both determines claim eligibility and pays benefits — produces a conflict of interest, the holding Morris proposes would effectively render the distinction between “normal” arbitrary and capricious review and the so-called “heightened” arbitrary and capricious review utilized in the presence of a conflict meaningless.2
In this case, we therefore apply ordinary arbitrary-and-capricious review to the Plan’s decision to deny long-term disability benefits to Morris.
Ill
Morris argues that Broadspire’s denial of claims was arbitrary and capricious in that there was no evidence in the record suggesting that his medical condition had improved between the time the Plan most recently found him to be disabled and the time his benefits were terminated.
In Kramer v. Paul Revere Life Ins. Co., 571 F.3d 499 (6th Cir.2009), we held cancellation of benefits by a plan administrator to be arbitrary and capricious when done in the absence of evidence showing that the claimant’s condition had improved, and no explanation existed for the apparent discrepancy from earlier assessments. Morris cites Kramer, along with three district courts in our circuit, as support for his argument that we have adopted an Eighth Circuit decision — McOsker v. Paul Revere Life Ins. Co., 279 F.3d 586 (8th [984]*984Cir.2002) — that holds, in Morris’s words, “that an administrator acts arbitrarily and capriciously when it terminates benefits absent an indication in the record that a claimant’s medical condition had improved sufficiently to justify terminating long-term disability benefits previously awarded.”
Morris overstates the holdings of the cases he cites from within our circuit and, in our view, also overstates the holding of McOsker. In Kramer, we indicated that:
Moreover, there is no explanation for the decision to cancel benefits that had been paid for some five years based upon the initial determination of total disability in the absence of any medical evidence that the plaintiffs condition had improved during that time. The best that can be said of the opinions of [the company’s consultants] is that they support the proposition that [the claimant] was, in fact, never disabled from her “own occupation.” But that conclusion flies in the face of all the other evidence in the record, and the plan administrator’s reliance upon it can only be described as arbitrary and capricious.
Kramer, 571 F.3d at 507 (emphasis in original). This is not, however, a wholesale endorsement of the notion that a lack of evidence of improvement is sufficient unto itself to prove a plan administrator’s decision arbitrary and capricious. Rather, Kramer coupled the absence of such evidence with a lack of explanation or support for the plan’s decision, among other factors.
Surely it is reasonable to require a plan administrator who determines that a participant meets the definition of “disabled,” then reverses course and declares that same participant “not disabled” to have a reason for the change; to do otherwise would be the very definition of “arbitrary and capricious.” It does not follow, however, either logically or from our decision in Kramer, that the explanation must be that the plan administrator has acquired new evidence demonstrating that the participant’s medical condition has improved. While Morris contends that the legal issue “is not whether there is ‘new evidence,’ but whether the evidence — new or old — establishes that Morris’s medical condition had improved,” the ultimate question is whether the plan administrator had a rational basis for concluding that Morris was not disabled at the time of the new decision. Under the any-occupation standard at issue in this case, any number of factors could be germane to such a determination — including evidence of improvement, certainly, but also including evidence better defining the participant’s medical condition, or even, given the plan’s definition of “disabled,” newly-acquired skills that would permit the previously disabled participant to perform an occupation he had not been qualified for at the time of his disability.3
Such an interpretation comports not only with the “no explanation” language in Kramer, but also with each of the other cases cited by Morris. These cases, at most, merely characterize the lack of evidence of improvement as a circumstance to be weighed by a reviewing court, rather than per se proof of arbitrariness. See McOsker, 279 F.3d at 589 (“We are not suggesting that paying benefits operates forever as an estoppel so that an insurer can never change his mind; but unless information available to an insurer alters [985]*985in some significant way, the previous payment of benefits is a circumstance that must weigh against the propriety of an insurer’s decision to discontinue those payments”) (emphasis added); Rabuck v. Hartford Life and Accident Ins. Co., 522 F.Supp.2d 844, 872 (W.D.Mich.2007) (characterizing prior payment of disability benefits as a circumstance that weighs against the propriety of discontinuing payments); Boyd v. Am. Elec. Power Sys. Long-Term Disability Plan, 2007 WL 2778667 (S.D.Ohio Sept.21, 2007) (“The fact that defendant’s decision to terminate benefits was not based on new significant evidence or evidence of improvement since the prior determination of disability weighs against the propriety of defendant’s decision to terminate plaintiffs LTD benefits.”). As a “circumstance to be weighed,” we may therefore accord greater or lesser import to a lack of evidence of improvement, depending upon whether that lack is explained by other relevant information.
