Ovist v. Unum Life Ins Co(.) of America

14 F.4th 106

Opinion

United States Court of Appeals For the First Circuit

No. 20-1464 RHONDA OVIST,

Plaintiff, Appellant,

v.

UNUM LIFE INSURANCE COMPANY OF AMERICA;

UNUM GROUP,

Defendants, Appellees.

APPEAL FROM THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF MASSACHUSETTS

[Hon. Timothy S. Hillman, U.S. District Judge]

Before

Lynch and Selya, Circuit Judges, and Laplante,* District Judge.

Jonathan M. Feigenbaum for appellant.

Katrina T. Liu, Trial Attorney, Plan Benefits Security Division U.S. Department of Labor, Kate S. O’Scannlain, former Solicitor of Labor, G. William Scott, Associate Solicitor for Plan Benefits Security U.S. Department of Labor, and Thomas Tso, Counsel for Appellate and Special Litigation U.S. Department of Labor, on brief for the Secretary of Labor, amicus curiae.

Joseph M. Hamilton, with whom Mirick, O’Connell, DeMallie & Lougee, LLP was on brief, for appellees.

* Of the District of New Hampshire, sitting by designation.

September 22, 2021

3

LAPLANTE, District Judge. This case involves a dispute over the applicability of a self-reported symptoms benefit limitation provision to a long-term disability claim. Plaintiff- appellant Rhonda Ovist is a participant in her employer’s long- term disability plan (“the Plan”), which is insured and administered by defendant-appellee Unum Life Insurance Company of America and governed by the Employee Retirement Income Security Act (“ERISA”), 29 U.S.C. §§ 1001 et. seq. Ovist, who suffers from Chronic Fatigue Syndrome (“CFS”), fibromyalgia, and associated symptoms including pain and fatigue, was granted benefits under the Plan in 2011. The Plan provides for a maximum benefit period of 24 months for “disabilities due to mental illness and disabilities based primarily on self-reported symptoms.” Unum terminated Ovist’s benefits under this provision (“the SRS limitation”) in February 2015, after paying benefits to Ovist for about 43 months.

Following an unsuccessful administrative appeal, Ovist filed an ERISA action in the U.S. District Court for the District of Massachusetts, seeking recovery and reinstatement of her benefits as well as attorneys’ fees and costs. The parties filed cross-motions for summary judgment, and the district court granted Unum’s motion but denied Ovist’s cross-motion. Ovist v. Unum Life Ins. Co. of Am., No. 4:17-cv-40113-TSH, 2020 WL 1931755 (D. Mass.

4

Feb. 21, 2020), report and recommendation adopted, No. 4:17-cv- 40113, 2020 WL 1931958 (D. Mass. Mar. 27, 2020).

Ovist challenges Unum’s decision to terminate her benefits; in particular, Ovist objects to Unum’s requirement that she provide objective evidence of her functional limitations1 in order to avoid the SRS limitation. This requirement, Ovist contends, runs counter to the parameters of the SRS limitation, as interpreted by the Seventh Circuit Court of Appeals in Weitzenkamp v. Unum Life Insurance Company of America, 661 F.3d 323 (7th Cir. 2011). In Weitzenkamp, the court concluded that a nearly identical SRS limitation is applicable to “disabling illnesses or injuries that are diagnosed primarily based on self-reported symptoms rather than to all illnesses or injuries for which the disabling symptoms are self-reported.” Id. at 330 (emphasis in original). Ovist further argues that the SRS limitation does not apply to her claim under Unum’s interpretation of the provision, since she provided objective evidence of her functional limitations, which Unum unreasonably rejected.

We decline Ovist’s invitation to adopt the Seventh Circuit Court of Appeals’ holding in Weitzenkamp, concluding

A “functional limitation” is a “restriction or lack of 1

ability in performing an action as a result of a disability.” American Psychological Association, APA Dictionary of Psychology, at https://dictionary.apa.org/functional-limitation. We use the terms “functional loss” and “functional limitations” interchangeably.

5

instead that Unum’s objective evidence requirement is permissible under this circuit’s precedent and is consistent with a reasonable interpretation of the SRS limitation provision. Further, we find that Unum’s determination that Ovist lacked objective proof of her functional limitations rests on substantial evidence in the record, and is thus not arbitrary or capricious. We accordingly affirm the entry of summary judgment to Unum.2 I. Facts Ovist began working as a sociology professor at Rollins College in 1999. Her responsibilities included teaching, advising students, developing courses, grading papers, and researching and writing. In its claim log, Unum noted that Ovist began reporting and seeking treatment for “severe fatigue and generalized diffuse pain in 2003,” and she was treated with narcotics and “several courses of antiviral agents.” Ovist’s complaints of pain and fatigue continued over the years. Rollins College approved Ovist’s request for short-term disability leave from January 9, 2011 to June 1, 2011, based on diagnoses of chronic fatigue syndrome (“CFS”), cytomegaloviral illness, sleep apnea, chronic sinusitis,

2 The Secretary of Labor filed a brief in this case as amicus curiae. In the brief, the Secretary supports placing the burden on the plan, and not the plan participant, to prove the applicability of a benefit limitation, such as the SRS limitation. We acknowledge and appreciate the Secretary’s assistance in this case.

6

and parvovirus. In June 2011, Ovist applied for long-term disability benefits under the Plan. A. Relevant Terms of the Plan The Plan “delegates to Unum . . . discretionary authority to make benefit determinations under the Plan[,] . . . includ[ing] determining eligibility for benefits and the amount of any benefits, resolving factual disputes, and interpreting and enforcing the provisions of the Plan.” An individual is considered “disabled” and eligible for benefits under the Plan when she is “limited from performing the material and substantial duties of [her] regular occupation due to [her] sickness or injury; and [has] a 20% or more loss in [her] indexed monthly earnings due to the same sickness or injury.”

This case centers on Unum’s interpretation and application of the SRS limitation, a provision of the Plan that limits the benefit period to 24 months for certain disabilities. The SRS limitation provides:

The lifetime cumulative maximum benefit period for all disabilities due to mental illness and disabilities based primarily on self-reported symptoms is 24 months. Only 24 months of benefits will be paid for any combination of such disabilities even if the disabilities:

are not continuous; and/or are not related.

The Plan defines mental illness as:

A psychiatric or psychological condition classified in the Diagnostic and Statistical Manual of Mental Health Disorders (DSM},

7

published by the American Psychiatric Association, most current as of the start of a disability. Such disorders include, but are not limited to, psychotic, emotional or behavioral disorders, or disorders relatable to stress.

