The American Medical Association is raising concerns about the use of artificial intelligence to review claims for prior authorization.
In a resolution passed at the association's annual meeting, delegates directed the association to advocate for more government oversight of AI review of patient claims.
According to a June 14 news release, the AMA wants AI prior authorization to:
- Be based on accurate clinical criteria.
- Include review by physicians not incentivized to deny care.
- Require human review before denying claims.
In the resolution, AMA members cited a March report from ProPublica that found Cigna physicians denied large batches of claims without reviewing them. The internal system is not used for prior authorization, and is used to review claims after care has been provided.
A Cigna spokesperson told Becker's the report is a "mischaracterization of our process for accelerating payment of claims for routine, low-cost screenings."
"As health insurance companies increasingly rely on AI as a more economical way to conduct prior authorization reviews, the sheer volume of prior authorization requirements continues to be a massive burden for physicians and creates significant barriers to care for patients," AMA trustee Marilyn Heine, MD, said in the news release. "The bottom line remains the same: We must reduce the number of things that are subject to prior authorization."
Federal lawmakers have also raised concerns about the use of AI in prior authorizations.
"If insurance companies are getting bigger and buying companies that specialize in developing algorithms, it strikes me that they are going to be in a position to invest in new ways to deny care," Senate Finance Committee Chair Ron Wyden said during a June 8 hearing.