PUBLISHED
UNITED STATES COURT OF APPEALS FOR THE FOURTH CIRCUIT
No. 20-1510
ANITA TEKMEN, Plaintiff – Appellee,
v. RELIANCE STANDARD LIFE INSURANCE COMPANY, Defendant – Appellant.
Appeal from the United States District Court for the Eastern District of Virginia, at Alexandria. Anthony John Trenga, Senior District Judge. (1:18-cv-01304-AJT-MSN)
Argued: September 14, 2022 Decided: December 16, 2022
Before WYNN and HARRIS, Circuit Judges, and KEENAN, Senior Circuit Judge.
Affirmed by published opinion. Judge Wynn wrote the opinion, in which Judge Harris and Senior Judge Keenan joined.
ARGUED: Joshua Bachrach, WILSON ELSER MOSKOWITZ EDELMAN & DICKER L.L.P., Philadelphia, Pennsylvania, for Appellant. Glenn R. Kantor, KANTOR & KANTOR, LLP, Northridge, California, for Appellee. ON BRIEF: Richard Dennis Carter, RICHARD D. CARTER PLLC, Annapolis, Maryland, for Appellee.
2
WYNN, Circuit Judge:
Reliance Standard Life Insurance Company denied Anita Tekmen’s claim for long-
term disability benefits after concluding that she was not “Totally Disabled” as defined by her disability insurance plan. Tekmen brought this action under the Employee Retirement Income Security Act of 1974 (“ERISA”), 29 U.S.C. § 1132(a)(1)(B), arguing that the denial of benefits violated that Act. After conducting a bench trial under Federal Rule of Civil Procedure 52, the district court awarded judgment to Tekmen. Reliance appeals.
On appeal, we affirm that the district court appropriately resolved the matter under Rule 52 and did not clearly err in its factual findings. We also affirm that Tekmen was entitled to long-term disability benefits under the terms of the plan.
I.
In October 2013, Tekmen, while employed as a Financial Analyst for Adsum, Inc., performed financial management services, including analyzing complex financial management systems, maintaining knowledge of business processes and financial software, and conducting technical evaluations and reports. The position required reasoning, cognitive exertion, and the ability to hear and understand easily. Through her employment with Adsum, Inc., Tekmen received coverage under a long-term disability insurance policy provided by Reliance.
Tekmen was involved in a rear-end car accident on October 24, 2013. At an urgent-
care facility several hours later, she reported experiencing neck and lower-back pain, dizziness, headache, and wooziness. The treating provider wrote a letter indicating that Tekmen could return to work four days later. Tekmen did so.
3
Shortly after returning to work, Tekmen was treated by Dr. Frederick W. Parker III, a Family Medicine physician, who diagnosed that she had a concussion. At that time, Tekmen reported experiencing dizziness and headaches. She continued to report significant symptoms three weeks after the accident, including dizziness, sensitivity to light and noise, and difficulty concentrating. She also reported vestibular symptoms, 1 including unsteadiness and difficulty with balance. Tekmen began participating in vestibular therapy, and by late December 2013, her symptoms had begun to improve.
In January 2014, however, Tekmen reported to Dr. Parker that she had developed “an exaggerated auditory response to light vibration, motion and noise.” J.A. 618. 2 This new sensitivity to sound and vibration—called “hyperacusis”—caused Tekmen “extreme distress.” Id. At this time she also reported experiencing tinnitus 3 and difficulty concentrating.
By March 2014, Tekmen’s symptoms had worsened; she continued to experience heightened sensitivity to vibration, light, and noise. Tekmen reported that her sensitivity to sound and vibration made her unable to tolerate being near a running microwave, flushing toilet, or passing truck. Dr. Parker indicated that these symptoms rendered Tekmen unable to concentrate or perform her job duties.
1
The vestibular system encompasses parts of the inner ear and brain responsible for helping to control balance and eye movements.
2
Citations to the “J.A.” refer to the Joint Appendix filed by the parties in this appeal.
3
Tinnitus is the perception of intermittent or constant noise, such as ringing, hissing, whistling, buzzing, or clicking, in the ears or head.
4
In April 2014, Tekmen again began to improve following rest and time off from work through a medical leave of absence. She used headphones and ear plugs and avoided the stimuli that triggered her symptoms. She returned to work, first part-time, and then full- time by July 2014.
But in January 2015, Tekmen’s hyperacusis drastically worsened when her employer moved her work location to a new building, Tower 1. Dr. Parker noted that the heightened symptoms appeared to be triggered by “generators in the lower floors and basement” of the new building, which caused “a low-grade vibration throughout the office.” J.A. 461. Tekmen catalogued her symptoms as including the following: “Heavy head dull pressure headache, severe vibration and noise sensitivity induced vertigo,” “feel dizzy drunk hungover, woozy, unstable, off balance,” “severe cognitive dysfunction and mental fatigue,” “confused, disoriented, CAN NO LONGER THINK OR FUNCTION.” J.A. 1688.
Tekmen’s employer moved her to a different building in February 2015, but in June 2015, Tekmen told Dr. Parker that her hyperacusis had been worsening over the preceding months and that she was using headphones and cotton in her ears to deaden sound. In August 2015, Dr. Parker noted that Tekmen had an episode involving slurred speech, unstable gait, and problems with motor function, among other symptoms. He recommended that Tekmen take a two-week leave of absence from work.
Tekmen attempted to return to work on August 31, 2015, but she had to leave after two hours due to her symptoms. She saw Dr. Parker the following day, and he recommended that Tekmen take a four- to eight-week leave of absence. Tekmen filed a
5
claim for short-term disability benefits with Reliance on September 16, 2015, which was granted. Two months later, she sought to convert her short-term disability benefits into long-term disability benefits, which are the subject of this appeal.