And the Plan defines self-reported symptoms as:

The manifestations of your condition which you tell your physician, that are not verifiable using tests, procedures or clinical examinations standardly accepted in the practice of medicine. Examples of self-

reported symptoms include, but are not limited to headaches, pain, fatigue, stiffness, soreness, ringing in ears, dizziness, numbness and loss of energy.

B. First Review of Disability Claim Ovist submitted her long-term disability claim form in June 2011, listing CFS as the illness causing her disability. In an accompanying Attending Physician Statement, Ovist’s primary care physician, Dr. John Hudson, confirmed that CFS is the “primary diagnosis preventing the patient from working[.]” He also listed secondary diagnoses of Parvovirus3 and fibromyalgia,4 along with

3 According to the Mayo Clinic, “most people with parvovirus infection have no signs or symptoms. When symptoms do appear, they vary greatly depending on how old you are when you get the disease.” For adults, “[t]he most noticeable symptom of parvovirus infection . . . is joint soreness lasting days to weeks.” Mayo Clinic, Parvovirus infection, at https://www.mayoclinic.org/diseasesconditions /parvovirus-infection/symptoms-causes/syc-20376085.

4 According to the National Institute of Health, fibromyalgia is “a long-lasting or chronic disorder that causes muscle pain and fatigue,” and “[t]he symptoms of fibromyalgia are pain and tenderness throughout your body.” National Institute of Health, Fibromyalgia, at https://www.niams.nih.gov/health-topics/fibromyalgia.

8

the following symptoms: severe chronic fatigue, severe muscle and joint pain, dizziness, insomnia, depression, and “unclear mental clarity, or brain fog[.]”

The following month, on July 15, 2011, Dr. Hudson completed a detailed assessment of Ovist, in which he repeated many of the same diagnoses and symptoms, along with acute and chronic sinusitis, opioid withdrawal, chronic bronchitis, and various viral infections. Dr. Hudson indicated that he referred Ovist to Dr. Nancy Klimas, a chronic fatigue specialist, for a second opinion on CFS and fibromyalgia.

In an Initial Progress Note dated July 29, 2011, Dr.

Klimas wrote that Ovist is “very disabled . . . with [symptoms] consistent with CFS.” Dr. Klimas prescribed a sleep study to determine the presence of apnea and ordered lab tests for “immune activation, function, [and] cytokines.”

In August 2011, Unum sent letters to three of Ovist’s physicians—Dr. Kent Hoffman, who treated Ovist for pain and opioid dependence; Dr. Klimas; and Dr. Cory Baill, a gynecologist who began treating Ovist as early as 2002—to ask whether they restricted Ovist from completing her work, which, according to Unum’s Vocational Rehabilitation Consultant, involved light physical demands. Specifically, Unum asked the doctors if they “restrict Ms. Ovist from performing full time work [that includes] sitting up to frequently and standing/walking up to frequently;

9

exerting 20 [pounds] of force occasionally or ten [pounds] of force frequently, and/or a negligible amount of force constantly to lift, carry, push, pull, or otherwise move objects, including the human body[.]” Each doctor confirmed that Ovist could not work under the physical demands listed.

On September 1, 2011, Dr. Freeman Broadwell, a medical consultant for Unum, reviewed Ovist’s file and found that she consistently reported pain and fatigue and received treatment for these symptoms from multiple providers over several years. Dr. Broadwell did not find, however, that the “existence, intensity, frequency, and duration of chronic pain and fatigue [were] consistent with the clinical examination / diagnostic findings.” Nor did he find from Ovist’s work-up that chronic infection or any other physical condition could explain a level of impairment that rendered Ovist unable to work. Dr. Broadwell nonetheless concluded that “[d]ue to the consistency of [Ovist’s] reports” of chronic pain and fatigue “corroborated by her providers and absence of evidence to the contrary, the [restrictions and limitations] of no work are supported.”

In a letter dated September 13, 2011, Unum approved Ovist’s claim for long-term disability benefits “due to the symptoms related to [her] medical condition of [CFS].” Unum noted that her file shows that she had “consistently reported and been treated for self-reported complaints of chronic fatigue and

10

chronic diffuse pain,” and added that the Plan “limit[s] [Ovist’s] benefits to 24 months due to [her] medical condition of [CFS].” Consequently, Unum provided an end date of June 29, 2013 for Ovist’s benefit payments. C. Second Review of Disability Claim During 2012 and 2013, Ovist’s physicians and Unum’s Personal Visit Consultant documented the persistence of Ovist’s conditions and symptoms. In a status update dated January 6, 2012, Dr. Klimas observed that Ovist continued to have physical and cognitive impairments, and that her pain was a “very serious issue . . . requiring increasing levels of pain med[ications].” She provided a primary diagnosis of immune deficiency based on “laboratory testing demonstrat[ing] poor cellular function . . . , proinflammatory cytokine expression, [and] serology consistent with viral reactivation.” Dr. Klimas also listed secondary diagnoses of fatigue and fibromyalgia. On May 4, 2012, Dr. Klimas reported no improvement in Ovist’s functional capacity and advised that Ovist have “unrestricted access to rest” and “avoid exposure to community acquired infections.”

On May 14, 2012, Unum’s Personal Visit Consultant, Mark Cox, conducted a field visit with Ovist. Cox noted in his report that Ovist remained seated throughout the visit, and he “did not observe the insured display any physical signs of pain / discomfort while she was seated . . . .” But Cox also found that Ovist

11

“appeared tired and fatigued throughout the entire field visit,” “appeared to have some difficulty staying focused on the topics being discussed,” and occasionally slurred her speech. A couple months later, in response to a request for information from Unum, Dr. Hoffman opined that Ovist’s diagnoses were chronic pain syndrome, fibromyalgia, and opioid dependence. He maintained that Ovist was unable to work.

Ovist’s behavioral health issues also begin to appear in the record during this time period. Around August 2012, Ovist began seeing a clinical psychologist, Dr. Catherine Segota. That same month, Ovist submitted a form to the Florida Department of Health, in which she reported experiencing daily panic and/or anxiety attacks that lasted from 15 minutes to hours. Roughly one month later, on September 14, 2012, Ovist was awarded social security disability insurance (“SSDI”) benefits retroactive to April 2012. The Social Security Administration determined that Ovist was disabled as of October 26, 2011, with diagnoses of “other unspecified arthropathy”5 and “anxiety related disorder.”

On July 19, 2013, Dr. Tony Smith, a medical consultant for Unum, reviewed Ovist’s file and wrote that multiple specialists

According to Johns Hopkins Medicine, arthropathy is “a 5

joint disease, of which arthritis is a type,” and its symptoms include “joint swelling, stiffness[,] [and] reduced range of motion.” Johns Hopkins Medicine, Arthropathy, at https://www.hopkinsmedicine.org/health/conditions-anddiseases /arthropathy.

