A report on artificial intelligence from conservative think tank Paragon Health Institute could serve as a roadmap for healthcare AI policy for the incoming Trump administration; and the author is coming out with a second paper, focused specifically on regulation, next month.
In July of this year, PHI released a policy paper on the costs and benefits of using artificial intelligence in healthcare, written by visiting fellow Kev Coleman. The paper touted the potential cost savings of AI through productivity gains, quality improvements and autonomous care delivery.
Coleman has a background in software programming, software startups and health insurance. He was VP of product management at Experion for a decade, founded the research and data division at HealthPocket and served as the president of Association Health Plans for the last five years.
His July paper made policy recommendations such as allowing some trusted AI solutions to operate without a human in the loop, to save money.
“Every time a medical service can be performed independently of a clinician, a sizable expense within health care delivery can be eliminated,” Coleman wrote in his first paper. “Because medical devices aren’t licensed by individual states, there is also the possibility of AI-enabled health care devices competing for patients across the nation, giving alternatives to regions dominated by a consolidated hospital system and lowering prices through market competition.”
Because of the positive response of the first paper, Coleman told Fierce Healthcare he is writing a second paper on more specific guidelines for the regulation of artificial intelligence. The paper is expected to be released in early December, he said.
He hopes it will be considered for use by the incoming Trump administration.
“There's no kind of rhetorical exaggeration, political bomb-throwing, but instead a recent approach that's trying to reflect many years in software as well as healthcare research,” Coleman told Fierce Healthcare in an interview.
Paragon Health Institute was founded in 2021 by Brian Blase, who served as special assistant at the White House National Economic Council from 2017-2019, under then-President Donald Trump. Paragon is a non-partisan think tank that promotes policy geared toward free-market solutions, innovation, patient choice and competition.
Other health policy experts who served in the Trump administration also helped found the Institute, like Demetrios Kouzoukas, the chief executive of the Medicare program at the Centers for Medicare and Medicaid Services from 2017 through 2021, and Joel Zinberg, the general counsel and a senior economist at the White House Council of Economic Advisers, 2017-2019.
In his upcoming paper, Coleman will recommend that the Trump administration, and the 119th Congress, approach AI regulation at a granular level. He said that regulation cannot broadly refer to “healthcare AI,” because AI is not a monolithic technology. Being too broad could inadvertently affect other software and create unintended consequences for the industry.
“What I'm trying to do is empower rule makers and policymakers to approach AI in a let's call it a meaningful engagement, and by that, by providing the resources to understand what are the basic tools within AI and what are the continuities between traditional software as well as traditional statistics and mathematical modeling, and the things that are really uniquely expressed in contemporary AI software,” Coleman said.
Instead, he recommends that the agencies use specific language to regulate specific types of AI, like large language models (LLM), generative AI and advanced neural networks (ANN). Coleman recommends that the operational definitions of these technologies in regulation be very specific.
Jodi Daniel, former director of the office of policy in the Office of the National Coordinator for Health Information Technology (ONC), now lead of the digital health practice at Crowell & Moring, told Fierce Healthcare she expects president-elect Trump to vacate President Biden’s October 2023 Executive Order on AI in favor of his own approach.
Coleman agrees, saying he expects the incoming administration to rely on a "distributed approach" across the agencies.
He advocates that the federal agencies are appropriately staffed with AI experts. In his July paper, he wrote that the FDA should hire more AI experts. He also advocates a risk-based approach to AI, so that algorithms with a greater potential to harm patients would be more scrutinized.
But Coleman said that the Trump administration is likely to let the agencies and offices under HHS stay in their own lanes with AI regulation, according to their congressional authority. He expressed skepticism that a Trump administration would appoint a Chief AI Officer within HHS because the agency’s purview has such a wide range.
Instead, he expects that AI developments within HHS will be siloed between agencies like the Food and Drug Administration and the Centers for Medicare and Medicaid. Coleman clarified he was not speaking on behalf of the incoming Trump administration.
He is also a proponent of enforcing existing regulations on data privacy and non-discrimination in healthcare and urges the incoming administration to avoid duplicative laws or regulations.
Coleman said he doesn’t expect Congress to step in on AI unless state laws become overly burdensome.