From the time of her accident in October 2013 until she filed the claim for long-
term disability benefits in November 2015, Tekmen saw many specialists and received multiple possible explanations for her symptoms. She was examined numerous times by Dr. Ruben Cintron, a neurologist who began treating Tekmen in May 2014 for her hyperacusis and balance problems. She also sought treatment from other physicians, including neurologists, neurotologists, vestibular therapists, and head injury specialists. She was diagnosed with post-concussion syndrome, hyperacusis, endolymphatic hydrops, tinnitus, and vestibular dysfunction, among other possible diagnoses. Tekmen also underwent several tests, including MRIs, EEGs, and CT scans, as well as tests of cognitive function and hearing ability. The results of many of these tests were normal or otherwise failed to fully explain her reported symptoms.
In reviewing Tekmen’s application for long-term disability benefits, Reliance hired two independent physicians—Dr. Julius Damion and Dr. Leonid Topper—to review Tekmen’s medical records and determine whether she was totally disabled. Dr. Damion concluded that Tekmen had “work capacity on a full time consistent basis on or around the 09/01/2015 Date of Loss and forward, with the stipulation that a quiet work environment in which there is no ambient vibration be provided for her.” J.A. 4076. Dr. Topper found that there was “no evidence of any objective neurological illness affecting the claimant” around September 1, 2015; that her clinical history could not be explained by “any
6
recognizable neurological diagnosis”; and that, “[i]n the absence of specific diagnosis, there is no evidence of impairment.” J.A. 4057.
Reliance denied Tekmen’s claim for long-term disability benefits on May 2, 2016.
In support of its determination, Reliance explained that Dr. Topper noted that “sensitivity to noise” like that experienced by Tekmen “is not expected from the neurological point of view in cases of concussion” and that “while concussion and whiplash injury” from the car accident were “believable,” the symptoms “would not increase over time, or be subject to new symptoms months later.” J.A. 90. And, Reliance explained, “Dr. Topper further notes that there is no conclusive evidence to establish any neurological diagnoses to explain [Tekmen’s] physical symptoms on or around 09/01/2015.” Id.
Reliance acknowledged Dr. Damion’s findings that Tekmen “may have symptoms of hyperacusis and tinnitus.” J.A. 91. Nevertheless, Reliance concluded that those conditions did not “prevent [Tekmen] from working full-time in [her] Regular Occupation” 4 because “Dr. Damion state[d] that [Tekmen] did not have any medical testing that would support that vibratory stimuli would preclude [her] ability to work in an office setting.” Id.
Additionally, Reliance noted that, although Tekmen was receiving treatment for her hyperacusis and tinnitus symptoms, demonstrated mild to moderate hearing loss, and
4
Reliance classified Tekmen’s Regular Occupation as that of a “Budget Analyst,”
defined as someone who “[a]nalyzes accounting records to determine financial resources required to implement [a] program” and makes “recommendations for budget allocations,” “[r]ecommends approval or disapproval of requests for funds,” and “[a]dvises staff on cost analysis and fiscal allocations.” J.A. 742.
7
showed an abnormal gait and unsteadiness in some examinations, other tests—including MRIs and neuropsychological evaluations—were normal. According to Reliance, those results belied Tekmen’s claim of “Total Disability.” Reliance concluded that the exacerbation of hyperacusis and tinnitus symptoms in early 2015 was due to “building- specific” noise and vibration. J.A. 92. Reliance acknowledged that Dr. Damion stated that Tekmen should have a quiet work environment with no ambient vibration but concluded that this “minor work-site provision” was “location-specific and building-specific” and “not applicable to [Tekmen’s] Regular Occupation in the national economy.” Id.
Therefore, Reliance concluded that Tekmen did not have an impairment that precluded her continued employment, stating: “We believe . . . that actual examination findings and the results of diagnostic studies fail to support the level of impairment that you are claiming. Rather, we find the opinions provided by independent physicians Dr. Damion and Dr. Topper, as portraying a more accurate depiction of your physical ability to perform the material duties of your Regular Occupation.” Id.
Tekmen filed an appeal, which was referred to Reliance’s Quality Review Unit for independent review. In evaluating the appeal, Reliance hired two additional independent examiners to review Tekmen’s file. Dr. David Foyt, an otolaryngologist, noted that although there is “really no objective way” of evaluating hyperacusis, from a purely “otologic or neuro-otologic standpoint,” he could not find any “objective evidence” of limitations. J.A. 4568–69. He therefore concluded that Tekmen had the capacity to work full-time from September 1, 2015 on. Dr. Laurie Truog, a psychiatrist, evaluated Tekmen’s claimed disability from a purely psychiatric perspective and concluded that she had “total
8
functional ability from a psychiatric standpoint as of 9/1/15 and ongoing.” J.A. 4599. Reliance affirmed the denial of benefits on March 13, 2017.
Tekmen brought this action in federal district court on October 18, 2018. The parties filed cross-motions for summary judgment. In her opposition to Reliance’s motion for summary judgment, Tekmen requested that the court instead resolve the case pursuant to Federal Rule of Civil Procedure 52. Reliance did not address this request in its reply.
The district court held a hearing on the summary judgment motions on August 23, 2019. On March 31, 2020, the district court denied both motions for summary judgment but awarded judgment to Tekmen on the merits after conducting a bench trial based on the administrative record pursuant to Federal Rule of Civil Procedure 52. Reliance appealed.
II.