12

diagnosed Ovist with CFS, fibromyalgia, and chronic pain syndrome, though “[n]o specific etiology for the opined diagnoses has been established to date.” Dr. Smith determined that “the medical data documenting Ovist’s consistent complaints,” Ovist’s level of treatment, Dr. Hoffman’s report of narcotic dependence, Cox’s description of Ovist’s lack of focus and slurred speech, and the 2012 SSDI award “reasonably support[] a finding that Ms. Ovist may have difficulty working in a sedentary or light capacity on a consistent basis.” He added that Ovist’s “improvement is reasonable and expected,” and he suggested a follow-up with Ovist in 8-12 months.

Unum’s claim log indicates that, on July 22, 2013, Unum notified Ovist over the phone that her claim would undergo further review and her benefits would continue, though the initial 24- month benefit period had passed in June. Unum also conveyed to Ovist that it still found that she was unable to work, but improvement was reasonable and expected. The following day, Unum noted in the claim log that Ovist’s file was transferred to the Special Benefits Unit CORE section, and the next steps were an annual telephone call with Ovist and a medical update in 12 months. One year later, in June 2014, Ovist’s file was “reassign[ed]” from the CORE to the “Comp.” section, “to hold” the annual telephone call with Ovist, “request records,” review her eligibility, and “discuss[] [] the SR[S] limitation.” Ovist asserts, based on an

13

affidavit from an employee of Unum, that the CORE section manages claims pertaining to conditions that are unlikely to improve, and the “Comp.” section “manages claims with great scrutiny.” D. Third Review of Disability Claim In August and September 2014, Unum reached out to Ovist’s treatment providers, Dr. Deborah Dube, Dr. Hoffman, Dr. Segota, and Dr. Klimas, requesting updates on Ovist’s conditions and functional limitations. Dr. Dube, Dr. Hoffman, and Dr. Segota responded to Unum in November 2014. Dr. Klimas’s response reached Unum after it completed its third review of Ovist’s claim, and Unum considered it in its next review, as discussed further below.

Dr. Dube, who treated Ovist for fatigue, generalized pain, and chronic sinus infections, stated that she was unable to provide information on Ovist’s functional capacity because her office did not perform functional capacity testing. Dr. Hoffman maintained that Ovist’s primary diagnoses were fibromyalgia and chronic pain syndrome, and her secondary diagnosis was opioid dependence. He explained that Ovist was still unable to work because her “mental and physical capacities are very time limited and easily exhausted.” Finally, Dr. Segota noted that she met with Ovist on fifteen occasions in the previous two years, and that she diagnosed Ovist with major depressive disorder, generalized anxiety disorder, panic disorder, posttraumatic stress disorder, and “unspecified neurocognitive” issues. Dr. Segota

14

listed sadness, rumination, hopelessness, sleep disturbance, and fatigue as “data which support[ed]” the diagnoses, and she concluded that Ovist was unable to work for an unknown period of time due to the severity and chronic nature of her symptoms.

Shortly thereafter, on December 14, 2014, Unum consulted with its legal counsel regarding any “concerns with possibly applying [the SRS Limitation] to dx [a diagnosis] which may include . . . fibromyalgia[.]” Counsel advised Unum that “[t]here is no binding precedent in the Eleventh Circuit that would preclude application of the [SRS] limitation when fibromyalgia has been diagnosed.”

On December 30, 2014, Nurse Sarah Curran, a medical consultant for Unum, reviewed Ovist’s file and agreed with Unum’s reviewers who determined in “2011 and again in 2013 . . . that there was no physical basis to explain the etiology of [Ovist’s] reported complaints of extreme fatigue and pain[,] and [her restrictions and limitations] were supported based on the consistency of [her] complaints.” Nurse Curran also noted that Ovist’s physical exams “remain unremarkable and there has been no diagnostic testing performed to explain the etiology of [Ovist’s] complaints.” Nurse Curran concluded that Ovist was unable to work due to her pain and fatigue. She added that, based on Dr. Segota’s report, Ovist was also functionally impaired “from a [behavioral health] perspective” beginning in July 2012.

15

A couple weeks later, Dr. Bryan Hauser, a medical consultant for Unum, reviewed Ovist’s file and consulted Dr. Hoffman. Based on that review, he agreed that Ovist was unable to work due to the same “non-verifiable medical conditions.”

In a letter dated February 17, 2015, Unum notified Ovist that, after reviewing updated information in her file, its determination from 2011 remained in place—Ovist was unable to work due to CFS and fibromyalgia, and the SRS limitation applied to her claim. Unum explained that Ovist’s diagnoses “are considered self- reported,” as they “cannot be verified or confirmed by physical examination findings that are not dependent on [Ovist’s] report and cannot be verified by diagnostic test findings.” Unum also repeated its prior finding that Ovist’s “physical examinations remain unremarkable,” and “[t]here had been no diagnostic testing performed to explain the etiology of [Ovist’s] complaints.” Finally, Unum noted that Ovist was functionally impaired from a behavioral health perspective since July 2012, but since Ovist’s maximum benefit period had already been exhausted in June 2013 due to her conditions of CFS and fibromyalgia, Unum could “no longer consider conditions or symptoms based on either behavioral health or self-reported symptoms.” Ovist’s final benefit payment was scheduled for the following day, February 18, 2015.

16

E. Fourth Review of Disability Claim Dr. Klimas responded to Ovist’s 2014 request for updates in a letter dated February 24, 2015; this letter initiated a fourth round of review of Ovist’s claim. In her letter, Dr. Klimas asserted that Ovist was unable to work, and her symptoms included fatigue, pain, cognitive dysfunction, sleep disturbance, and headaches. Dr. Klimas listed diagnoses of CFS, Immune Deficiency Syndrome, Myalgia, Orthostatic hypotension, and sleep disturbance, and she noted that water and mold damage in Ovist’s home “probably contributed to the worsening of her health.” She added that her findings were “confirmed by physical examination, medical history, and laboratory data . . . show[ing] impaired immune function . . . and latent virus reactivations.”

According to the claim log, Unum tried to contact Dr.

Klimas multiple times in March 2015 “to clarify the etiologies of [Ovist’s] impairing symptoms and the basis for the diagnosis of [the] conditions” Klimas listed. Dr. Klimas’s response, dated June 12, 2015, was considered by Unum’s Appeals Committee.6 On April 14, 2015, Dr. Hauser reviewed Ovist’s file again. He disagreed with Dr. Klimas’s “assertion that [Ovist’s] impairing symptoms are attributable to (either directly or

6Unum added Dr. Klimas’s June 2015 statement to Ovist’s claim file on July 17, 2015. The content and opinions that Dr. Klimas’s expressed in her June 2015 statement were similar to those expressed in her February 2015 letter to Unum.

