UnitedHealthcare sued for using faulty AI algorithm to deny claims
UnitedHealthcare was sued in November 2023 by two families alleging the company used a faulty AI algorithm with a reported 90% error rate to deny insurance claims that doctors had approved as medically necessary. A 2024 Senate investigation found the company's denial rate for post-acute care coverage more than doubled between 2020 and 2022, coinciding with automation initiatives.
What happened
In November 2023, a lawsuit filed by two families accused UnitedHealthcare of using an AI algorithm developed by NaviHealth to deny coverage for patients despite doctors approving the care as necessary. Both families had a family member die, and the lawsuit alleged that UnitedHealthcare knew the AI tool had a '90% error rate.'
UnitedHealthcare defended the tool, stating it is not used to make coverage determinations but instead serves as a guide to inform providers and families about potential care needs. A U.S. Senate Permanent Subcommittee on Investigations report from October 2024 found that UnitedHealthcare's prior authorization denial rate for post-acute care jumped from 10.9% in 2020 to 22.7% in 2022, with denied claims in 2022 nine times higher than in 2019, during a period when the company implemented multiple initiatives to automate the process.
Reported impact
The lawsuit alleged that the scheme provided UnitedHealthcare a financial windfall in the form of policy premiums without having to pay for promised care. The lawsuit claimed that elderly patients were prematurely kicked out of care facilities nationwide or forced to deplete family savings to continue receiving necessary medical care because the AI model disagreed with their doctors' determinations.
UnitedHealthcare's denial rate for post-acute care was reportedly nearly twice as high as other insurance companies, according to reports cited by Forbes. Some hospital systems have stopped accepting the company's insurance coverage.
Index analysis
The AI algorithm reportedly failed to align with medical professionals' coverage determinations, resulting in wrongful claim denials. The dramatic increase in denial rates coinciding with automation initiatives suggests the algorithm's deployment amplified claim rejections.
Governance controls
UnitedHealthcare stated that the AI tool was designed as a guide to inform stakeholders rather than make final coverage decisions, suggesting internal governance intended to limit the algorithm's role. The lawsuit remained ongoing as of the tracker date, and no official finding has been confirmed regarding UnitedHealthcare's knowledge of the tool's error rate or intentional misuse.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.