Cusson v. Liberty Life Assur. Co. of Boston

United States Court of Appeals For the First Circuit No. 08-2381 DONNA CUSSON, Plaintiff, Appellant, v. LIBERTY LIFE ASSURANCE COMPANY OF BOSTON; and THE FLEETBOSTON FINANCIAL CORPORATION LONG-TERM DISABILITY PLAN, Defendants, Appellees. APPEAL FROM THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF MASSACHUSETTS [Hon. George A. O'Toole, U.S. District Judge] Before Torruella, Selya, and Howard, Circuit Judges. Mala M. Rafik, with whom Alyssa A. Yenikomshian, S. Stephen Rosenfeld, and Rosenfeld & Rafik, P.C., was on brief for appellant. Samuel Berk and Margaret Arenas Malek, on brief for amicus curiae Massachusetts Employment Lawyers Association, in support of reversal. George M. Thompson and Sager & Schaffer, LLP, on brief for amicus curiae Health Law Advocates, in support of reversal. Andrew C. Pickett, with whom Matthew D. Freeman and Jackson Lewis LLP, was on brief for appellees. January 14, 2010 TORRUELLA, Circuit Judge. In this appeal, appellant Donna Cusson ("Cusson") challenges the district court's decision to grant summary judgment to appellees Liberty Life Assurance Company of Boston ("Liberty") and the FleetBoston Financial Corporation Long-Term Disability Plan ("LTD Plan"). After Liberty terminated Cusson's long-term disability ("LTD") benefits, Cusson sued in the U.S. District Court of Massachusetts, arguing that the termination of her benefits was improper under the Employee Retirement and Income Security Act ("ERISA"), 29 U.S.C. §§ 1001-1431 (2006). The district court granted summary judgment in favor of the defendants, and Cusson now appeals. After careful consideration, we affirm the district court's judgment. I. Background1 A. The Long-Term Disability Plan Cusson was employed at FleetBoston as a Senior Facilities Planner. She was covered under both a short-term disability plan ("STD Plan") and the LTD Plan. Liberty is both the administrator of and payer of benefits under the LTD Plan. Under the terms of the insurance policy ("the Policy") for the LTD Plan, the definition of "Disability" or "Disabled" for LTD purposes varies over time with the duration of the disability. For the initial twenty-four-month period after a claimant switches into the LTD 1 For additional background, see the decision below. Cusson v. Liberty Life Assurance Co., No. 78-3386, 2008 U.S. Dist. LEXIS 77084 (D. Mass. Sep. 30, 2008). -2- Plan from the STD Plan, a person is considered "disabled" if she is unable to perform "all of the material and substantial duties of [her] own occupation." However, after twenty-four months, a person is considered "disabled" if she is "unable to perform, with reasonable continuity, all of the material and substantial duties of [her] own or any other occupation for which [s]he is or becomes reasonably fitted by training, education, experience, and physical or mental capacity." B. Cusson's Fibromyalgia Diagnosis and Initial LTD Benefits Award In March 2002, Cusson was diagnosed with breast cancer. She began to receive benefits under the STD Plan in April 2002 because of the pain and fatigue caused by her chemotherapy. Cusson was notified on October 9, 2002 that she qualified for LTD benefits under the Policy. Liberty also informed Cusson that she was required to apply for Social Security Disability Income ("SSDI") benefits. Cusson's chemotherapy ended in October 2002. However, she continued to feel symptoms similar to what she felt during her cancer treatment, including dizziness, fatigue, headaches, and widespread body pain. In February 2003, Cusson's rheumatologist, Dr. James Figueroa ("Figueroa"), diagnosed her with fibromyalgia. Fibromyalgia is characterized by "chronic and frequently difficult to manage pain in muscles and soft tissues surrounding joints." Rose v. Hartford Fin. Servs. Group, 268 F. App'x. 444, 446 n.4 (6th -3- Cir. 2008) (citing Taber's Cyclopedic Medical Dictionary 402 (19th ed. 2001)). A patient may be diagnosed with fibromyalgia if she has (1) a history of widespread pain for at least three months and (2) pain in at least eleven out of eighteen tender point sites upon digital palpation. Id. Symptoms of fibromyalgia can include stiffness, fatigue, and disturbed sleep. See Sarchet v. Chater, 78 F.3d 305, 306 (7th Cir. 1996). On March 28, 2003, at Liberty's request, Cusson completed an Activities Questionnaire, a document in which a claimant is asked to describe her day-to-day activities and her functional limitations. Cusson indicated that she was able to sit for approximately two hours, stand for fifteen to thirty minutes, and walk for thirty to forty-five minutes. She noted that these lengths of time were "longer/shorter depending on the day and how I feel." She also reported memory and concentration problems, and indicated that she was never able to predict how she would feel from one day to the next. On August 21, 2003, at Liberty's request, Figueroa filled out a questionnaire regarding Cusson's condition. Figueroa was asked to describe "specific restrictions and limitations . . . preventing your patient from being able to work." Figueroa also submitted a Physical Capacities Form, on which he reported that Cusson could lift no more than ten pounds twice a day, could sit for four hours per day but needed to stand up every fifteen -4- minutes, could stand for up to one hour but needed to sit every five to ten minutes, and could walk for an hour but needed to rest every few minutes. Figueroa concluded that Cusson could not work a full eight-hour day until her pain and fatigue subsided. On March 15, 2004, Liberty notified Cusson that beginning September 29, 2004, the definition of "disabled" under the Policy would change, after which Cusson would have to be unable to perform all of the material and substantial duties of any occupation for which she was qualified. On May 7, 2004, Figueroa, at Liberty's request, completed a Functional Capacities Form and a Restrictions Form. The Functional Capacities Form asked Figueroa to list, for various activities, whether Cusson had "[n]o [r]estrictions," could perform the activity "[f]requently (1/3 to 2/3 of the time)," "[o]ccasionally (up to 1/3 of the time)," or was "[u]nable to" perform the activity. Figueroa indicated that Cusson was "[u]nable to" squat, bend, kneel, climb, push or pull objects, or lift any weight over ten pounds. C. 2004 Review and Termination of Cusson's LTD Benefits On May 18, 2004, Liberty hired Omega Insurance Services, Inc. ("Omega") to conduct video surveillance of Cusson. Eleven days later, Liberty again told Cusson to apply for SSDI