In this case, there is no reason to believe that the purported lack of evidence of improvement rendered the Plan’s decision arbitrary, given the extent to which it obtained new information about his current condition. The Plan’s decision was based upon the recommendations of ten reviewing physicians and an employability assessment report. These sources were, in turn, provided with contemporaneously-issued reports from an independent psychologist, an independent functional capacity evaluator, and, in most cases, the entire contents of the Plan’s file relating to Morris dating back to 1993. There is, in fact, nothing in the record to indicate that Morris even underwent an EAR prior to September 15, 2005, let alone that the Plan had access to the information generated by the FCE and the independent psychiatrist’s examination. Even assuming that these reports did not explicitly demonstrate “improvement” in the sense of showing that the specific physical and psychological bases for Morris’s initial disability determination had been ameliorated, the additional detail and analysis they provide serve as adequate explanation for the administrator’s decision to revoke those benefits under the arbitrary and capricious standard.
Morris also argues that Broad-spire’s decision was arbitrary and capricious because it did not address the fact that Morris had successfully applied for Social Security disability benefits in 1993.
We have held that if a plan administrator (1) encourages an applicant to apply for Social Security disability payments, (2) financially benefits from the applicant’s receipt of Social Security, and then (3) fails to explain why it is taking a position different from the Social Security Administration on the question of disability, a court reviewing that determination should weigh this in favor of a finding that the decision was arbitrary and capricious. Bennett v. Kemper Nat’l Svcs., Inc., 514 F.3d 547, 553 (6th Cir.2008). The rationale for according this factor any weight is that it serves as a form of estoppel, reducing fraud in the legal process by forcing a modicum of consistency on a repetitive litigant. Glenn, 461 F.3d at 667-68 (quoting Ladd v. ITT Corp., 148 F.3d 753, 756 (7th Cir.1998)).
Morris asserts (and the Plan does not dispute) that the Plan encouraged him to apply for Social Security disability payments when he first became disabled, and that the Social Security Administration’s subsequent grant of disability benefits lightened the Plan’s financial obligation to Morris under the terms of the plan document. It is similarly undisputed that Broadspire did not refer at all to Morris’s Social Security award in its initial letter [986]*986denying him benefits, and merely listed it as one of seventy or so documents that had been reviewed in its second and third denial letters.
Again, however, the language of Bennett indicates that a failure to take into account a Social Security disability award is to be weighed in favor of a finding that the decision was arbitrary and capricious, not that such a decision is arbitrary and capricious per se. As such, we may give it greater or lesser weight, depending on the circumstances. Given that the purpose of weighing it at all is to prevent the plan, as an interested party, from effectively committing fraud on the court by taking inconsistent positions to minimize its financial exposure, it makes a great deal of sense to weigh an unexplained inconsistency in this area heavily when those positions are taken in quick succession.4
In this case, however, the dissonance between the plan’s encouragement of Morris’s Social Security claim and its subsequent denial of benefits is muted, because more than twelve years had passed between the time Morris was determined to be disabled by the Social Security Administration and the time Broadspire informed him that they no longer considered him disabled. It seems highly unlikely that a plan would encourage a claimant to apply for Social Security disability benefits as part of a disingenuous scheme to lessen its potential exposure on a claim, then abide twelve years while paying its own share of disability benefits before cutting those benefits off for no legitimate reason. It is far more reasonable to believe that a 1993 determination of disability was simply not very relevant to a 2005 decision on the same issue.
To be sure, Broadspire’s decision to cut off benefits is inconsistent with the Plan’s previous encouragement of Morris to apply for Social Security disability, and to that extent weighs slightly in Morris’s favor when it comes to evaluating whether that decision was arbitrary and capricious. We will not merely assume, in the absence of an explanation by Broadspire, that it considered the Social Security Administration’s determination to be obsolete and thereby excuse their failure to take it into account entirely. Given that the Plan’s encouragement of Morris to apply for those benefits happened so long ago, however, and that it obtained new information about Morris’s disability from numerous sources in a contemporaneous investigation, we do not find the Plan’s behavior to be particularly troubling.
C
We now turn to the question of whether Broadspire acted arbitrarily and capriciously in evaluating the medical evidence available to it while making the decision to terminate Morris’s benefits. Morris contends that Broadspire ignored the medical evidence and the conclusions contained in the reports of his own doctors, as well as those set forth in the independent functional capacity evaluation and the independent psychiatric evaluation.
Generally speaking, a plan may not summarily reject the opinions of a treating physician, but must instead give reasons for adopting an alternative opinion. Elliott, 473 F.3d at 620. Giving greater weight to a non-treating physician’s opinion for no apparent reason lends force to the conclusion that a plan administrator’s decision is arbitrary and capricious. Ibid. Plan administrators, however, “are not ob[987]*987ligated to accord special deference to the opinions of treating physicians.” Black & Decker Disability Plan v. Nord, 538 U.S. 822, 825, 123 S.Ct. 1965, 155 L.Ed.2d 1034 (2003). To that extent, a lack of objective medical evidence upon which to base a treating physician’s opinion has been held sufficient reason for an administrator’s choice not to credit that opinion. See, e.g., Boone v. Liberty Life Assur. Co. of Boston, 161 F. App’x 469, 473 (6th Cir.2005) (administrator’s decision not to credit treating physicians’ assessments not arbitrary because the assessments were not supported by the objective evidence, as required by the plan document).
