Synthetic Intelligence has the potential to enormously enhance scientific care and help the transition to value-based care, but in addition comes with dangers. So what function ought to AI play in value-based care fashions? The query was posed final week throughout a panel dialogue at CES 2024 that was moderated by Dr. Jesse Ehrenfeld, president of the American Medical Affiliation.
Chris Jagmin, vice chairman and chief medical officer of state packages and medical coverage at CVS Well being’s Aetna, mentioned it’s vital to make use of AI in areas the place there may be little threat. This consists of scientific decision-making and approving one thing as an insurer. He added that AI presents alternatives in areas that “drive us loopy,” similar to automating name facilities or automating claims cost.
Nevertheless, “denying one thing based mostly upon AI with out human contact I feel is unsuitable for scientific,” Jagmin mentioned.
Dr. Maria Ansari, CEO and government director of The Permanente Medical Group, echoed Jagmin’s feedback, stressing the significance of getting a human contact when utilizing AI.
“It’s the doctor who’s treating the affected person, not AI,” Ansari mentioned. “AI [should be used] to enhance our work in scientific choice help instruments.”
She added that in a value-based care mannequin, AI ought to be used responsibly to foretell who wants a further degree of care, similar to who’s liable to having a stroke or a coronary heart assault.
That is particularly vital getting back from the Covid-19 pandemic since so many individuals weren’t coming in for preventive care, Ansari famous. For instance, she mentioned that The Permanente Medical Group had a two-year backlog of ladies who didn’t are available in for a mammogram throughout Covid-19.
“How do you prioritize that? … As an alternative of simply utilizing conventional threat components for who is perhaps in danger for breast most cancers, we used AI to have a look at their previous mammograms, which had been learn as regular, however picked up 1000 information factors that may really predict who’s going to go on to have most cancers and prioritized these sufferers,” Ansari mentioned. “That prediction mannequin turned out to be 60% extra correct than conventional threat components.
“I feel we are able to use AI after we discuss value-based care to enhance well being outcomes in the appropriate approach that’s accountable and never denying care, however really enhancing our care,” she continued.
The dialog got here after President Joe Biden signed an government order in October on the protected improvement and use of AI. In December, the White Home introduced that 28 suppliers and payers made commitments for accountable use of AI in healthcare. These organizations embrace CVS Well being, Mass Normal Brigham, Premera Blue Cross and Sanford Well being.
Photograph: metamorworks, Getty Photos