benefits, warning that if she did not, Liberty would reduce her disability benefit by the amount she was eligible to receive from the Social -5- Security Administration ("SSA"). Liberty also informed Cusson that it would help her obtain SSDI benefits. Omega conducted surveillance of Cusson on June 9-12, 17, and 24, 2004. Omega videotaped Cusson when she was observed outside of her home, and it obtained fifty minutes and fifty-eight seconds of video over the course of these six days. The video footage that was obtained appeared to show Cusson engaged in normal activities. For example, on June 9, Cusson was observed walking her dog for approximately thirty minutes; according to the Omega report, Cusson walked "in a smooth, fluid manner without exhibiting any external signs of impairment." Later that day she drove to a nearby Home Depot store and was seen exiting the store an hour later carrying a shopping bag. On June 24, Cusson was observed having lunch with a friend at a restaurant and then going to a Wal- Mart. Cusson was inside the Wal-Mart for almost one hour, after which she exited with a loaded shopping cart. She loaded the items from her cart, including what Omega's report described as a large bag of cat litter, into her car, and she was also observed bending at the waist to pick up items that had fallen on the ground. On July 20, 2004, Liberty told Cusson that if she did not provide proof of her SSA application by August 3, 2004, Liberty would immediately reduce her benefits by the estimated amount of the SSDI benefit. Liberty also instructed Omega to conduct a second round of surveillance. This surveillance occurred between -6- July 26 and 28, during which time Cusson was seen on the surveillance tape once, on July 27, for a total of one minute and seven seconds. In August 2004, Liberty ordered a third round of video surveillance. Omega performed this surveillance on August 2, 9-10, 14, and 26. Cusson was seen leaving her home only on August 26. Cusson was observed bending fully at the waist to retrieve an item outside her front door. She was then seen driving to a local restaurant and eating lunch with another person, and was then seen driving to a dry cleaner's, a Wal-Mart, and other department stores. During the course of her errands, she was observed carrying various items into and out of stores and bending at the waist to retrieve items from or place items into her car. On September 15, 2004, Cusson completed another Activities Questionnaire from Liberty, in which she described her functional limitations, including pain, fatigue, memory loss, and inability to concentrate. Cusson reported that since her last Activities Questionnaire from March 28, 2003, she could not remember going to a mall because malls were "just too big"; instead, Cusson noted, she went "mostly to small stores [with] easy access in + out." Cusson also reported that she had "good" days and "bad" days. She said that on a "good" day she would be able to perform activities such as running errands, visiting other people, and having lunch outside her home. However, she reported that on -7- a "bad" day she would be unable to sleep because of her pain and would often be unable to get out of bed for several days. On October 13, 2004, Figueroa reported to Liberty that, during an eight-hour work day, Cusson could sit for less than two hours, stand for less than thirty minutes, and walk for fifteen minutes. Figueroa also noted that Cusson could not lift or carry more than ten pounds and could not work eight hours per day in any occupation. Two days later, Figueroa reported to Liberty that Cusson's "limitations tend to vary with the severity of [her] symptoms." Upon receipt of Figueroa's reports, Liberty referred Cusson's file to Dr. Robert Millstein ("Millstein"), a full-time Liberty employee, for review. Millstein reviewed Cusson's claim and concluded that "the video surveillance demonstrates functional capacity which exceeds that reported on the most recent activities questionnaire and confirms that the claimant retains the physical capacity to perform full-time work at a light level." Millstein noted a number of apparent inconsistencies between Cusson's reported limitations and the video surveillance. For example, Millstein noted that although Cusson reported fatigue, the surveillance showed that "on consecutive days, [Cusson] is seen out of her house for long periods of time performing activities which would be best characterized as requiring light to medium physical capacity." Thus, Millstein concluded, "the record does not support -8- the presence of fatigue which would preclude full-time work." Millstein also noted that Cusson was seen going into large stores, such as Wal-Mart and Home Depot, even though she had previously reported that she did not go to malls because malls were too big. Millstein also noted that Cusson was seen bending at the waist and placing a large object, a bag of cat litter, in her car. Millstein's report contained inaccurate statements about the surveillance results. Millstein frequently confused sightings of Cusson returning home with sightings of Cusson leaving her home. Millstein also said that on certain days Cusson left her home and remained out of her home for long periods, when in fact the surveillance for those days clearly shows that Cusson returned to her home at various points throughout the day. As a result of these errors, Millstein's report overstates the amount of time Cusson spent outside of her home on certain days, sometimes by as much as a factor of three. In November 2004, Liberty ordered a fourth round of surveillance, this time conducted by MJM Investigations, Inc. ("MJM"). MJM observed Cusson on November 2-6 and 11, 2004. Cusson was seen outside of her home for a total of roughly half an hour over the course of the surveillance. On November 2, Cusson was observed exiting her house carrying various bags and leaving her home as the passenger in a car. On November 4 and 11, Cusson was -9- briefly seen driving to various locations to run errands. Cusson was not seen outside of her home on November 3, 5, or 6. On November 29, 2004, Liberty conducted a Transferable Skills Analysis and Labor Market Survey ("TSA") of Cusson's claim, relying solely on Millstein's report. The analysis concluded that Cusson could perform her occupation as a Facilities Manager as well as other occupations, such as Construction Manager, Property/Real Estate Manager, and Civil Engineer. On December 8, 2004, Liberty sent Cusson a letter informing her that her benefits were being terminated. The letter