On appeal, we first address two interrelated questions: the method a district court uses to resolve an ERISA denial-of-benefits case, like this one, and the standard of review we employ on appeal. Reliance argues that courts in this Circuit are required to resolve ERISA denial-of-benefits cases via summary judgment and that the district court erred in dispensing with this case through a bench trial. Reliance also argues that this Court must review the district court’s legal conclusions—and, seemingly, its factual findings as well— de novo. We disagree with both assertions.
A.
Our sister circuits take diverging views on how ERISA denial-of-benefits cases should be handled by the district courts. Some have concluded that, although summary judgment may be appropriate when there is no genuine issue as to any material fact, a bench
9
trial is appropriate when fact-finding is required. See, e.g., Casey v. Uddeholm Corp., 32 F.3d 1094, 1098–99 (7th Cir. 1994); Kearney v. Standard Ins. Co., 175 F.3d 1084, 1094– 95 (9th Cir. 1999). At least one has concluded that neither summary judgment nor a bench trial is appropriate in ERISA denial-of-benefits cases and that an alternative form of review unique to ERISA cases is appropriate. See Wilkins v. Baptist Healthcare Sys., Inc., 150 F.3d 609, 618–19 (6th Cir. 1998) (Gilman, J., concurring; constituting the opinion of the court as to this issue). Still others provide that a modified, quasi-summary-judgment procedure is appropriate. See Orndorf v. Paul Revere Life Ins. Co., 404 F.3d 510, 517 (1st Cir. 2005); LaAsmar v. Phelps Dodge Corp. Life, Accidental Death & Dismemberment & Dependent Life Ins. Plan, 605 F.3d 789, 795–96 (10th Cir. 2010). Because Reliance appeared to endorse this quasi-summary-judgment procedure at oral argument, Tekmen v. Reliance Standard Life Ins. Co., No. 20-1510 (4th Cir. Sept. 14, 2022), https://www.ca4.uscourts.gov/OAarchive/mp3/20-1510-20220914.mp3, we will describe that procedure in more detail before explaining why we do not find it to be the appropriate mechanism for resolving ERISA denial-of-benefits cases.
To begin, the well-established, traditional summary judgment standard provides that a party is entitled to summary judgment when there is no genuine dispute as to any material fact and the moving party is entitled to judgment as a matter of law. Fed. R. Civ. P. 56(a). In considering a motion for summary judgment, the court construes “all facts and reasonable inferences therefrom in the light most favorable to the nonmoving party.” United States v. 8.929 Acres of Land in Arlington Cnty., 36 F.4th 240, 252 (4th Cir. 2022) (quoting Carter v. Fleming, 879 F.3d 132, 139 (4th Cir. 2018)). In other words, summary
10
judgment is appropriate when the evidence “is so one-sided that one party must prevail as a matter of law.” Id. (quoting Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 252 (1986)).
Ordinarily in the summary judgment context, the court “cannot weigh the evidence or make credibility determinations.” Jacobs v. N.C. Admin. Off. of the Cts., 780 F.3d 562, 569 (4th Cir. 2015). Indeed, the Supreme Court has made clear that “at the summary judgment stage the judge’s function is not himself to weigh the evidence and determine the truth of the matter but to determine whether there is a genuine issue for trial.” Anderson, 477 U.S. at 249. Thus, although a judge may address facts when ruling on a motion for summary judgment, “[b]y definition, no findings of material facts that were ‘in genuine issue’ are possible in granting summary judgment.” PHP Healthcare Corp. v. EMSA Ltd. P’ship, 14 F.3d 941, 944 n.3 (4th Cir. 1993). This is so because, if material facts were in genuine dispute, summary judgment would not be warranted. Fed. R. Civ. P. 56(a).
Some circuits, however, use a modified summary judgment standard unique to ERISA denial-of-benefits cases that are based solely on an administrative record. This modified standard “differs in one important aspect from the review in an ordinary summary judgment case.” Orndorf, 404 F.3d at 517. The First Circuit has explained the procedure as follows:
When deciding whether a party is entitled to summary judgment, we typically view the record evidence in the light most favorable to the non-
moving party, drawing all reasonable inferences in that party’s favor. Our approach is different, however, in the ERISA benefit-denial context, where the record before us is the same record that was before the plan administrator.
In such a case, “summary judgment is simply a vehicle for deciding the [benefits] issue” and “the non-moving party is not entitled to the usual inferences in its favor.”
11
Gent v. CUNA Mut. Ins. Soc’y, 611 F.3d 79, 82–83 (1st Cir. 2010) (citations omitted); see also LaAsmar, 605 F.3d at 796.
This Court has never endorsed such an ERISA-specific quasi-summary-judgment procedure. See Neumann v. Prudential Ins. Co. of Am., 367 F. Supp. 2d 969, 977–78 (E.D. Va. 2005) (describing lack of Fourth Circuit authority analyzing the appropriateness of summary judgment in ERISA § 1132(a)(1)(B) cases); Weisner v. Liberty Life Assurance Co. of Boston, 192 F. Supp. 3d 601, 608 (D. Md. 2016) (same). Although many ERISA cases in this Circuit arising under § 1132(a)(1)(B) are resolved via summary judgment, see, e.g., Johnson v. Am. United Life Ins. Co., 716 F.3d 813, 819 (4th Cir. 2013); Shupe v. Hartford Life & Accident Ins. Co., 19 F.4th 697, 706, 712 (4th Cir. 2021), some district courts have employed bench trials, see, e.g., Neumann, 367 F. Supp. 2d at 977–80; Watson v. Unum Life Ins. Co. of Am., 126 F. App’x 604, 605 (4th Cir. 2005). Indeed, rather than endorsing the quasi-summary-judgment mechanism, we have questioned the propriety of employing summary judgment in the ERISA-benefit-denial context. See Phelps v. C.T. Enters., 394 F.3d 213, 218 (4th Cir. 2005) (noting, in dicta, that the Court had “reservations” regarding “[t]he propriety of importing the summary judgment standard whole-cloth into the ERISA context” and favorably citing the Sixth Circuit’s discussion of the issue in Wilkins, 150 F.3d at 617). We decline to endorse the quasi-summary-judgment procedure now.