In looking to the various communications sent from Broadspire to Morris during the process of terminating Morris’s benefits, we find that Broadspire set forth a lack of objective data as its primary reason for choosing not to credit the assessments of Drs. Dadmehr and Stark. In addressing the conclusions of Dr. Dad-mehr, Broadspire’s letter denying Morris’s first internal appeal indicated that
... there were no physical examination findings included in this note substantiating Mr. Morris’ inability to perform the duties of any occupation. Therefore based on the medical evidence, Mr. Morris can sit, stand and ambulate. He is also capable of lifting up to 8 to 10 pounds on a repetitive basis. Therefore, based on the medical information from an orthopedic standpoint Mr. Morris is capable of performing the duties of any occupation.
With regard to Morris’s alleged neurological problems, the appeal decision observed that “there weren’t any clinical neurological abnormalities documented by Dr. Dad-mehr from the neurological examinations. There also was no clinical documentation provided to support recurrent seizure activity in the last ten years. Additionally, there was no clinical documentation of a recurrence of syncopal episodes since the early 1990’s.”
Similarly, Broadspire’s first appeal denial indicated that “there were no documented physical examination findings, lab abnormalities including blood sugar values such as a hemoglobin AIC provided for review to determine the degree of control of his diabetes.” With regard to his psychiatric condition, the letter notes that “Dr. Stark did not provide any specific examination findings substantiating the presence of impairment in psychological functioning,” and “did not provide findings to substantiate impairment in cognitive, emotional, or behavioral functioning that would preclude Mr. Morris from performing the duties of any occupation.”
The Plan’s references to a lack of objective data to support Morris’s physicians’ conclusions are consistent with its overall conclusion that “[bjased on a review of the aforementioned medical data, there was a lack of medical evidence to substantiate ... inability to perform any occupation.” It is also consistent with the Plan Summary, which states:
To continue receiving benefits, you will be required to provide continuing proof of your disability at least once each year. Objective medical evidence must be supplied supporting your case for disability. A letter from your treating physician merely stating that you are unable to work without any supporting information will not be considered as conclusive proof of your disability.
Nor was there any great mystery as to what kinds of objective medical data Broadspire was looking for in this case. Broadspire’s first appeal denial informed Morris that it would consider relevant data to include specific psychiatric findings, a complete orthopedic examination, complete neurological testing, other abnormal diag[988]*988nostic test results, and examination findings to substantiate the presence of cognitive, emotional, or behavioral impairment. Under the terms of the plan document, it was Morris’s responsibility to obtain this data; he did not, and therefore the Plan was not arbitrary and capricious in choosing to discount the conclusions of his treating physicians.5
Moreover, Broadspire did not ignore the findings of its own independent examinations. Morris points out that neither the functional capacity evaluation nor the independent psychological examination came to a conclusion as to whether Morris could return to work. Yet a definitive conclusion as to whether Morris met the any-occupation standard does not appear to have been the point of either exam. Rather, they provided information on Morris’s relevant physiological and psychological limitations, which were reviewed by the vocational counselor who prepared the Employability Assessment Report that concluded Morris could work as an automatic presser or shirt presser. While Morris makes much of a single sentence in the FCE noting that “[h]e appears rehabable if pain can be controlled,” (implying that he cannot return to work without rehab, and cannot be rehabbed without pain control), the FCE puts Morris in the “Medium Physical Demand Category,” whereas the ultimate conclusion of the EAR and Plan was that Morris could return to sedentary or light work. The jobs in the sedentary/light work category would presumably involve lesser physical demands than Morris was already capable of during the FCE, and could also include the “frequent change in position, allowances for self-management of symptoms while working or exercising, and ... work conditioning” recommended by the FCE evaluator. Similarly, while the independent psychological examination conducted by Dr. Mclntire did not opine one way or another as to the question of whether Morris could “return to work,” it pointed out that he had difficulties with depression and anxiety, and in interacting with others. These conclusions, too, were reviewed for the EAR.6
D
The remainder of Morris’s arguments may be disposed of in short order.