discussed the video surveillance and Millstein's review of Cusson's file. The letter also noted the results of the TSA and informed Cusson that she was capable of performing her original job as a Facility Planner. In addition, the letter noted that Liberty had written to Figueroa requesting a response to Millstein's report and the video surveillance, but that Figueroa had not responded. D. Cusson's Appeal and Further Tests On June 1, 2005, Cusson notified Liberty that she was appealing its decision to terminate her benefits. Cusson submitted various documents, including her own sworn statement; affidavits from her friends, family, and former colleagues; and a chart comparing the surveillance reports with her own recollection of her activities on the days she was surveilled. Cusson's sworn statement and activity chart indicate that on some of the days she -10- was seen outside of her home, she would "hit a wall" and return home "crying in pain." However, Cusson's activity chart is completely silent as to what she was doing, or what symptoms she was experiencing, on many days when she was not observed outside of her home. Cusson also submitted medical reports from in-person examinations by Dr. Peter Schur ("Schur"), a rheumatologist, and Drs. Daniel Cohen ("Cohen") and Jean Matheson ("Matheson"), who specialize in sleep disorders. Schur's examination was paid for by Cusson's health insurer as part of her treatment. Schur's examination indicated that Cusson had limited motion in her shoulders, hips, and back; had "fair muscle tone at best"; and was painfully sensitive to touch. Schur concluded that Cusson was disabled "at present," and recommended treatment options. Cohen and Matheson indicated that Cusson had trouble sleeping because of her pain, but they did not render an opinion as to disability. Cusson also submitted a report from Paul Blatchford, Ed. M. ("Blatchford"), a vocational rehabilitation expert, who conducted an in-person evaluation on February 15, 2005. Blatchford's report detailed Cusson's impaired performance on a variety of standardized vocational tests approved by the United States Department of Labor ("DOL"). The report noted that Cusson appeared exhausted and distracted throughout the entire proceeding. Blatchford noted that under DOL guidelines, the occupations for -11- which Cusson was qualified required "sedentary" to "light" exertion levels. Blatchford concluded that Cusson's functional capacity precluded her from performing and sustaining any work-related activities. Blatchford also disputed Liberty's interpretation of the surveillance videos. He argued that the videos failed to address Cusson's cognitive limitations. He also noted that although the video showed that Cusson was capable of short periods of activity, a fact that she herself had reported, it did not show Cusson engaged in sustained activity that would be necessary for any kind of employment. On July 26, 2005, Liberty informed Cusson that it had referred her file for two peer reviews. Both reviews were paid for by Liberty. One review was conducted by Dr. Walter Longo ("Longo"), an oncologist, who determined that there was "no objective data supporting ongoing impairment due to prior therapy" for breast cancer, but that it was possible that Cusson could have limitations due to fibromyalgia. The other was conducted by Dr. Jeffrey Liebermann ("Liebermann"), a rheumatologist, who concluded that Cusson "should be able to perform the duties of light or sedentary occupations" despite her fibromyalgia, and that Cusson's impairments were "entirely subjective." Liebermann also claimed to have spoken on the phone with Figueroa. Liebermann reported that during his conversation with Figueroa, Figueroa said that he had told Cusson that he no longer wished to be consulted about the -12- disability aspects of her case. On August 11, 2005, Liebermann sent Figueroa a letter purporting to summarize their conversation and requesting that Figueroa review and make comments or corrections to the letter and fax it back to him. Figueroa did not sign the letter or return it. On September 7, 2005, Cusson underwent an examination by Dr. Mitchell Abramson ("Abramson"). Abramson reported that Cusson had "multiple trigger points" for pain and was unable to bend forward more than forty degrees due to back pain. Abramson concluded that Cusson "appear[ed] to be disabled due to her medical illness." On September 9, 2005, Blatchford wrote an addendum to his February 15, 2005 report, stating that Liberty's interpretation of the surveillance reports was "inaccurate, and contradicted by the videotapes themselves." Blatchford noted that the video of Cusson running errands was "consistent with [Cusson's] ability to perform activities for limited periods of time," but that Cusson's need for frequent rest and her "tendency to 'fade' after engaging in any task for a sustained period of time" showed that she "lack[ed] the ability to engage in sustained work, whether sedentary or otherwise." Cusson submitted Abramson's report, Blatchford's addendum, and other documentation to Liberty on September 15, 2005. In response, Liberty conducted a second round of peer reviews. Both of these additional reviews were conducted by MLS National -13- Medical Evaluations, Inc. ("MLS") and paid for by Liberty. The first review was conducted by Dr. Reynold Karr ("Karr"), a rheumatologist, who concluded that he did not "identify any objective medical basis for impairment or need for physical restrictions based upon review of the accompanying medical record." The second MLS review was conducted by Dr. Matthew Kaufman ("Kaufman"), an oncologist. Kaufman indicated that, from his perspective as an oncologist, there was no "identifiable or medically supported level of functional impairment that can be objectively verified." However, Kaufman noted that fibromyalgia fell outside his area of expertise, and conceded that Cusson "may have a level of functional impairment due to . . . fibromyalgia and/or psychological overlay." On November 1, 2005, Liberty upheld its decision to terminate Cusson's benefits. Liberty stated that "based on the totality of clinical and objective medical documentation contained in Ms. Cusson's file, sedentary to light capacity has been established." E. Award of SSDI Benefits On January 21, 2006, an Administrative Law Judge ("ALJ") awarded Cusson SSDI benefits, with a disability date of March 15, 2002. In granting the benefits, the ALJ found that Cusson was "unable to do sustained work activities in an ordinary work setting on a regular and continuing