17

indirectly) viral infection, immune deficiency, or any other verifiable (through physical examination or diagnostic testing) medical condition.” He also noted that Dr. Klimas’s tests for viral infection only provided “evidence of past (and not acute or ongoing)” infection. Overall, Dr. Hauser came to a similar conclusion regarding Ovist’s claim as he did the previous year— that Ovist was unable to work due to fatigue, pain, and depression, and these symptoms were attributed to diagnoses of CFS, fibromyalgia, and depression, which were not objectively verified. With respect to fibromyalgia, Dr. Hauser stated that Ovist’s diagnosis “is supported by the history of chronic generalized . . . pain and the finding of at least 11 [of] 18 fibromyalgia tender points.” He elaborated on the subjective nature of the tender point examination, noting that, during the examination, “pain elicited by the application of pressure by the examiner (i.e., tenderness) is experienced and reported by the examinee; its existence cannot be verified.”

A couple days later, Unum medical consultant Dr. James Bress completed an independent analysis of Ovist’s file. Dr. Bress agreed that Ovist was unable to work, and he identified that Dr. Klimas was the only treatment provider to state that Ovist’s functional loss was “[due to] verifiable medical problems,” namely immune impairment and latent virus reactivations. Like Dr. Hauser, Dr. Bress disagreed with this aspect of Dr. Klimas’s opinion. He

18

determined that Dr. Klimas’s finding was not supported by any evidence, and that the tests Dr. Klimas administered to assess immune function and viruses were not “standard medical testing” and/or had no confirmed association with Ovist’s symptoms.

Unum sent Ovist a letter dated April 17, 2015, stating that it maintained its decision to terminate her benefits under the SRS limitation as of February 18, 2015. Unum explained that Ovist was functionally impaired based on the non-verifiable conditions of CFS, fibromyalgia, and depression. Unum added that the diagnoses were “supported in part” by Ovist’s repeated reports of pain and fatigue, as well as “the finding of at least 11/18 fibromyalgia tender points.”

On June 15, 2015, Ovist sent Unum a fungal report and a November 2013 mold analysis of her home. Dr. Hauser and Dr. Bress each reviewed the documentation and concluded that it did not change their prior conclusions, as there is no proven association between mold contamination or elevated fungal spores and Ovist’s symptoms or diagnoses. Accordingly, on July 14, 2015, Unum informed Ovist that its benefits determination and supporting rationale remained the same. F. Ovist’s Administrative Appeal Ovist appealed Unum’s decision in a letter dated July 4, 2015. Unum defined the “medical question” on appeal as whether “the records support functional loss and/or [restrictions and

19

limitations] due to physical medical condition/symptoms that are verifiable using tests, procedures or clinical examinations as of [February 17, 2015] forward[.]”

Ovist submitted additional information for review on appeal—office notes from a gastrointestinal specialist from May 2013; office notes from Dr. Hoffman from January to July 2015; records and diagnostic test results from primary care physician Dr. Dube and cardiologist Dr. Potts from January and February 2015; and records and lab reports from Dr. Klimas from February and March 2015. Ovist also provided Unum with the results of a Cardiopulmonary Exercise Test (“CPET”) conducted by exercise physiologist Jeffrey Cournoyer on September 24-25, 2015. The CPET is designed “to determine functional capacity and assess the recovery response to a standardized physical stressor.”7 In his report, Cournoyer explained that the CPET consisted of two identical tests that were administered on consecutive days, in order to “establish changes in work function capacity.” Cournoyer noted that Ovist “demonstrated maximal

7 According to Massachusetts General Hospital, “[t]he primary purpose of . . . [CPET] is to carefully assess how your lungs, heart, blood vessels and muscles perform during an exercise challenge. . . . CPET is used to define how conditions that affect heart, lung, blood vessel or muscle function contribute to exercise intolerance. Massachusetts General Hospital, Cardiopulmonary Exercise Testing Lab, at https://www.massgeneral.org/medicine/pu lmonary/treatments-and-services/cardiopulmonary-exercisetesting .

20

effort in some, but not all of the testing measures,” but he also found “both tests to be of maximal nature.”

As for the test results, Cournoyer observed that Ovist’s testing measures varied by roughly 6% between the two days, suggesting that the results were not abnormal on either day. He also found that Ovist’s metabolic responses indicated “a higher probability of running out of energy” on the second day; her “[b]reathing values showed more strain” on the second day; and her “wattage at maximal effort” on both days was “pathologically low.” Seemingly consistent with this, Cournoyer observed that Ovist left the office fatigued on the first day of testing, but after the second day of testing, Ovist’s “posture and walking gait suggested a severely weakened state, and [Cournoyer] was not comfortable with allowing her to leave in that condition.” The CPET also included an element of cognitive testing. Cournoyer found that Ovist “showed the most drastic changes” across the two days in her “immediate and delayed recall of simple information” and in her concentration.

Unum’s Angela Malan-Elzawahry reviewed Ovist’s file on January 5, 2016. Malan-Elzawahry detailed the medical conditions with which Ovist had been diagnosed or for which she had been evaluated and/or treated, including infections, sinusitis, CFS, fibromyalgia, wrist pain, ankle sprain, shingles, thyroid goiter, chronic pain syndrome, opioid dependence, depression,

21

cardiopulmonary conditions, left knee pain and related joint issues, and mold exposure. She also described the results of various laboratory or diagnostic tests administered to Ovist, which she found to be unremarkable. According to Malan-Elzawahry, tests for viral illnesses, which were repeatedly administered to Ovist beginning in 2006, had not provided evidence of current infection; a 2012 sleep study did not “identify sleep apnea at a severity to be expected to cause the level of fatigue that [Ovist] has reported”; and a 2015 cardiac evaluation did not show “restrictions and limitations . . . related to a cardiopulmonary condition[.]”

Malan-Elzawahry concluded that Ovist consistently reported worsening pain and fatigue, but the “conditions identified by testing, such as thyroid goiter, a joint effusion, and anterior cardiac wall soft tissue attenuation [were] not correlating with symptoms that are at a severity to limit [Ovist’s] function,” nor were they “expected to generate the fatigue and generalized pain symptoms reported by [Ovist].” Malan-Elzawahry deferred to Unum’s medical consultant, Dr. Scott Norris, for further, independent analysis of the appeal.

