Amid Lawsuits, Regulatory Scrutiny, AI Is Risky Business for Medicare Advantage Plans

With the filing of a proposed class action lawsuit this month, Humana Inc. became the third major insurer in recent history to be accused of using artificial intelligence to wrongfully deny patients’ care and the second insurer to face allegations specific to Medicare Advantage members. While industry experts agree that AI holds promise for improving the patient experience, it also comes with risks, and lawsuits and other regulatory actions offer a warning to insurers of all types to come up with a proper risk mitigation strategy as they increasingly deploy AI to streamline certain operations.  

In the Dec. 12 complaint, which was filed in the U.S. District Court for the Western District of Kentucky, Humana MA members accuse the insurer of relying on the nH Predict AI model to make “rigid and unrealistic” projections for how long a patient will require post-acute care after an inpatient hospital stay. The AI model was developed by naviHealth, a subsidiary of UnitedHealth Group, and was the subject of a highly critical investigation published by STAT in November and subsequent lawsuit filed against UnitedHealth by the estates of two deceased MA members.