The difficulty with employing summary judgment in the ERISA denial-of-benefits context arises where the parties disagree as to key facts. Although in such cases the court will decide the ultimate legal question of whether the individual meets the relevant plan
12
definition of disability, see Orndorf, 404 F.3d at 518, it may first need to resolve competing factual contentions within the administrative record about the cause, severity, or legitimacy of an individual’s impairment. But the court’s role at the summary judgment stage is not to resolve disputed questions of fact. Anderson, 477 U.S. at 249.
The present case is illustrative. Here, Dr. Parker, who examined Tekmen many times over the years following her accident, stated that, although Tekmen may not present as the “classic postconcussion patient,” he would nevertheless “vouch for the fact that her disability is severe and she is significantly impaired to the point where gainful work related activities are essentially impossible.” J.A. 2198. Dr. Cintron, a neurologist who treated Tekmen continually starting in 2014, described her as “incapacitated” and “debilitated” by her symptoms, supporting his “clinical opinion that she is not able to maintain a job in the foreseeable future.” J.A. 1731. On the other hand, the independent examiners who evaluated Tekmen’s claims in connection with her application for long-term disability benefits uniformly stated that Tekmen was not impaired to an extent that would preclude her full-time employment.
Rather than addressing the apparent tension between the summary judgment standard and the need to resolve competing factual contentions in cases involving ERISA benefits denials, some courts have concluded that “[w]here review is properly confined to the administrative record before the ERISA plan administrator, . . . there are no disputed issues of fact for the court to resolve.” Orndorf, 404 F.3d at 518. Instead, the argument goes, courts employing the quasi-summary judgment procedure simply take the administrative record as it stands and “conduct[] de novo review of [the] ultimate
13
conclusion” as to whether the plaintiff has established a disability by the plan’s terms. Id. at 518–19. In this review, “the non-moving party is not entitled to the usual inferences in its favor.” Id. at 517.
We disagree that there can never be disputed issues of fact where review is based on the administrative record. Where, as here, the district court is faced with directly at-odds contentions regarding whether the individual’s claimed impairment is genuine, we see no alternative to the district court making findings of fact. And where such findings implicate material issues, summary judgment simply is not appropriate. See Anderson, 477 U.S. at 248–49; Jacobs, 780 F.3d at 568–69.
Further, if the district court were to resolve a denial-of-benefits case involving disputed facts at summary judgment but without the attendant summary-judgment presumptions, it would effectively be engaging in factfinding that is subject to a de novo standard of appellate review. Yet as the Supreme Court has made clear in a different context, duplicating the district court’s factfinding efforts in this Court “would very likely contribute only negligibly to the accuracy of fact determination at a huge cost in diversion of judicial resources.” Anderson v. City of Bessemer City, 470 U.S. 564, 574–75 (1985) (reviewing the Court of Appeals’ application of clear-error review to the district court’s factual finding of discriminatory intent in an action under Title VII of the Civil Rights Act of 1964).
In conducting de novo review of a denial of benefits under ERISA, the district court undertakes a careful examination of the often-voluminous administrative record to determine whether the claimant was entitled to benefits. This may involve assessing
14
credibility and determining the appropriate weight to assign evidence. See, e.g., Neumann, 367 F. Supp. 2d at 980–94 (resolving an ERISA denial-of-benefits case pursuant to a bench trial on the administrative record). And although Courts of Appeals are also capable of reviewing cold records, district courts are institutionally assigned the role of finder of fact. Yet the approach Reliance endorses would convert the district court’s resolution of the facts from the “main event” to a mere “tryout on the road,” City of Bessemer City, 470 U.S. at 575 (citation omitted), with no apparent benefit to the accuracy of the factual determinations. Because either party could appeal whatever findings the district court rendered, the district court’s careful and time-consuming determinations of fact would be essentially superfluous. This would give district courts little reason to invest the time in factfinding necessary in cases with genuine disputes of material fact and would require redundant factfinding by the appellate courts, which are less institutionally equipped to take on that role. For that reason, “review of factual findings under the clearly-erroneous standard—with its deference to the trier of fact—is” supposed to be “the rule, not the exception.” Id.
At bottom, the ERISA-specific quasi-summary-judgment procedure is a solution in search of a problem. The Federal Rules of Civil Procedure already provide a mechanism for district courts to resolve disputed facts and render a judgment, and that mechanism was employed by the district court here: a Rule 52 bench trial. We see no reason to contort the traditional summary-judgment analysis to fill a nonexistent procedural void. Indeed, in our view, we are not permitted to alter the Federal Rules in the manner Reliance suggests.
15
Accordingly, in the context of de novo review of ERISA denial-of-benefits cases as in any other context, district courts should employ the appropriate procedural mechanism for resolving the case before them as defined by the Federal Rules of Civil Procedure. 5 Summary judgment in such cases is appropriate when there is no genuine dispute as to any material fact and the moving party is entitled to judgment as a matter of law. Where there are disputed issues of material fact, a Rule 52 bench trial, which will typically be limited to the administrative record that was before the plan administrator, 6 is appropriate. Put simply, the court may not “recite[] the familiar rules governing summary-judgment proceedings” but “not follow them.” Avenoso v. Reliance Standard Life Ins. Co., 19 F.4th 1020, 1024 (8th Cir. 2021). As we have observed, findings of material facts in genuine issue are “[b]y definition” not possible in granting summary judgment. PHP Healthcare Corp., 14 F.3d at 944 n.3.