On January 21, 2016, Dr. Norris concluded that Ovist’s functional limitations were “based on [Ovist’s] reported [symptoms] and [were] not consistent with the minimal and nonspecific findings documented on examinations or the diagnostic

22

testing/imaging studies included in the file records.” He elaborated on this conclusion, noting, for example, that “multiple laboratory tests were negative for infections, metabolic, immunologic, hematologic, inflammatory, or other verifiable causes of [Ovist’s] reported fatigue”; Ovist’s “[r]ecords [did] not catalogue examinations or testing consistent with cognitive impairment related to physical conditions”; Ovist’s February 2015 cardiac tests were normal; and “although [Ovist] reported mold exposure in her home, the records do not reveal evidence of impairment related to mold.”

Finally, Dr. Norris expressed reservations about Ovist’s CPET results. He wrote that the September 2015 CPET was not “time- relevant regarding [Ovist’s] functional capacity as of February 2015,” the date when Unum closed Ovist’s claim. He also indicated that Ovist did not “exhibit maximal effort” at certain points of the CPET, meaning that “a true . . . maximal aerobic capacity[] was not measured.”

Unum denied Ovist’s appeal in a letter dated January 27, 2016, concluding that the SRS limitation applied to Ovist’s claim. In the letter, Unum repeated Dr. Norris’s conclusion that her reported impairing symptoms and her functional loss were “inconsistent with and/or not supported” by clinical examinations, diagnostic findings, or other objectively verifiable evidence in her file. Unum listed examples: there was no “correlation between

23

. . . [the] environmental mold in [Ovist’s] home [and] the severe illness and functional impairment being reported”; Ovist’s infectious disease titers8 were “consistent with a past history of infection with no evidence of recurrent infection”; Ovist’s temperature readings were normal; “the cardiology work up was normal with no findings . . . consistent with [orthostatic hypotension]”; “[t]he 2011 sleep study was negative for obstructive sleep apnea”; and “[t]he records do not document cognitive impairment on exam,” though Ovist’s “significant anxiety and depression can increase her perception of poor concentration and/or cognitive dysfunction.”

Unum further noted that Ovist’s CPET was not time-

relevant, and “Ovist did not exhibit full effort on all tests.” Unum asserted that “[r]egardless of [Cournoyer’s] findings” from the CPET, Ovist’s “functional limitation is based primarily on self-reported pain and fatigue.”

Unum then concluded:

We do not refute [] your client’s perceived physical and/or functional limitations.

However, we determined that any and all loss of function is based on disability due to mental illness and based primarily on self-

reported symptoms. For all the reasons

8 “Antibody titer is a laboratory test that measures the level of antibodies in a blood sample,” and it is used to identify, among other things, if a patient has “had a recent or past infection . . . .” University of California San Francisco Health, Antibody titer blood test, at https://www.ucsfhealth.org /medical-tests/antibody-titer-blood-test.

24

stated, we determined that no further benefits are payable under the policy’s [SRS limitation].

G. Procedural History On August 2, 2017, following the adverse determination on her administrative appeal, Ovist initiated an ERISA action in the U.S. District Court for the District of Massachusetts, alleging that Unum unlawfully terminated her benefits. On August 30, 2019, the parties filed cross-motions for summary judgment. A magistrate judge issued a Report and Recommendation on February 24, 2020, recommending that the district court grant Unum’s motion and deny Ovist’s motion.

In the Report and Recommendation, the magistrate judge first determined that Ovist, as the claimant, bears the burden to prove that the SRS limitation does not apply to her claim. Then, he found that Unum acted reasonably, under the terms of the Plan and this circuit’s precedent, when it terminated Ovist’s benefits under the SRS limitation based on a finding that Ovist’s impairing symptoms, such as pain and fatigue, were based on her self- reporting, as opposed to objectively verifiable diagnostic or other tests. The magistrate judge relied, in particular, on case law from this court establishing that it is unreasonable for an insurer to require a claimant to provide objective evidence of diagnoses that do not lend themselves to objective verification,

25

but an insurer can reasonably require objective evidence of a claimant’s resulting functional limitations.

In March 2020, Ovist filed an objection to the Report and Recommendation, and Unum filed a reply. The district court adopted the Report and Recommendation in full on March 27, 2020. This timely appeal followed. We affirm the district court’s order. II. Standard of Review We review the district court’s grant of summary judgment de novo. Arruda v. Zurich Am. Ins. Co., 951 F.3d 12, 21 (1st Cir. 2020). When the underlying plan affords the insurer discretion to determine eligibility for benefits, “[a] federal court reviews an insurer’s termination decision ‘under a deferential arbitrary and capricious standard . . . .” Cook v. Liberty Life Assur. Co. of Bos., 320 F.3d 11, 18 (1st Cir. 2003) (quoting Pari-Fasano v. ITT Hartford Life & Acc. Ins. Co., 230 F.3d 415, 418 (1st Cir. 2000)).

Here, the parties do not dispute that the Plan grants discretionary authority to Unum. The Plan expressly “delegates to Unum and its affiliate Unum Group discretionary authority to make benefit determinations under the Plan,” including “determining eligibility for benefits and the amount of any benefits, resolving factual disputes, and interpreting and enforcing the provisions of the Plan.” Accordingly, we will review Unum’s benefit decision under the arbitrary and capricious standard, as the district court did.

26

Under this standard, “we need not decide what is the best reading of the words in the insurance policy.” Stamp v. Metro. Life Ins. Co., 531 F.3d 84, 94 (1st Cir. 2008). We will uphold the plan administrator’s benefit decision if it “is reasonable and supported by substantial evidence on the record as a whole.”9 McDonough v. Aetna Life Ins. Co., 783 F.3d 374, 379 (1st Cir. 2015) (citing Colby v. Union Sec. Ins. Co. & Mgmt. Co. for Merrimack Anesthesia Assocs. LTD Plan, 705 F.3d 58, 61 (1st Cir.2013)). “‘Substantial evidence’ is ‘evidence reasonably sufficient to support a conclusion.’” Arruda, 951 F.3d at 21 (quoting Doyle v. Paul Revere Life Ins. Co., 144 F.3d 181, 184 (1st Cir. 1998)). Importantly, a conclusion can still be supported by substantial evidence if contrary evidence exists. See Boardman v. Prudential Ins. Co. of Am., 337 F.3d 9, 15 (1st Cir. 2003). In short, we must uphold Unum’s determination unless it was “unreasonable in light of the information available” to Unum. Pari-Fasano, 230 F.3d at 419.

9 In ERISA cases in which the plan administrator has discretion to make benefits determinations, we have variously described the standard of review as review for “abuse of discretion,” “arbitrariness and capriciousness,” and “substantial evidence.” These terms are interchangeable in this context. See, e.g., McDonough, 783 F.3d at 379 (“A court that undertakes abuse of discretion review in an ERISA case must determine whether the claims administrator’s decision is arbitrary and capricious or, looked at from another angle, whether that decision is reasonable and supported by substantial evidence on the record as a whole.”).