basis." -14- F. District Court Proceedings On December 7, 2005, Cusson filed suit against Liberty and the LTD Plan in the District of Massachusetts under ERISA, 29 U.S.C. § 1001 et seq. Cusson sought to recover her benefits pursuant to 29 U.S.C. § 1132(a)(1)(B). Liberty answered and asserted a counterclaim seeking reimbursement for overpayment of benefits under the Policy as a result of Cusson having received SSDI benefits. The parties filed cross-motions for summary judgment on all issues. Because the LTD Plan gave Liberty discretion to determine eligibility for benefits, the district court applied an abuse of discretion standard to Liberty's decision. Under this standard, Liberty's decision would be upheld if it was "reasoned and supported by substantial evidence." Gannon v. Metro. Life Ins. Co., 360 F.3d 211, 213 (1st Cir. 2004). The district court recognized that Liberty had a structural conflict of interest, since it both evaluated benefit claims and paid the benefits. However, the district court, citing the Supreme Court's decision in Metropolitan Life Insurance Co. v. Glenn, held that even in the face of a conflict, the standard of review remained deferential, and that the conflict was "one factor among many that a reviewing judge must take into account." 128 S. Ct. 2343, 2351 (2008). Cusson made two main arguments in her district court case opposing the denial of benefits. First, she argued that Liberty -15- and its reviewers improperly used and interpreted the surveillance videos. Second, she argued that Liberty's review process was flawed, and that these flaws showed that Liberty's decision was an abuse of discretion. The district court held that Liberty's decision to terminate her benefits was not an abuse of discretion. With respect to the surveillance results, the court held that, although the surveillance supported the inference that Cusson could not perform sustained activity, the surveillance also supported the inference that "Cusson and Dr. Figueroa had significantly overstated [Cusson's] functional limitations." The court also rejected Cusson's various claims of procedural unfairness in Liberty's review process. In addition to upholding Liberty's decision to terminate Cusson's benefits, the court also granted summary judgment to Liberty on its counterclaim. Cusson now appeals the district court's decision. II. Discussion A. Applicable Law and Standard of Review We review a district court's grant of summary judgment de novo. Thompson v. Coca-Cola Co., 522 F.3d 168, 175 (1st Cir. 2008). Normally, when we review a district court's grant of summary judgment, "[w]e will reverse only if, after reviewing the facts and making all inferences in favor of the non-moving party [here, Cusson], the evidence on record is sufficiently open-ended to permit a rational factfinder to resolve the issue in favor of -16- either side." Id. (internal quotation marks omitted). However, in an ERISA benefit-denial context, "the district court sits more as an appellate tribunal than as a trial court." Leahy v. Raytheon Corp., 315 F.3d 11, 18 (1st Cir. 2002). In such cases, "summary judgment is simply a vehicle for deciding the issue," and, consequently, "the non-moving party is not entitled to the usual inferences in its favor." Orndorf v. Paul Revere Life Ins. Co., 404 F.3d 510, 517 (1st Cir. 2005). When an ERISA plan gives the administrator the discretion to determine eligibility for benefits (as in this case), a reviewing court must uphold that decision unless it is "arbitrary, capricious, or an abuse of discretion." Gannon, 360 F.3d at 213 (citing Vlass v. Raytheon Employees Disability Trust, 244 F.3d 27, 30 (1st Cir. 2001)). However, Cusson argues that the Supreme Court's decision in Glenn changed the standard of review from an abuse of discretion standard to a "combination of factors" standard in cases where, as here, there is a structural conflict of interest. We disagree. In deciding what effect a conflict would have on the standard of review, the Supreme Court in Glenn "elucidate[d] what [the] Court said in Firestone, namely, that a conflict should 'be weighed as a factor in determining whether there is an abuse of discretion.'" Glenn, 128 S. Ct. at 2350 (quoting Firestone Tire & Rubber Co. v. Bruch, 489 U.S. 101, 115 (1989)). However, the Court -17- went on to say, "We do not believe that Firestone's statement implies a change in the standard of review, say, from deferential to de novo review." Id. Rather, the Court said that "Firestone means what the word 'factor' implies, namely, that when judges review the lawfulness of benefit denials, they will often take account of several different considerations of which a conflict of interest is one." Id. at 2351. Because Glenn did not alter the standard of review, and because our review of the lower court's grant of summary judgment is de novo, we will review Liberty's decision to deny Cusson's benefits under the same abuse of discretion standard used by the district court. See Denmark v. Liberty Life Assurance Co. (Denmark III), 566 F.3d 1, 9 (1st Cir. 2009) (where the plan administrator has discretion, "Glenn's baseline principle, consistent with this circuit's prior precedents, [is] that judicial review of such a benefit-denial decision is for abuse of discretion"). B. Appropriate Weight Given to the Conflict of Interest Although the presence of a conflict of interest does not change the standard of review in this case, the conflict itself can, under certain circumstances, be accorded extra weight in the court's analysis. In Glenn, the Supreme Court held that "[t]he conflict of interest at issue . . . should prove more important (perhaps of great importance) where circumstances suggest a higher likelihood that it affected the benefits decision, including, but -18- not limited to, cases where an insurance company administrator has a history of biased claims administration." 