B.
Nonetheless, Reliance contends that our case law establishes that summary judgment is the only appropriate vehicle for resolution of ERISA denial-of-benefits claims. Its argument goes as follows: We have previously stated that our review of ERISA benefits
5
Because the district court employed de novo review and neither party challenged the use of that standard, we express no view on the appropriate procedural mechanism for resolving cases in which review in the district court is for abuse of discretion.
6
When conducting de novo review of ERISA benefits denials, district courts in this Circuit may allow new evidence that was not before the plan administrator “only when circumstances clearly establish that additional evidence is necessary to conduct an adequate de novo review of the benefit decision.” Quesinberry v. Life Ins. Co. of N. Am., 987 F.2d 1017, 1025 (4th Cir. 1993).
16
denials is de novo. But a district court’s factual findings under Rule 52 must be reviewed for clear error. Fed. R. Civ. P. 52(a)(6) (findings of fact “must not be set aside unless clearly erroneous”). So, Reliance posits, Rule 52 cannot be used to resolve ERISA claims because our review even of factual findings has to be de novo, not for clear error. We disagree.
To be sure, we have indicated that our review in these cases is de novo. In Johnson v. American United Life Insurance Co., we stated: “In an appeal under ERISA, we review a district court’s decision de novo, employing the same standards governing the district court’s review of the plan administrator’s decision.” 716 F.3d at 819 (quoting Williams v. Metropolitan Life Ins. Co., 609 F.3d 622, 629 (4th Cir. 2010)). We then recited the Supreme Court’s instruction in Firestone Tire & Rubber Co. v. Bruch that courts must review a denial of benefits “under a de novo standard unless the benefit plan gives the administrator or fiduciary discretionary authority to determine eligibility for benefits or to construe the terms of the plan.” Id. (quoting Firestone Tire & Rubber Co. v. Bruch, 489 U.S. 101, 115 (1989)). “Accordingly,” we continued, “our review of the denial of benefits in this case is de novo.” Id.; see also Shupe, 19 F.4th at 706 (explaining that because “[t]he Supreme Court has held that ‘a denial of benefits challenged under [ERISA] is to be reviewed under a de novo standard’” unless the relevant decisionmaker is vested with discretion, review in this Court should be de novo (quoting Firestone, 489 U.S. at 115)).
But Reliance’s argument ultimately rests on a faulty premise, because while these statements indicate that review of the district court’s legal conclusions is de novo, they do not suggest that this Court must review factual findings de novo. If we can review factual findings for clear error—which, as noted, is the typical rule for our review of a Rule 52
17
judgment—then there is no contradiction between our prior case law and the clear-error standard of review we apply to factual findings made in bench trials.
The Supreme Court in Firestone did not suggest it intended to create a new, ERISA-
specific rule that Courts of Appeals should conduct de novo reviews of factual findings in the ERISA denial-of-benefits context. And we have not interpreted the case that way. We have previously stated that the standards of review outlined in Firestone—that is, de novo review where the plan does not vest the administrator with discretion and review for abuse of discretion where it does—“apply to a district court’s review of a plan administrator’s coverage determinations” and “also are applicable on appellate review.” Bynum v. Cigna Healthcare of N.C., Inc., 287 F.3d 305, 311 (4th Cir. 2002) (emphasis added), abrogated on other grounds by Carden v. Aetna Life Ins. Co., 559 F.3d 256 (4th Cir. 2009). But the Bynum Court nevertheless instructed that the district court’s factual findings made after a bench trial should be reviewed for clear error. Id. at 311–12 (“Our review is affected by the principle that when we review a district court’s decision on the merits, we are, absent clear error, bound by its factual findings.” (citation omitted)). Thus, as explained above, we reject Reliance’s view.
C.
In its briefing, Reliance alleges two other errors by the district court. Neither holds.
First, Reliance argues that the district court’s resolution of this case under Rule 52 rather than Rule 56—after the parties had moved for summary judgment under Rule 56 and the court held a hearing on the motions—violated the party presentation principle by improperly reframing the legal question presented by the parties. The party presentation
18
principle provides that courts “rely on the parties to frame the issues for decision and assign to courts the role of neutral arbiter of matters the parties present.” Greenlaw v. United States, 554 U.S. 237, 243 (2008). Reliance cites United States v. Sineneng-Smith, in which the Ninth Circuit panel effected a “takeover” of the case by inviting three organizations to participate as amici, ordering additional briefing from those organizations to address new legal questions not previously raised by the parties, and then deciding the case on the basis of those arguments. 140 S. Ct. 1575, 1580–81 (2020). The Supreme Court vacated the judgment and remanded.