27

III. Analysis Unum approved Ovist’s long-term disability claim in 2011 due to her symptoms related to CFS. In its initial determination letter, Unum notified Ovist that her benefit payments would cease after 24 months under the SRS limitation, which applies, in pertinent part, to disabilities “based primarily on self-reported symptoms,” including “headaches, pain, fatigue, . . . and loss of energy.” Roughly 43 months later, in its final adverse benefit determination letter, Unum maintained that the SRS limitation applied to her claim because Ovist’s functional limitations were supported by her reports of pain and fatigue, rather than clinical examinations, diagnostic findings, or other objectively verifiable evidence.

Ovist primarily challenges Unum’s interpretation and application of the SRS limitation as follows. She argues that it was unreasonable for Unum to require objective evidence of her functional loss after concluding that she was unable to work. In the alternative, she claims that she did provide the requested evidence—her CPET results—and Unum rejected the results “on the flimsiest of grounds.” Next, Ovist contends that the Seventh Circuit Court of Appeals’ interpretation of a provision almost identical to the SRS limitation, in Weitzenkamp, should control,

28

and the SRS limitation does not apply to her claim under the holding in Weitzenkamp. We consider each challenge in turn.10 A. Unum’s Objective Evidence Requirement is Reasonable To begin, Ovist contends that it was unreasonable for Unum to require objective proof of her functional limitations after conclusively determining that she was unable to work. This argument fails under settled precedent within this circuit, which we revisit below.

We have repeatedly held that it is unreasonable for an insurer to require objective evidence in support of diagnoses, like fibromyalgia and CFS, which are not subject to objective verification. See Denmark v. Liberty Life Assur. Co. Of Bos., 481 F.3d 16, 37 (1st Cir. 2007) (explaining that it is unreasonable for an insurer “to require objective evidence to support a diagnosis of a condition that is not subject to verification through laboratory testing[,]” and identifying fibromyalgia as one such condition), vacated on other grounds, 566 F.3d 1 (1st Cir. 2009); Cook, 320 F.3d at 21-22 (finding it unreasonable for the

Ovist also challenges the district court’s holding that she 10

bears the burden to prove that the SRS limitation does not apply to her claim, arguing instead that the burden of proof lies with Unum. We need not decide this issue because it will not affect the outcome of this case. Under the applicable standard of review, we must determine whether Unum’s decision is “reasonable and supported by substantial evidence.” McDonough, 783 F.3d at 379 (citing Colby, 705 F.3d at 61). As discussed below, Unum’s decision passes muster under this standard, regardless of where we place the burden of proof.

29

insurer to expect the claimant to provide “convincing ‘clinical objective’ evidence that she was suffering from CFS” because there are no accepted laboratory tests associated with the condition).

In Boardman v. Prudential Insurance Company of America, however, we drew a distinction between requiring objective evidence of conditions that do not lend themselves to objective verification and requiring objective evidence of the functional limitations resulting from a claimant’s conditions. 337 F.3d at 16-17 & n.5. We held that the latter is permissible.

The claimant in Boardman presented varying diagnoses of conditions that are associated with pain and fatigue, including CFS and myalgias. Id. at 12-14 & n.4. She was granted long-term disability benefits upon showing that “due to her illness, she was unable to perform the duties of her job . . . .” Id. at 11. In order to remain eligible for benefits after the first 24 months of payments, the claimant needed to show “that she was disabled from duties of ‘any job for which [she was] reasonably fitted . . . .’” Id. at 13. Though the plan administrator, Prudential, “was willing to accept that [the claimant] suffered from the illnesses she reported to her doctors[,]” it terminated the claimant’s benefits, in pertinent part, because her file did not indicate “any limitations or restrictions, based on objective findings, that would preclude [her] from performing any occupation for which she is suited.” Id. at 15, 16 n.5. We affirmed the district court’s

30

grant of summary judgment to Prudential, reasoning that, “while the diagnoses of [CFS] and fibromyalgia may not lend themselves to objective clinical findings, the physical limitations imposed by the symptoms of such illnesses lend themselves to objective analysis.” Id. at 16 n.5, 17.

We have since repeatedly invoked this principle and Boardman’s diagnosis-disabling symptom distinction when reviewing plan administrators’ benefit determinations. See Cusson v. Liberty Life Assur. Co. of Bos., 592 F.3d 215, 227 (1st Cir. 2010) (observing that the plan administrator did not “question the diagnosis of fibromyalgia,” but “instead . . . questioned the effect of the disease on [the claimant’s] ability to work . . . . Because it is permissible to require documented, objective evidence of disability, it was not inappropriate for [the plan administrator] to rely on the lack of such documented evidence, or on the footage that contradicted [the claimant’s] reports of limitations, in making their recommendations” that the plaintiff was able to work and thus not disabled under the terms of the plan), abrogated on other grounds by Montanile v. Bd. of Trustees of Nat. Elevator Indus. Health Benefit Plan, 577 U.S. 136 (2016); Denmark, 481 F.3d at 37 (holding that the plan administrator acted “within the parameters defined in Boardman” when it required the claimant, who was diagnosed with fibromyalgia, to “provide objective evidence of functional limitations or restrictions that

31

would prevent her from working” in order to qualify for long-term disability benefits). Here, we endorse our holding in Boardman once again and conclude that it was reasonable for Unum to require objective proof of Ovist’s functional loss.11 Ovist attempts to escape this outcome by emphasizing that, prior to imposing this evidentiary requirement, Unum repeatedly found that Ovist was unable to work and did not harbor any suspicions that Ovist was falsifying her functional limitations. This argument is misplaced, as it constitutes an objection to the design of the Plan—over which the ”employer ha[s] large leeway”—rather than a viable challenge to the reasonableness

11We do acknowledge, as the district court did, that Unum’s claim log and correspondence with Ovist also intermittently focused on the absence of objective evidence supporting her diagnoses. For example, in February and April 2015 letters notifying Ovist that her benefits would be terminated under the SRS limitation, Unum stated that her functional limitations were associated with fibromyalgia and CFS, two conditions that “were considered self-reported” since they could not be confirmed by “clinical signs” or “diagnostic test findings.” Relatedly, when Unum sought guidance from legal counsel on Ovist’s claim, its question turned on Ovist’s diagnosis, as opposed to her functional limitations. Specifically, Unum asked counsel if it was legally permissible to apply the SRS limitation to a diagnosis of fibromyalgia. While these examples show an inconsistency in Unum’s handling of Ovist’s claim, they are peripheral to our analysis on appeal. This court’s analysis centers on Unum’s final adverse benefit determination. See Terry v. Bayer Corp., 145 F.3d 28, 35 (1st Cir. 1998) (stating that the court “must focus, as in the usual case, on the determination of the final decision-maker” when reviewing an appeal of the plan administrator’s decision to terminate benefits). And in this case, Unum’s medical question on appeal, subsequent internal analysis, and denial of Unum’s appeal all primarily centered on the insufficient objective evidence of Ovist’s functional limitations.

