ERISA De Novo Review and the Burden of Proof: California Court Affirms Unum’s Termination of Long-Term Disability Benefits

In Syed v. Unum Life Insurance Company of America, No. CV 25-01052-MWF (CTSx), 2026 WL 2807048 (C.D. Cal. Sept. 18, 2026), United States District Judge Michael W. Fitzgerald affirmed Unum’s termination of a corporate lawyer’s long-term disability benefits under an ERISA-governed group policy and entered judgment for the insurer. Reviewing the claim de novo by stipulation, the court held that Plaintiff failed to prove by a preponderance of the evidence that she remained disabled from her usual occupation when benefits were terminated. For anyone with an ERISA group disability claim, the decision is a reminder that on de novo review the claimant carries the burden of proof, and that a diagnosis alone does not establish disability.

What standard of review and burden of proof applied?

The parties stipulated to de novo review, meaning the court gave no deference to Unum’s decision and made its own merits determination on the administrative record. Under that standard, the claimant bears the burden of proving entitlement to benefits, and must show she was more likely than not disabled under the terms of the policy. The policy defined total disability as an inability to perform with reasonable continuity the substantial and material acts necessary to pursue the claimant’s usual occupation in the usual and customary way. Because the court proceeded under Federal Rule of Civil Procedure 52, it did not ask whether a factual dispute existed but instead weighed the conflicting evidence and decided which was more likely true.

How did the court weigh the contemporaneous medical record?

The court began with the evidence closest in time to the April 2024 termination, reasoning that a physician’s opinion is more credible when supported by contemporaneous functional limitations. Plaintiff’s initial claim identified major depressive disorder and generalized anxiety disorder as the disabling conditions, and her early treatment records documented significant depression and anxiety while repeatedly connecting her symptoms to workplace stress and personal and family circumstances. The court found that the contemporaneous records reflected stabilization and improvement: Plaintiff reported decreased depression in December 2023, reduced sadness and irritability in January 2024, and stable and manageable symptoms in February and March 2024, while her nurse practitioner recorded fair concentration, adequately managed attention, and largely unremarkable mental status examination findings. Unum’s reviewing psychiatrists concluded that this record was inconsistent with continued work-preclusive impairment, and the court found their conclusions persuasive.

Did the court give less weight to the file-reviewing physicians?

No. Plaintiff argued that Unum’s consultants should be discounted because they conducted file reviews rather than in-person examinations. The court rejected that argument, noting that courts are not obliged to accord special deference to examining physicians over non-examining physicians. The court observed that the rule had particular force here because much of Plaintiff’s own evidence came from providers who could not meaningfully be described as treating providers, including a neurologist who based his opinion on a telehealth interview and record review and a physician who did not appear to have personally examined Plaintiff.

Why did the later diagnoses of POTS and chronic fatigue syndrome not change the outcome?

On appeal to the insurer, Plaintiff reframed her claim as depression and anxiety secondary to dysautonomia and POTS, supported by an independent neurological evaluation and opinions from her cardiologist and primary care physician. The court held that these materials did not alter the result. While later-identified diagnoses may prove or explain earlier disability, the bare fact of a diagnosis does not by itself establish disability; the claimant must connect the diagnoses to actual functional limitations that rendered her disabled under the policy. The court found the neurologist’s report had limited persuasive force because he did not reconcile his POTS conclusion with the administering electrophysiologist’s description of the tilt table results as at the upper limit of normal, and did not appear to have considered the contemporaneous evidence that Plaintiff’s leave resulted from a particular mental health episode. The cardiologist likewise reported normal stress test and echocardiogram results and agreed the tilt table results were at the upper limits of normal, leaving her disability opinion resting largely on Plaintiff’s subjective descriptions.

How did the court treat the treating providers and the lay statements?

The court declined to categorically discount Unum’s reviewers, and equally declined to overly credit Plaintiff’s treating providers, explaining that a treating physician’s opinion can be discounted when it lacks supportive evidence, is contradicted by other assessments, or rests on subjective descriptions of limitations. It found that neither the therapist’s records, the nurse practitioner’s later opinion, nor the clinical social worker’s statements provided specific observations, functional assessments, or mental status examination findings demonstrating that Plaintiff could not perform corporate legal work as of April 2024. The narrative statements from Plaintiff, her sister, and her friend supported that she experienced symptoms including social withdrawal, and reduced the force of Unum’s reliance on her attendance at certain social gatherings, but the court held such lay statements present a potential for bias and cannot, on their own, assess work-preclusive functional impairment when weighed against a record of unremarkable examination findings.

What happened to the motion to exclude the insurer’s arguments?

Plaintiff moved to exclude certain defense arguments as new rationales barred under Collier v. Lincoln Life Assurance Co. of Boston, which prohibits a district court from adopting a newly presented rationale the insurer did not raise during the administrative process. The court denied the motion, finding it had not relied on any new rationale. The arguments on which it relied, including the lack of substantiated functional restrictions, the largely unremarkable test results, the documented improvement in symptoms, and the degree to which Plaintiff’s later opinions depended on subjective reports, were either identified in Unum’s denial letters or were subsidiary to those rationales, and to the extent Plaintiff identified other arguments, the court had not relied on them.

What was the outcome?

Applying the policy standard, the court held that the record did not establish by a preponderance of the evidence that Plaintiff had a disability preventing her from performing the substantial and material acts of her occupation as of April 2024. Unum’s decision to terminate benefits was affirmed, judgment was entered in favor of the insurer, and the motion to exclude was denied.