This case is a far cry from Sineneng-Smith and does not implicate the party presentation principle. The district court here did not reshape the legal question presented by the parties; it simply adjusted the procedural mechanism it would use to address the correctness of Reliance’s decision. Reliance had notice that a Rule 52 bench trial was on the table: Tekmen specifically requested the use of Rule 52, and both parties acknowledged at the summary judgment hearing that the district court’s decision would implicate factfinding. Further, Reliance points to no authority providing that district courts may not conduct a Rule 52 bench trial following the parties’ motions for summary judgment, and we hold that it was appropriate for the district court to resolve this case pursuant to Rule 52. 7
7
We recognize that, in some cases, a party may develop or change its strategy for presenting evidence based on whether the district court is ruling at the summary judgment stage or after a bench trial. Compare Fed. R. Civ. P. 56(a) (“The court shall grant summary judgment if the movant shows that there is no genuine dispute as to any material fact . . . .” (emphasis added)), and E.D. Va. Adm. R. 56(B) (requiring that a summary judgment brief
19
Second, Reliance contends that the district court erred when it “ignored” Tekmen’s alleged deemed admissions. Opening Br. at 13. Under the Eastern District of Virginia’s Local Rule 56, each brief in support of a motion for summary judgment “shall include a specifically captioned section listing all material facts as to which the moving party contends there is no genuine issue and citing the parts of the record relied on to support the listed facts as alleged to be undisputed,” and a brief in response “shall include a specifically captioned section listing all material facts as to which it is contended that there exists a genuine issue necessary to be litigated and citing the parts of the record relied on to support the facts alleged to be in dispute.” E.D. Va. Adm. R. 56(B). “In determining a motion for summary judgment, the Court may assume that facts identified by the moving party in its listing of material facts are admitted, unless such a fact is controverted in the statement of genuine issues filed in opposition to the motion.” Id. (emphasis added).
Reliance contends that Tekmen failed to respond to Reliance’s statement of material facts and, therefore, that Tekmen should be deemed to have admitted these facts. Tekmen acknowledged that there were deficiencies in the initial statement of material facts and filed
include a list of “undisputed” facts), with Fed. R. Civ. P. 52(a)(1) (“In an action tried on the facts without a jury . . . , the court must find the facts specially . . . .”), and Bench Trial Instructions Order, at 2, Tekmen v. Reliance Standard Life Ins. Co., No. 1:18-cv-01304- AJT-MSN (E.D. Va. June 20, 2019), ECF No. 15 (explaining that the parties “should file proposed findings of fact” for the bench trial). Although this distinction does not alter our conclusion here, see J.A. 6601 (Reliance’s counsel’s acknowledgement that the court will act as “the finder of fact”); J.A. 6605 (Reliance’s counsel’s request for the court to “compare[]” evidence), we acknowledge that a district court should generally follow its established procedures for either ruling on a summary judgment motion or, in the alternative, proceeding to a bench trial.
20
a motion to amend the memorandum to correct the errors, which the court granted. But, Reliance argues, “[b]y taking this case outside of Rule 56, the district court was able to avoid” Tekmen’s admissions. Opening Br. at 20. Reliance’s argument on this point fails because this provision of the Local Rules is plainly permissive, not mandatory. Therefore, even if the district court had decided the case on summary judgment, the court would not have been required to deem admitted the facts in Reliance’s statement.
III.
Having dispensed with the arguments concerning standard of review, we reach the merits of Tekmen’s claim. Because we conclude that it was correct for the district court to conduct a Rule 52 bench trial, we will review the court’s factual findings for clear error, Fed. R. Civ. P. 52(a)(6), and review de novo its legal conclusion that Tekmen was entitled to benefits. Brundle ex rel. Constellis Emp. Stock Ownership Plan v. Wilmington Tr., N.A., 919 F.3d 763, 773 (4th Cir. 2019) (citing Nat’l Fed’n of the Blind v. Lamone, 813 F.3d 494, 502 (4th Cir. 2016)). We conclude that the district court did not clearly err in its findings of fact, and we agree with the district court’s conclusion that Tekmen was entitled to benefits.
A.
“A finding is clearly erroneous when although there is evidence to support it, the reviewing court on the entire evidence is left with the definite and firm conviction that a mistake has been committed.” Wall v. Rasnick, 42 F.4th 214, 220 (4th Cir. 2022) (quoting United States v. Blackledge, 751 F.3d 188, 194 (4th Cir. 2014)).
21
The district court’s most consequential factual determination was its decision to give more weight to the reports of the two physicians who repeatedly treated Tekmen following her accident, as well as to the additional treating physicians, than to the physicians hired by Reliance who merely reviewed Tekmen’s file. In so doing, the district court credited the accounts of Dr. Parker, who consistently expressed his view that Tekmen was profoundly impaired in her daily functioning, and Dr. Cintron, who similarly believed Tekmen to be completely unable to work due to her symptoms.
The district court also credited the opinions of Tekmen’s other treating physicians.
These included Dr. Yvette Sandoval, a neurologist who diagnosed Tekmen with Postconcussion Syndrome; Dr. Josef Gurian, an ear, nose, and throat doctor who diagnosed Tekmen with hearing loss, vertigo/dizziness, tinnitus, hyperacusis, and “possible third window syndrome,” J.A. 6004–05; Dr. Bryan McKenzie, a neurotologist and ear, nose, and throat specialist who believed that Tekmen “could have either perilymphatic fistula or superior semicircular canal dehiscence,” J.A. 4913; Dr. Joseph Furman, an otolaryngologist and otoneurologist who diagnosed Tekmen with posttraumatic endolymphatic hydrops; and Dr. Sanjay Prasad, a neurotologist and ear, nose, and throat specialist who suspected possible “[e]ndolymphatic hydrops with perilymphatic fistula or hyperacusis from sensitivity to the cochlear fluid wave,” both of which he suggested could be treated with surgery, J.A. 2793.
Reliance points to three other treating clinicians who, according to Reliance, cast doubt on Tekmen’s claims. First, Dr. Michael Jaffee, a brain injury specialist, believed Tekmen’s symptoms were “embellished,” but also suggested that a “post-concussive
22
sustained migraine variant” might be contributing to her symptoms. J.A. 2061. Steven Singer, a psychiatric specialist, indicated that Tekmen had a family history of “hypochondriases,” but that Tekmen “made no connection” between that history and her own symptoms, which she “very firmly believe[d] ha[d] a medical etiology.” J.A. 1325. He concluded that Tekmen’s symptoms and family history warranted further psychiatric analysis. Dr. Nadia Robertson, a psychiatrist, evaluated Tekmen in connection with Tekmen’s reported feelings of depression and hopelessness stemming from her hyperacusis and concluded that those psychiatric symptoms were indicative of an adjustment disorder with depressed mood.