128 S. Ct. at 2351. On the other hand, the conflict "should prove less important (perhaps to the vanishing point) where the administrator has taken active steps to reduce potential bias and to promote accuracy." Id. We have interpreted this language from Glenn to mean that "courts are duty-bound to inquire into what steps a plan administrator has taken to insulate the decisionmaking process against the potentially pernicious effects of structural conflicts." Denmark III, 566 F.3d at 9. Although Cusson argues that it is Liberty's burden to show that the conflict did not affect its decision, the Supreme Court explicitly declined to specify a burden-of-proof rule for determining the weight that a conflict should be given. See Glenn, 128 S. Ct. at 2351 ("Neither do we believe it necessary or desirable for courts to create special burden-of-proof rules, or other special procedural or evidentiary rules, focused narrowly upon the evaluator/payer conflict."). We therefore apply the same burden of proof to the conflict issue that we do to any other aspect of an ERISA claim for improper denial of benefits; hence, Cusson bears the burden of showing that the conflict influenced Liberty's decision. See Terry v. Bayer Corp., 145 F.3d 28, 34 (1st Cir. 1998) ("Here, [claimant] bears the burden of making a showing sufficient to establish a -19- violation of ERISA, namely, that the benefit termination was unreasonable." (internal quotation marks omitted)). Cusson has not raised (and, thus, the parties have not briefed) the issue of the adequacy of Liberty's internal procedures to insulate the review process from the conflict. Instead, Cusson argues that certain alleged flaws in the way the review was conducted prove that Liberty's decision was influenced by its conflict. As discussed below, we do not find that these alleged flaws show that Liberty was improperly influenced by its conflict. Firstly, Cusson argues that Liberty relied exclusively on Millstein's report, including his inaccurate statements about the surveillance results, rather than relying on the totality of the record. Cusson correctly points out that in its initial denial letter, Liberty quoted language that it claimed was from the surveillance reports, but that was actually from Millstein's interpretation of the surveillance. Liberty also claimed that it relied on its TSA, but the record indicates that the TSA was based solely on Millstein's report. We find nothing wrong with the degree of Liberty's reliance on Millstein's report. Although the denial letter from Liberty quoted Millstein's summary of the surveillance data, and not the surveillance reports themselves, the record also indicates that Liberty's case manager did review the tapes. Moreover, as the district court noted, even though Millstein made inaccurate -20- statements about the surveillance data in his report, Millstein's main substantive point about the surveillance -- that it showed Cusson engaged in activities that she claimed she could not do -- was accurate. Many of the activities that Cusson was observed doing were activities that she specifically reported being unable to do, such as bending at the waist and lifting objects heavier than ten pounds. Moreover, when Cusson was seen outside, she did not appear to show any signs of fatigue or impaired physical capabilities. While this is by no means ironclad proof of ability to work, it is evidence that Liberty was entitled to consider. We read Liberty's decision as being based not simply on the amount of activity observed, but also on the nature of the activity observed, and Millstein's report contained no errors with respect to the latter. Next, Cusson argues that Liberty improperly relied on Liebermann's hearsay testimony about his conversation with Figueroa. Liebermann claimed that he spoke with Figueroa and that Figueroa "understood that there was a significant discrepancy between the claimant's reported functionality and what was seen on the surveillance tapes." Liebermann further claimed that Figueroa indicated that he had told Cusson that he no longer wished to be involved in the disability aspect of her case. However, Cusson noted that Figueroa never responded to Liebermann's request to sign off on Liebermann's account of their conversation. Thus, Cusson -21- argues, Liebermann's entire account of the conversation is hearsay that neither Liberty nor the district court should have relied on in its decision. However, "[a] plan administrator is not a court of law and is not bound by the rules of evidence." Speciale v. Blue Cross & Blue Shield Ass'n, 538 F.3d 615, 622 n.4 (7th Cir. 2008). Because Figueroa never responded to Liebermann's request to comment on his account of the conversation, it was not an abuse of discretion for Liberty to assume that Liebermann's account of the conversation was accurate. Since Cusson never presented Liberty with any evidence to challenge directly Liebermann's account of the conversation, we find nothing inappropriate about Liberty's reliance on that account. Cusson next argues that Liberty's reviewing physicians were biased against patients with fibromyalgia. Although Liebermann's report mentions some of the evidence submitted by Cusson, Cusson alleges that Liebermann reached his conclusion about her ability to work without actually analyzing the medical evidence in her file. In his report, Liebermann stated: "This claimant has a diagnosis of [f]ibromyalgia. As such, she should be able to perform the duties of light or sedentary occupations." Cusson argues that this statement indicates that Liebermann had a preconceived notion that fibromyalgia could never be a disabling diagnosis, and thus did not fairly consider her file. Similarly, -22- Millstein cited a 1986 study of fibromyalgia patients that indicated that most of them were eventually able to return to work. Cusson points out that the Seventh Circuit has held that "[t]he fact that the majority of individuals suffering from fibromyalgia can work is the weakest possible evidence that [an individual claimant] can." Hawkins v. First Union Corp. Long Term Disability Plan, 326 F.3d 914, 919 (7th Cir. 2003). If Liebermann or Millstein had said that fibromyalgia patients are never disabled, those statements would be clearly wrong. However, although Liebermann and Millstein made statements that might be interpreted as improper generalizations, it does not necessarily follow that they relied on such generalizations when issuing their reports. Had either doctor concluded that Cusson was not disabled because fibromyalgia is never or not always disabling, that conclusion would have been inappropriate. However, both doctors' reports indicate that they reviewed the surveillance footage and based their conclusions on the apparent contradiction between the footage and Cusson's reported impairments. Cusson also argues that Liberty's reliance on paper file reviews instead of direct medical examinations shows that it was influenced by its conflict. Cusson acknowledges that this court has found "a nonexamining physician's review of a claimant's file [to be] . . . reliable medical evidence." Gannon, 360 F.3d at 214. However, Cusson argues that the particular circumstances of this -23- case warrant a preference for reports by