32

of Unum’s benefit determination. Black & Decker Disability Plan v. Nord, 538 U.S. 822, 833 (2003).

Indeed, under the terms of the Plan, individuals are considered “disabled” and eligible for benefits if they are unable to complete the “material and substantial duties” of their jobs due to their illnesses. The SRS limitation confines the benefit period to 24 months for the same, disabled individuals if they have, in pertinent part, “disabilities based primarily on self- reported symptoms.” Unum simply followed the Plan’s blueprint, then, by first determining that Ovist was unable to work (and thereby granting her benefits), and then reasonably requiring objective proof of her functional loss in order to determine if her disabilities were “based primarily on self-reported symptoms,” and thus subject to the associated benefit limitation. It follows that Ovist takes issue with her employer’s decision to establish a different evidentiary hurdle (centered on objective evidence) for claimants seeking to obtain benefits over the long term, as opposed to claimants seeking benefits for only 24 months or less. This objection to the Plan’s design does not alter our analysis under Boardman.

B. Unum’s Denial of Long-term Disability Benefits on the Information Before it was Reasonable

Ovist next argues that Unum reached its benefit determination by unreasonably rejecting her CPET results, which,

33

according to Ovist, provide objective proof of her functional limitations. Under the arbitrary and capricious standard, our task is not to re-weigh the evidence in the record. Instead, we must uphold the plan administrator’s decision if it “is reasonable and supported by substantial evidence on the record as a whole.” McDonough, 783 F.3d at 379 (citing Colby, 705 F.3d at 61). This also means, however, that “[p]lan administrators . . . may not arbitrarily refuse to credit a claimant’s reliable evidence[.]” Black & Decker Disability Plan, 538 U.S. at 834. Upon reviewing the record, we find that Unum’s conclusion that Ovist’s functional limitations were “inconsistent with and/or not supported based on clinical exam and/or diagnostic findings, procedures, and/or other clinical findings” is both supported by substantial evidence and consistent with a reasonable review of the record as a whole, including the CPET results.

To begin, Unum reviewed Ovist’s file on at least five separate occasions, allowing Ovist to supplement the file with medical providers’ opinions and other evidence in the interim. Unum also followed up with Ovist’s physicians in order to obtain updates on her conditions and the basis of their opinions. Further, Unum alerted Ovist to the absence of objective evidence of her functional limitations; for example, in a February 2015 adverse benefit determination letter, Unum stated that Ovist’s “physical examinations remain unremarkable,” and “[t]here had been

34

no diagnostic testing performed to explain the etiology of [Ovist’s] complaints.” Still, Unum’s medical consultants repeatedly found that there were no diagnostic findings or clinical examinations explaining Ovist’s consistent reports of pain and fatigue and her associated functional loss.

Dr. Broadwell was the first consultant to make this observation, in 2011. In 2013, Dr. Smith concluded that Ovist was unable to work based on her consistent complaints of pain and fatigue, her treatment history, Dr. Hoffman’s report of her narcotic dependence, observations from Unum’s field visit, and her 2012 SSDI award. Dr. Smith did not base his conclusion, then, on any clinical exams, diagnostic findings, or objectively verifiable physical exams evidencing Ovist’s functional limitations. Next, in 2014, Nurse Curran determined that Ovist’s functional limitations “were supported based on the consistency of [her] complaints,” but “there was no physical basis to explain the etiology of [those] complaints.”

Only one of Ovist’s doctors, Dr. Klimas, claimed that Ovist’s condition and/or symptoms were “confirmed by physical examination . . . and laboratory data” showing “impaired immune function . . . and latent virus reactivations.” Two of Unum’s medical consultants considered and rejected Dr. Klimas’s claim. Dr. Bress noted that Dr. Klimas’s tests for immune function and viruses were not “standard medical testing” and/or were unrelated

35

to Ovist’s symptoms. Dr. Hauser also asserted that Dr. Klimas’s tests for viral infection provided evidence of past, but not ongoing, infections.12 Dr. Hauser accordingly concluded that Ovist’s impairing symptoms were not “attributable to (either directly or indirectly) viral infection, immune deficiency, or any other verifiable (through physical examination or diagnostic testing) medical condition,” and Dr. Bress concurred.

Unum’s reviewers also considered other diagnostic and clinical exam findings and objectively verifiable evidence in Unum’s file and found that they were normal and/or could not explain the severity of Ovist’s disabling symptoms and her resulting functional limitations. Malan-Elzawahry found that Ovist’s 2012 sleep study and 2015 cardiac tests did not identify conditions or symptoms that could explain Ovist’s functional limitations. Dr. Norris determined that Ovist’s laboratory tests were negative for “infections, metabolic, hematologic, inflammatory, or other verifiable causes” of Ovist’s fatigue. He also noted that Ovist’s cardiac tests were normal, and no tests or examinations in Ovist’s file were consistent with “cognitive impairment related to physical conditions.” Finally, Dr. Bress, Malan-Elzawahry, and Dr. Norris reviewed Ovist’s fungal report and

Consistent with this finding, when reviewing Ovist’s 12

administrative appeal, Unum’s Malan-Elzawahry also found that Ovist’s tests for viral illnesses, dating back to 2006, did not provide evidence of current infections.

36

the mold analysis of her home and agreed that there was no known association between mold or fungus exposure and Ovist’s disabling symptoms.

Ovist does not dispute Unum’s assessment of the evidence above. Rather, Ovist points to her September 2015 CPET results as the singular source of objective proof of her functional limitations, and she argues that Unum rejected the CPET results on “the flimsiest of grounds.” We disagree.

Unum’s medical consultant, Dr. Norris, reviewed the CPET results and concluded that the test was not time-relevant and did not reflect Ovist’s maximal effort. Dr. Norris’s assessment of Ovist’s maximal effort was, at least in part, supported by Cournoyer’s own statement that Ovist “demonstrated maximal effort in some, but not all of the testing measures.” On the other side of the ledger, Cournoyer opined that the tests on both days were “of maximal nature.” His report lists respiratory, metabolic, and other markers showing that Ovist experienced fatigue and cognitive impairment following physical activity on both days. And Ovist argues that the seven-month gap between the February 2015 termination of her claim and the administration of the CPET is not meaningful because there is no evidence that her symptoms changed during that period.