As discussed, Reliance hired four independent examiners who only reviewed Tekmen’s paper file. Dr. Damion, an otolaryngologist, concluded that there was no worsening of Tekmen’s symptoms as of September 1, 2015—the date that she initially claimed disability—and that she had “capacity” to work “on a full time consistent basis” from that date forward, so long as she was provided with “a quiet work environment in which there is no ambient vibration” or noise. J.A. 1460–61. Dr. Topper, a neurologist, concluded that Tekmen had no change in condition as of September 1, 2015, and concluded that there was no evidence to support a neurological diagnosis that would explain Tekmen’s symptoms. He emphasized Dr. Jaffee’s conclusion that Tekmen’s symptoms were embellished and concluded that Tekmen was not impaired. Dr. Truog, a psychiatrist, concluded that Tekmen’s psychiatric symptoms were consistent with an adjustment disorder with depressed mood and that she was not impaired from a psychiatric standpoint. Finally, Dr. Foyt, an otolaryngologist and neurotologist, found no objective evidence to
23
support Tekmen’s claimed symptoms, determined that there were no significant changes in condition as of September 1, 2015, and concluded that there was nothing from an otologic or neuro-otologic standpoint that would prevent her from working.
Reliance asserts that it was error for the district court to give more weight to the opinions of treating physicians than to the opinions of physicians who only reviewed Tekmen’s paper file. The crux of Reliance’s argument on this point is its contention that the Supreme Court’s decision in Black & Decker Disability Plan v. Nord, 538 U.S. 822 (2003), prohibits plan administrators—and by extension, the district court—from giving more weight to the opinions of treating physicians than those of non-treating physicians.
But Nord did not create such a rule. Instead, Nord provided that “[n]othing in [ERISA] itself . . . suggests that plan administrators must accord special deference to the opinions of treating physicians.” Id. at 831 (emphasis added). Thus, that case provides no support for the contention that administrators and the district court are not permitted to give additional weight to the opinions of treating physicians; it merely said they are not required to do so. Where, as here, the district court determines that the accounts of treating physicians are more persuasive than those of physicians who only examined a paper record, it is not error for the district court to assign those opinions more weight.
True enough, there are conflicting views even among the treating physicians as to the extent, cause, and severity of Tekmen’s symptoms. But most physicians who treated Tekmen believed her to have legitimate impairment in functioning, even in the face of normal test results. And the two physicians who consistently examined and treated Tekmen—Drs. Parker and Cintron—believed that her symptoms were legitimate and
24
disabling. Dr. Cintron’s view is particularly compelling, given that his specialty is neurology.
Reliance also suggests that the fact that Tekmen was able to work for most of the two years following her accident undermines the conclusion that she is unable to work due to her symptoms. Not so. The evidence in the record establishes that, contrary to Reliance’s assertions that there was no change in Tekmen’s condition between her accident in 2013 and the alleged onset of disability in late summer 2015, Tekmen’s symptoms significantly worsened in January 2015 and remained elevated until she applied for disability benefits. See J.A. 6581 (district court finding that Tekmen had continuous symptoms from the time of her 2013 accident that were “exacerbated” in January 2015 and “consistent[ly] worsen[ed]” from there).
But further, as a number of our sister circuits have concluded, there is no “logical incompatibility between working full time and being disabled from working full time.” Hawkins v. First Union Corp. Long-Term Disability Plan, 326 F.3d 914, 918 (7th Cir. 2003); see also Doe v. Standard Ins. Co., 852 F.3d 118, 125 n.5 (1st Cir. 2017); Rochow v. Life Ins. Co. of N. Am., 482 F.3d 860, 865 (6th Cir. 2007); Lasser v. Reliance Standard Life Ins. Co., 344 F.3d 381, 392 (3d Cir. 2003); cf. Seitz v. Metro. Life Ins. Co., 433 F.3d 647, 651 (8th Cir. 2006); Levinson v. Reliance Standard Life Ins. Co., 245 F.3d 1321, 1326 n.6 (11th Cir. 2001). Although “[a] desperate person might force [her]self to work” despite her illness, “even a desperate person might not be able to maintain the necessary level of effort indefinitely.” Hawkins, 326 F.3d at 918. And a person with a disability “should not be punished for heroic efforts to work by being held to have forfeited [her] entitlement to
25
disability benefits” when she must ultimately stop working. Id. As the district court implicitly recognized, the fact that Tekmen was able to work for a period despite her symptoms does not necessarily mean that she was not disabled.
Upon review of the entire record, we are not “left with the definite and firm conviction that a mistake has been committed.” Wall, 42 F.4th at 220 (quoting Blackledge, 751 F.3d at 194). To the contrary, the district court accurately represented the evidence in the record and drew reasonable conclusions as to its import. Therefore, we conclude that the district court did not clearly err in its findings of fact.
B.
We review de novo the district court’s conclusion that, as a matter of law, Tekmen was entitled to benefits under the terms of the plan.
The long-term disability plan provided that Reliance would pay long-term disability benefits if, inter alia, Tekmen “[was] Totally Disabled as the result of a Sickness or Injury covered by the Policy” and was able to “submit satisfactory proof of Total Disability to [Reliance].” J.A. 46. As relevant to Tekmen’s case, “Totally Disabled” was defined to mean that, “as a result of an Injury or Sickness[,] . . . during [a specified period of time starting on the first day of Total Disability] and for the first 24 months for which a Monthly Benefit is payable, you cannot perform the material duties of your Regular Occupation.” J.A. 38.