examining physicians. She notes that fibromyalgia is a disease that can only be diagnosed by a clinical examination of a patient based on subjective reports of pain. Certain circuits have held that in such cases some deference to examining physicians is appropriate. See, e.g., Kalish v. Liberty Life Assurance Co., 419 F.3d 501, 509-10 (6th Cir. 2005); Hawkins, 326 F.3d at 919 (the "gravest problem" with the opinion offered by an ERISA plan administrator's paper-reviewing doctor was his refusal to accept "subjective symptoms" of pain and fatigue known to be associated with fibromyalgia). Moreover, Cusson notes that Liberty's reviewing physicians were paid for their reports by Liberty, whereas Cusson's treating physicians were paid by her health insurance company. The Supreme Court has recognized that "physicians repeatedly retained by benefits plans may have an incentive to make a finding of 'not disabled' in order to save their employers money and to preserve their own consulting arrangements." Black & Decker Disability Plan v. Nord, 538 U.S. 822, 832 (2003) (internal quotation marks omitted). We see no reason why the nonexamining physicians' reports in this case are unreliable. We recognize that fibromyalgia is a disease that is diagnosed primarily based on a patient's self- reported pain symptoms. However, Liberty's reviewers did not question the diagnosis of fibromyalgia; instead, they questioned the effect of the disease on Cusson's ability to work. "[T]his -24- court draws a distinction between requiring objective evidence of the diagnosis, which is impermissible for a condition such as fibromyalgia that does not lend itself to objective verification, and requiring objective evidence that the plaintiff is unable to work, which is allowed." Denmark v. Liberty Life Assurance Co. (Denmark II), 481 F.3d 16, 37 (1st Cir. 2007), vacated on other grounds, 566 F.3d 1 (1st Cir. 2009). Because it is permissible to require documented, objective evidence of disability, it was not inappropriate for Liberty's reviewers to rely on the lack of such documented evidence, or on the footage that contradicted Cusson's reports of limitations, in making their recommendations. The fact that Liberty's reviewers were paid for their reports does not, by itself, lead us to believe that Liberty was influenced by its conflict, since Cusson has provided no evidence that Liberty retained its reviewers specifically because they have a record of denying claims. Next, Cusson claims that Liberty failed to provide her complete medical record to its reviewers. Liberty's internal files indicated that "[a]ll medical records" had been sent to its reviewers. However, the reviewers did not list certain documents that were favorable to Cusson -- such as various medical records, her activity log, and portions of the surveillance reports that showed no activity -- in their reports. The district court held that under our holding in Tsoulas v. Liberty Life Assurance Co., it -25- would be improper for the court automatically to assume that unless the medical report lists each item the examiner reviewed, he or she did not review it. See 454 F.3d 69, 77 (1st Cir. 2006). However, Cusson argues that the district court's reliance on Tsoulas was incorrect. Cusson points out that in Tsoulas, the plaintiff's claim that Liberty failed to provide certain documents to the reviewers was obviously incorrect because some of the documents that the plaintiff said were missing were explicitly listed in those same reviewers' reports. Id. In contrast, the documents that Cusson claims were not provided are not listed in any of Liberty's reviewers' reports.2 Cusson's attempt to distinguish Tsoulas from this case fails. Contrary to Cusson's position, the reviewing physician reports in Tsoulas did not explicitly mention certain reports that the claimant said were missing; rather, the district court inferred that those documents were provided because the physicians' reports listed some, but not all, of the documents. See id. Likewise, in this case, each of the reviewing physicians' reports list at least some of the supplementary material provided by Cusson. Hence, in this case, as in Tsoulas, we find no reason to believe that Liberty did not supply the entire record. 2 Cusson also argues that, in the wake of Glenn, the district court was wrong to place the burden of proof on her to show that the records were not provided. However, as we noted above, the Supreme Court explicitly chose not to impose any new burden-of- proof rules with respect to conflicts. -26- Finally, Cusson argues that Liberty's failure to credit the SSA's award of disability benefits shows that Liberty was influenced by its conflict. Cusson notes that Liberty insisted that she apply for SSDI benefits, which she was awarded, but now seeks to benefit financially from that award by claiming the right to reimbursement for overpayment. Cusson notes that in Glenn, the Supreme Court treated this behavior as evidence that a conflict influenced the plan administrator's judgment. In Glenn, the plan administrator encouraged the claimant to apply for SSDI benefits and then recovered the bulk of those benefits by virtue of the reimbursement provision. The administrator then turned around and denied benefits for the same disability that it had encouraged the claimant to plead to the SSA. The Court held: This course of events was not only an important factor in its own right (because it suggested procedural unreasonableness), but also would have justified the court in giving more weight to the conflict (because MetLife's seemingly inconsistent positions were both financially advantageous). Glenn, 128 S. Ct. at 2352. We find nothing suspicious about Liberty's failure to credit Cusson's successful application for SSDI benefits, because Cusson was not awarded SSDI benefits until after Liberty denied her LTD benefits. We have held that when reviewing an insurer's decision to deny benefits, we determine whether "the insurer's eligibility determination was unreasonable in light of the -27- information available to it when it made its decision." Denmark II, 481 F.3d at 39 (internal quotation mark omitted). Because the SSA's award of benefits had not yet occurred when Liberty made its decision, that fact could not have been known to Liberty. In summary, as discussed above, we do not find that Liberty's decision was improperly influenced by its structural conflict of interest. We therefore do not accord any special weight to the conflict in our analysis of whether Liberty's decision was proper, but rather consider it