37

When considering the CPET,13 Unum credited Dr. Norris’s opinion over Cournoyer’s findings. We have “treated a nonexamining physician’s review of a claimant’s file as reliable medical evidence on several occasions,” Gannon v. Metro. Life Ins. Co., 360 F.3d 211, 214 (1st Cir. 2004) (citing cases), and we find no basis on the record to conclude that Dr. Norris’s opinion is unreliable. Thus, Dr. Norris’s critique of the CPET provides a reasonable basis for Unum to find that the CPET results alone did not compensate for the considerable absence in the record of objective evidence of Ovist’s functional loss, and therefore to conclude that Ovist’s “functional limitation was based primarily on self-reported pain and fatigue.” This conclusion holds, even though the CPET arguably provides some objective proof of Ovist’s functional loss. See Boardman, 337 F.3d at 15 (“The existence of contrary evidence does not necessarily render [the claim administrator’s] decision arbitrary and capricious.”). Accordingly, we conclude that, notwithstanding Ovist’s CPET results, Unum’s decision to apply the SRS limitation to Ovist’s

13 We use the term ‘considering’ instead of ‘rejecting’

because it is not clear, based on Unum’s final adverse benefit determination letter, that Unum discredited the CPET results entirely. Rather, Unum noted that the results were not time- relevant, and that Ovist did not exhibit full effort on all of the tests. Then, Unum concluded that, “[r]egardless of [Cournoyer’s] findings[,]” Ovist’s “functional limitation is based primarily[,]” though not necessarily wholly, “on self-reported pain and fatigue.”

38

claim was reasonable and rests on substantial evidence in the record as a whole.14

C. The Seventh Circuit Weitzenkamp Test is in Conflict With First Circuit Law

Having found that Unum’s decision to terminate Ovist’s benefits was reasonable under this circuit’s precedent, we are unconvinced by Ovist’s argument that Unum’s application of the SRS limitation is improper because this court should adopt the standards under Weitzenkamp, a Seventh Circuit Court of Appeals decision. In Weitzenkamp, the court interpreted an SRS limitation almost identical to the one at bar. Under the SRS limitation in Weitzenkamp, benefits ceased after 24 months for individuals with “disabilities, due to sickness or injury, which are primarily based on self-reported symptoms, and disabilities due to mental illness, alcoholism or drug abuse.” 661 F.3d at 326-27. Self-reported symptoms are also defined identically in the plan in Weitzenkamp and the Plan in this case. Id. at 327.

14 Ovist also argues that Unum unreasonably applied the SRS limitation to her claim based on her mental illness. Since the SRS limitation, by its terms, applies to “any combination” of “disabilities due to mental illness and disabilities based primarily on self-reported symptoms,” Unum can reasonably apply the limitation when only one of the categories of disabilities is present. Since we already determined that Unum reasonably applied the limitation after completing more than 24 months of payments for the second category of “disabilities based primarily on self- reported symptoms,” we need not consider whether it was also reasonable to apply the limitation under the first category of disabilities “due to mental illness.”

39

The Seventh Circuit Court of Appeals concluded that the SRS limitation “applies to disabling illnesses or injuries that are diagnosed primarily based on self-reported symptoms rather than to all illnesses or injuries for which disabling symptoms are self-reported.” Id. at 330 (emphasis in original). The court noted that, if the SRS limitation applied to the latter category of conditions, it would limit benefits for individuals with most any disease with symptoms such as pain, weakness, and fatigue— symptoms which “are difficult if not impossible to verify using objective medical evidence.” Id. The court then determined that the plaintiff’s diagnosis of fibromyalgia was supported by objective evidence in the form of a positive tender point exam, and thus the SRS limitation did not apply. Id. at 331. Ovist posits that the same outcome should follow here, as she was positive for at least 11 of 18 tender points when examined, and this circuit has determined that the tender or trigger points provide “the clinical findings necessary for a diagnosis of fibromyalgia under established medical guidelines[.]” Johnson v. Astrue, 597 F.3d 409, 412 (1st Cir. 2009).

Respectfully, we decline to follow the reasoning and holding in Weitzenkamp, as they are in tension with this circuit’s long-held diagnosis-disabling symptom distinction as articulated in Boardman, and the underlying principle that “the physical limitations imposed by the symptoms of such illnesses [as CFS and

40

fibromyalgia],” including pain and fatigue, do “lend themselves to objective analysis.” Boardman, 337 F.3d at 16 n.4. Accordingly, even if we accept that Ovist tested positive for fibromyalgia based on tender points, we still conclude that it was reasonable for Unum to require that Ovist provide objective evidence of her functional limitations, and to apply the SRS limitation based on the relative absence of this evidence.

We also find Unum’s objective evidence requirement to be reasonable (contrary to the conclusion in Weitzenkamp) for at least two reasons. First, it merely calls for the claimant to establish a causal connection between his or her disability and his or her alleged functional limitation(s) before being awarded long-term disability benefits beyond 24 months. Far from being arbitrary or capricious, this type of inquiry into causation is often necessary for a claim administrator to ensure that benefits are paid as intended by the operative policy. See, e.g., Arruda, 951 F.3d at 21-22 (finding that the claimant was not eligible for death benefits under an insurance policy issued for accidental death or injury, since substantial evidence in the record supported the conclusion that the death was caused or contributed to by the decedent’s pre-existing health conditions). In this case, the causal connection must be established to confirm that Ovist is unable to work due to her recognized, diagnosed medical conditions, as opposed to her unverifiable perceptions.

41

Unum’s objective evidence requirement is also reasonable because it furthers the purpose of the SRS limitation, as defined by Unum. Under the terms of the Plan, Unum “ha[s] the discretionary authority to construe the [P]lan”; thus, Unum also “ha[s] the discretion to determine the intended meaning of the [P]lan’s terms.” Stamp, 531 F.3d at 93-94 (internal quotation omitted). And according to Unum, the purpose of the SRS limitation is “to address conditions” that manifest themselves in a manner that renders the resulting functional limitations “inherently difficult to determine.” It should not be deemed an abuse of discretion, then, for Unum to further this goal by requiring objective evidence of Ovist’s functional limitations, as such evidence is a more reliable indicator of the severity of Ovist’s limitations than her self-reporting of pain and fatigue. Unum’s requirement is particularly reasonable here, where Ovist had the opportunity to take the CPET, a test that can provide objective evidence of her functional limitations, but she failed to do so in a timely manner. Ultimately, we do not adopt Weitzenkamp, and instead adhere to this circuit’s law and conclude that Unum’s interpretation of the SRS limitation is not arbitrary or capricious. IV. Conclusion For the reasons stated above, the judgment of the district court is AFFIRMED.