We agree with the district court that Tekmen was “Totally Disabled” and thus entitled to benefits. Tekmen submitted ample evidence demonstrating that she was totally
26
impaired under the terms of the plan, and we find Reliance’s arguments challenging various conclusions of the district court as to the plan’s meaning unavailing.
1.
Reliance first asserts that Gallagher v. Reliance Standard Life Insurance Co., 305 F.3d 264 (4th Cir. 2002), required Tekmen to submit “objective evidence” supporting her claim, suggesting that evidence such as MRI, CT, or EEG results supporting Tekmen’s claims would have been sufficient to establish disability but that self-reported symptoms were not. Gallagher created no such requirement.
The relevant plan language in Gallagher—mirroring the plan language at issue here—required the individual to submit “satisfactory proof of Total Disability.” Id. at 269. Reliance, in that case, asserted that the requirement of “satisfactory proof” granted Reliance discretion to determine benefit eligibility because it required Gallagher to submit “proof of his disability that is satisfactory to Reliance.” Id. A grant of discretion to Reliance would have required the district court to review the administrator’s decision for abuse of discretion rather than de novo. See id. at 269–70; see also Firestone, 489 U.S. at 115. This Court rejected that interpretation, concluding that the language of the plan did not require Gallagher to submit “proof of his disability that is satisfactory to Reliance,” but instead required “objectively satisfactory” proof. Gallagher, 305 F.3d at 269–70. Thus, the Gallagher Court was not called upon to distinguish between medically “objective” and “subjective” proof and consequently expressed no opinion on that question. Rather, the Court determined that proof that is satisfactory in an objective sense—rather than satisfactory to Reliance’s particular preferences—was required.
27
Further, a plan administrator may not require objective proof of disability if the plan does not contain such a requirement. See Cosey v. Prudential Ins. Co. of Am., 735 F.3d 161, 171 (4th Cir. 2013) (holding that because the relevant plans did not contain a requirement that Cosey submit objective proof, the district court erred in its determination that Prudential could deny Cosey’s claims on the basis that her proof lacked an objective component); DuPerry v. Life Ins. Co. of N. Am., 632 F.3d 860, 873 (4th Cir. 2011) (explaining that because the relevant policy “contained no provision precluding DuPerry from relying on her subjective complaints as part of her evidence of disability, [the defendant] could not reasonably deny her claim because of such reliance”).
Because the plan at issue here did not require objective proof of disability, we reject Reliance’s contention that Tekmen’s claim fails for the lack of such evidence.
2.
Reliance also argues that the district court failed to correctly apply the terms in the plan—specifically, the definition of “Regular Occupation”—in assessing Tekmen’s claim. We disagree.
The plan provides that “Total Disability” means, in relevant part, that “you cannot perform the material duties of your Regular Occupation” for a specified period of time. J.A. 38. It defines “Regular Occupation” to mean “the occupation you are routinely performing when Total Disability begins,” looking only at the occupation “as it is normally performed in the national economy, and not the unique duties performed for a specific employer or in a specific locale.” J.A. 37 (emphasis added).
28
Reliance argues that, because Tekmen claimed that the vibration in her new office in Tower 1 caused her symptoms, her disability only prevents her from working in that specific location and is therefore a locale-specific disability for which she cannot receive benefits. The district court, however, found that Tekmen’s move to the new office “did more than just make it impossible for her to work there, it essentially led to a significant relapse and consistent worsening of her various, underlying symptoms that first accrued after [her] accident in 2013.” J.A. 6581.
In support of its argument, Reliance cites Nichols v. Reliance Standard Life Insurance Co., 924 F.3d 802 (5th Cir. 2019), a case in which the underlying plan contained the same definition of “Regular Occupation” as is at issue here. Plaintiff Juanita Nichols, who worked at a chicken processing plant in which she was exposed to cold conditions, had to stop working because a circulatory condition rendered her unable to tolerate the cold. Id. at 805. The Fifth Circuit held that because Nichols was a sanitarian—a job that exists in both cold facilities and other facilities—Nichols’ disability was, indeed, location- specific. Id. at 810–11.
But here, ample evidence in the record demonstrates that Tekmen’s symptoms were improving prior to the move to Tower 1, that they drastically worsened immediately following her move to Tower 1, and that they continued to be severe and disabling after she was removed from Tower 1. And Tekmen’s symptoms “continued to worsen after she took a leave of absence from work and remained out of work into 2016.” J.A. 6581. Thus, Tekmen’s circumstance is unlike Nichols’s because the record does not show that Tekmen’s impairment was only present in response to the specific conditions of Tower 1.
29
As Dr. Damion concluded, the only environment in which Tekmen may not be impaired is one with “no ambient noise or vibration,” J.A. 4079, a setting that is unlikely to exist in a standard office building.
In short, the record supports the district court’s determination that Tekmen’s disability was not limited to a “specific locale.”
3.
Based on our de novo review of the district court’s legal conclusion that Tekmen was entitled to benefits, we agree that Tekmen was “totally disabled” under the terms of the plan.
As we have already explained, the district court did not err in crediting the opinions of the treating physicians—particularly Drs. Parker and Cintron—over those of non- treating physicians. Thus, given that the opinions of the treating physicians, taken as a whole, strongly supported Tekmen’s claims of non-location-specific impairment, and given that the lack of objective test results indicating disability cannot be fatal to Tekmen’s claim, the district court was correct to conclude that Tekmen could not perform the material duties of her Regular Occupation.
IV.
Based on the foregoing, we affirm the judgment of the district court.
AFFIRMED