along with all of the factors present in this case to determine if Liberty's ultimate conclusion regarding Cusson's benefits was "reasoned and supported by substantial evidence." Gannon, 360 F.3d at 213. We turn to this question below. C. Whether Liberty's Decision Was an Abuse of Discretion Cusson argues that the factors discussed in Section II(B), supra, which she claims show that Liberty was influenced by its conflict, also show that Liberty's ultimate decision was an abuse of discretion. For the same reasons we articulated above, we do not believe that these factors prove that Liberty's decision was an abuse of discretion. Cusson further argues that, contrary to the conclusions of Liberty and its reviewers, the surveillance video does not actually contradict any of her self-reported disabilities. Cusson never claimed that she was entirely incapable of performing -28- normally; rather, she only claimed that she was unable to function normally most of the time. Thus, Cusson argues, given how infrequently she was seen outside of her home during the course of the surveillance, the surveillance is consistent with her claim of disability, and it was an abuse of discretion for Liberty to use it as evidence against her claim. Cusson notes that the surveillance video represents a mere two hour snapshot from twenty days of her life over the course of four months.3 Cusson cites a number of cases in which courts have held that it was an abuse of discretion to rely on such limited surveillance data. See, e.g., Soron v. Liberty Life Assurance Co., 2005 WL 1173076, at *11 (N.D.N.Y. May 2, 2005) (surveillance cannot be used to discount plaintiff's self-reported symptoms where "it shows her performing isolated activities for brief periods of time with no revelation of consequences"); Osbun v. Auburn Foundry, Inc., 293 F. Supp. 2d 863, 870 (N.D. Ind. 2003) (1.5 hours of surveillance video over two days fell short of demonstrating that plaintiff was capable of 3 Cusson argued to the district court that Liberty's internal training manuals regarding video surveillance should have been admitted to the record. The district court denied this request because Cusson did not demonstrate a "very good reason" for the inclusion of the manuals. See Liston v. UNUM Corp. Officer Severance Plan, 330 F.3d 19, 23 (1st Cir. 2003). Cusson challenges this decision on appeal. However, Cusson does not clearly explain why these documents would assist this court in determining whether Liberty's decision was an abuse of discretion. Because we do not see why these documents would be relevant to this case, we do not reach the issue of whether the district court properly excluded them. -29- sustaining a job); Cross v. Metro. Life Ins. Co., 292 F. App'x. 888, 892 (11th Cir. 2008) (five days of surveillance resulting in two hours of video provided a "mere snapshot" of plaintiff's activities and failed to take into account impact of those activities on plaintiff). We conclude that it was reasonable to use the surveillance as evidence against Cusson's claim of disability. We have permitted ERISA plan administrators to use this type of sporadic evidence in the past. See, e.g., Denmark II, 481 F.3d at 38 (it was not arbitrary and capricious for Liberty to consider reports and photographs from four days of surveillance that showed claimant outside only for short periods of time on two of the four days); Tsoulas, 454 F.3d at 79 (no abuse of discretion where Liberty considered surveillance evidence consisting of four hours per day for three days even though it only represented a "small -- impliedly non-representative -- fraction of each day"). More importantly, although the limited amount of time she was seen outside her home is a factor that weighs in Cusson's favor, Liberty was certainly entitled to take notice of the fact that the video shows Cusson doing particular activities that she claimed she could not do. For example, she is seen going into large stores on two separate occasions, despite having claimed that she avoided malls because they were too big and preferred small stores. While we of course recognize that a Home Depot or a Wal-Mart is not the same -30- thing as a mall, we cannot say that it was an abuse of discretion for Liberty and its reviewers to conclude that Cusson's apparent ability to navigate very large stores contradicted her claim that her illness required her to frequent small stores. The surveillance also shows Cusson bending, kneeling, picking up large objects such as a bag of cat litter, and pushing a loaded cart, despite the fact that the Functional Capacities Form completed by Figueroa on May 4, 2004 indicated that Cusson was physically unable to perform these activities. Because we find that Liberty was entitled to credit the surveillance, we are left to determine whether Liberty's ultimate conclusion was supported by the totality of the evidence. We find that it was. Cusson argues that Liberty failed to consider her objective evidence of disability. However, the record reflects that Liberty reached its decision not because it failed to consider the evidence in Cusson's favor, but because it determined that the surveillance results undermined the credibility of important portions of that evidence. Because we review Liberty's decision under an abuse of discretion standard, we must uphold it if we find that it was "reasoned and supported by substantial evidence in the record."4 Vlass, 244 F.3d at 30 (internal quotation marks 4 In our prior ERISA cases, we have variously described the standard of review, applicable when an ERISA plan grants the plan administrator the discretion to determine eligibility for benefits, as review for "abuse of discretion," "arbitrariness and capriciousness," and "substantial evidence." In the ERISA benefit- -31- omitted). "Evidence is substantial when it is reasonably sufficient to support a conclusion. Evidence contrary to an administrator's decision does not make the decision unreasonable, provided substantial evidence supports the decision." Wright v. R. R. Donnelley & Sons Co. Group Benefits Plan, 402 F.3d 67, 74 (1st Cir. 2005). Because we cannot say that the evidence in this case was insufficient to support Liberty's conclusion, we affirm Liberty's decision to terminate Cusson's benefits. D. Whether Liberty's Counterclaim Is Allowed Liberty asserted a counterclaim to recover Cusson's SSDI benefits under 29 U.S.C. § 1132(a)(3), which states that a plan fiduciary can bring a civil action to obtain "appropriate equitable relief . . . to enforce . . . the terms of the plan." Under