Skip to content
Benefits

Challenge to Use of Automated Algorithm in Benefit Decisions Allowed to Proceed

EBIA  

· 5 minute read

EBIA  

· 5 minute read

Kisting-Leung v. Cigna Corp., 2025 WL 958389 (E.D. Cal. 2025)

A group of participants in different employer-sponsored health plans brought a proposed class action lawsuit against the plans’ common claims administrator for wrongful denial of benefits and breach of fiduciary duty due to the administrator’s use of an automated algorithm to process benefit claims. The participants provided examples of benefit claims that they said should have been covered under the terms of the respective plans but were denied as not medically necessary or for other reasons such as coding errors. The administrator asked the court to dismiss the claims.

With respect to the claims for wrongful benefit denials, the court concluded that the participants had not specifically identified plan terms that entitled them to benefits. For example, some participants relied on provisions in the administrator’s medical coverage guidelines, but the court pointed out that the coverage guidelines were not part of the respective plans and did not necessarily mirror plan terms. The court dismissed these claims but gave the participants permission to amend them for the court’s reconsideration. In support of their fiduciary breach claims, the participants pointed out that the administrator’s policies indicated that medical necessity determinations would be made by a “medical director” but that such determinations were in fact made by the automated algorithm. The administrator argued that it was acting within its discretionary authority in interpreting plan terms to permit use of the algorithm, but the court concluded that this interpretation was an abuse of discretion. Because several participants had presented arguments sufficient to support their assertion that entrusting medical necessity decisions to the automated algorithm violated plan terms, the court allowed their fiduciary breach claims to proceed. Other fiduciary breach claims were dismissed because the administrator had presented evidence that the algorithm was not used for the benefit decisions in question, but the court granted the affected participants permission to amend their claims to add evidence to the contrary. The court also rejected the administrator’s argument that the participants could not pursue both their wrongful denial claims and fiduciary breach claims, explaining that the claims were not duplicative because the remedies—payment of unpaid benefits (for wrongful denials) and changes to the administrator’s claims handling practices (for fiduciary breaches)—would be distinct.

EBIA Comment: It seems likely that the participants will amend their claims and refile to address the issues raised by the court. Although this ruling is generally preliminary, plan sponsors will want to watch for a decision on the merits of using automated algorithms in benefit plan decision making. For more information, see EBIA’s ERISA Compliance manual at Sections XXXVI.I (“What Relief Is Available on Successful Benefit Claims?”), XXVIII.I (“Fiduciary Liability and Litigation”) and XXXIV.N.2 (“Factors and Rules to Apply When Denying Benefit Claims”). See also EBIA’s Self-Insured Health Plans manual at Section XXVI.J (“Litigation Issues”).

 

Take your tax and accounting research to the next level with Checkpoint Edge and CoCounsel. Get instant access to AI-assisted research, expert-approved answers, and cutting-edge tools like Advisory Maps and State Charts. Try it today and transform the way you work! Subscribe now and discover a smarter way to find answers.

More answers