Morris argues that the plan administrator’s decision to terminate his benefits was arbitrary and capricious because it was done without considering whether, after more than twelve years of being out of the workforce, he was “employable.” This argument has some basis in the language of [989]*989the plan document, which requires that the Plan may only consider Morris to be not disabled under the any-occupation standard if Morris can work in an occupation that is “consistent with the participant’s formal education, training, and work experience.” However, it appears that the EAR relied upon by the plan administrator did take those factors into account. Generally, the purpose of the report was to make a determination regarding the potential to find Morris “feasible employment”; though the meaning of “feasible” is not defined in the EAR, the report specifically details Morris’s education, notes his job skills, and makes an explicit “transferability of skills analysis” in order to identify current jobs that made use of his previously-existing skills. Morris presents no evidence that his prior skills have declined, nor that the occupations identified by the EAR require particularly up-to-date knowledge.
Morris also argues that Broadspire did not provide his full medical file to the vocational consultant who prepared the EAR, instead creating a misleading picture of Morris’s condition by withholding the reports of Morris’s own treating physicians. See Spangler v. Lockheed Martin Energy Systems, Inc., 313 F.3d 356, 361-62 (6th Cir.2002) (“cherry-picking” claimant’s file in hopes of obtaining a favorable report from a vocational consultant as to the claimant’s ability to work held arbitrary and capricious). In Spangler, however, the decision to terminate was based solely on a transferability of skills analysis performed by an expert who had been forwarded a single aberrant physician’s report. In this case, the vocational consultant was provided with both of the independent examination reports, along with the report of a reviewing physician considered the most favorable to Morris of the ten such reports prepared at the behest of the plan. Moreover, the decision to terminate was not based solely on the EAR, but also on those reviewing physicians’ reports.7
Morris further contends that the Plan’s decision to terminate his benefits was arbitrary and capricious because it relied on peer-review physicians who had conflicts of interest stemming from long-term professional relationships with Broadspire. Morris cites our decision in Kalish v. Liberty Mutual/Liberty Life Assur. Co. of Boston, 419 F.3d 501, 507-08 (6th Cir.2005) for the proposition that “a plan administrator, in choosing the independent experts who are paid to assess a claim [has] a clear incentive to contract with individuals who were inclined to find it its favor that [a claimant] was not entitled to continued [disability] benefits.” In Kalish, however, we also noted that conclusory allegations of bias with respect to a plan-chosen reviewer, without statistical evidence that the reviewer consistently opined the claimants were not disabled, could not permit a conclusion that relying on that reviewer’s opinion was arbitrary and capricious. Id. at 508. Morris seeks to forestall this objection by averring that a WESTLAW database search identified 30 cases since 2005 in which Dr. Cohan served as a physician-reviewer for Broadspire, along with various lesser numbers for eight of the other reviewing physicians. Yet Kalish requires evidence that reviewers consistently opine that claimants are not disabled, not that they frequently do so. Morris provides no context for his numbers; we do not know, for example, if Dr. [990]*990Cohan’s “not disabled” findings account for 30 cases out of 30 in which he was retained by Broadspire, or 30 out of 300.
Finally, Morris claims that the Plan was arbitrary and capricious in relying upon peer-review physicians who made “critical credibility determinations” as to his psychiatric condition without having evaluated him personally. We have held that whether a doctor has physically examined the claimant is one factor that may be considered in determining whether a plan administrator acted arbitrarily and capriciously in giving greater weight to the opinion of its own consulting physician. Kalish, 419 F.3d at 509. However, “reliance on a file review does not, standing alone, require the conclusion that [a plan administrator] acted improperly.” Id. Importantly, this is not a situation in which the reviewing psychiatrists and psychologists are flatly contradicting the conclusions of those who examined the patient, nor is it the case that the Plan did not have Morris examined personally. Dr. Stark, Morris’s treating psychiatrist, consistently indicated that she believed Morris could perform some “parttime, low-stress work.” Dr. Mcln-tire, the independent psychiatric evaluator, saw Morris personally and recommended limitations that were taken into account by the EAR. Dr. Mclntire’s observations were also provided to the reviewing psychiatrists who made the final recommendations to the administrator as to Morris’s psychiatric limitations. Thus, to the extent that the Plan’s determinations relied upon an accurate assessment of Morris’s credibility, the Plan was not arbitrary and capricious in relying on recommendations that were based on the in-person evaluation of an independent medical professional and were not inconsistent with the patient’s own treating psychiatrist.
IV
For the reasons discussed above, we conclude that the Plan appears to have put forth a reasoned explanation based on the evidence for the termination of Morris’s benefits. The opinions of Morris’s own physicians, to they extent that they supported Morris’s disability claim, were unsupported by the objective evidence required in the plan document. By contrast, an employability assessment report that was based on objective evidence, an interview with Morris, and a relatively claimant-favorable report by a reviewing physician concluded that he could perform certain kinds of light-duty work. Though we find Broadspire’s failure to address the grounds for Morris’s Social Security disability award relevant as a matter of law, when weighed in the balance and considered in context it does not suffice to render the administrator’s decision arbitrary and capricious. The judgment of the district court is therefore AFFIRMED.