the LTD Plan, if Cusson received an overpayment on her disability claim from any source, Liberty "ha[d] the right to recovery of such overpayments from [Cusson] or [Cusson's] estate." Although Liberty terminated Cusson's LTD benefits before Cusson was awarded SSDI benefits, the SSA awarded benefits retroactive to March 15, 2002. Liberty argues that because it paid STD and LTD benefits from this date until December 8, 2004, the SSDI benefits Cusson received for this period are an overpayment, and hence that § 1132(a)(3) allows it to sue to enforce the right to recover the overpayment. denial context, these terms are interchangeable. See Denmark III, 566 F.3d at 7. -32- Cusson argues that Liberty's counterclaim is a legal claim, rather than an equitable claim, and hence that it is not permitted under § 1132(a)(3). Cusson relies on Great West Life & Annuity Insurance Co. v. Knudson, a case involving a similar overpayment provision in an ERISA plan, in which the Supreme Court held that the plan administrator could not use the overpayment provision to recover the proceeds from the plan beneficiaries' tort suit against a third party because the claim was a legal claim rather than an equitable claim. 534 U.S. 204 (2002). However, the district court, in allowing Liberty's claim, relied on Sereboff v. Mid Atlantic Medical Services, Inc., in which the Court reached the opposite result and allowed a plan to recover for overpayment of medical expenses when the claimant received money from a third party in tort. 547 U.S. 356 (2006). In Knudson, the funds recovered in the suit were not actually in the possession of the plan beneficiaries; rather, they had been placed into a "Special Needs Trust" under California law. 534 U.S. at 214. Thus, the Court held that the plan's monetary claim against the beneficiaries would effectively operate as "the imposition of personal liability for the benefits that [the plan] conferred upon [the beneficiaries]," which was a legal rather than equitable claim. Id. In contrast, in Sereboff, the beneficiaries had deposited the proceeds from the suit into their personal investment account, and the plan specifically asserted a claim for -33- some portion of those funds. 547 U.S. at 360, 362-63. Thus, the Court held that the claim was a type of "equitable restitution" because it was for "specifically identifiable funds that were within the possession and control of the Sereboffs." Id. at 362-63 (internal quotation marks omitted). We find that Sereboff, rather than Knudson, controls in this case. Here, like in Sereboff, the LTD Plan targets specific funds for recovery -- Cusson's LTD payments -- and identifies the specific portion to which Liberty is entitled -- the amount of the overpayment while Cusson was receiving benefits under the LTD Plan. We are persuaded by the Eighth Circuit's holding in a similar case that a claim such as this is a claim for equitable relief. See Dillard's Inc. v. Liberty Life Assurance Co., 456 F.3d 894, 901 (8th Cir. 2006) (finding that Liberty's claim was equitable when it sought "a particular share of a specifically identified fund -- all overpayments resulting from the payment of social security benefits"). Moreover, unlike in Knudson, the SSDI benefit was paid to Cusson rather than into a separate trust over which she has no control. It is true that, unlike the insurer in Sereboff, Liberty has not identified a specific account in which the funds are kept or proven that they are still in Cusson's possession. However, the Court in Sereboff noted "'the familiar rul[e] of equity that a contract to convey a specific object even before it is acquired will make the contractor a trustee as soon as he gets a title to -34- the thing.'" 547 U.S. at 363-64 (quoting Barnes v. Alexander, 232 U.S. 117, 141 (1914)). Here, the contract between Cusson and Liberty put Cusson on notice that she would be required to reimburse Liberty for an amount equal to what she might get from Social Security. We therefore find that Liberty's counterclaim is an equitable claim and is allowed under 29 U.S.C. § 1132(a)(3). Cusson further argues that Liberty's claim is barred under the Social Security Act, Title II, 47 U.S.C. §§ 401 et seq. The Act provides, in relevant part: The right of any person to any future payment under this title shall not be transferable or assignable, at law or in equity, and none of the moneys paid or payable or rights existing under this title shall be subject to execution, levy, attachment, garnishment, or other legal process, or to the operation of any bankruptcy or insolvency law. 42 U.S.C. § 407(a). The district court held that this section of the Act only applies to future payments, and not to retroactive payments. However, Cusson argues that the reference to "moneys paid or payable" indicates that the statute's protections apply to all payments, whether future payments or retroactive payments. We agree with the district court that 42 U.S.C. § 407(a) does not bar Liberty's claim. However, we disagree with the district court's reasoning in reaching this conclusion. The district court held that because the statute bars the transfer or assignment of the right to "future payment," it did not bar claims for retroactive payments. However, the statute clearly goes on to -35- say that "none of the moneys paid or payable . . . shall be subject to execution, levy, attachment, garnishment, or other legal process." Id. Thus, the protections afforded by § 407(a) apply to retroactive SSDI payments. See Philpott v. Essex County Welfare Bd., 409 U.S. 413, 416 (1973) (holding that a state could not recover SSDI benefits paid because "the protection afforded by § 407(a) is to 'moneys paid'"); Hoult v. Hoult, 373 F.3d 47, 56 (1st Cir. 2004) ("There is no question that § 407(a) . . . applies to benefits after they have been distributed to beneficiaries."). Instead, we find that § 407(a) does not bar Liberty's claim because Liberty is not attempting to recover Cusson's SSDI benefits. Rather, Liberty seeks to recover in equity from funds Liberty itself already paid under the LTD plan. Although the amount in question happens to be the same as the amount of Cusson's retroactive SSDI payment, the funds Liberty is targeting do not come from SSDI, and thus § 407(a) does not prohibit Liberty's claim. See Holmstrom v. Metro. Life Ins. Co., 615 F. Supp. 2d 722, 753 (N.D. Ill. 2009) (collecting cases in which courts have held that § 407(a) does not bar recovery for overpayment). III. Conclusion For the foregoing reasons, the judgment of the district court is affirmed. Affirmed. -36-