|
| | | |  |
| By Ben Leonard, Daniel Payne and Erin Schumaker | | | | |
Leading House Republicans want HHS to back away from labs that vet AI tools. | Francis Chung/POLITICO |
Leading House Republicans are taking a hard look at plans to regulate artificial intelligence in health care. They want HHS to back away from assurance labs intended to vet AI tools, our Ben Leonard reports. The details: Reps. Dan Crenshaw (R-Texas); Brett Guthrie (R-Ky.), who is running to chair the Energy and Commerce Committee; Jay Obernolte (R-Calif.); and Mariannette Miller-Meeks (R-Iowa) say the agency must stop any plans to create “government-administered assurance labs” and “walk back support” for a model for such labs from the Coalition for Health AI. Micky Tripathi, HHS’ assistant secretary for technology policy, was an adviser to CHAI and has signaled support for the group’s plan, which includes efforts to evaluate models based on “consensus-driven standards.” CHAI — whose founding partners included Google, Microsoft, Stanford Medicine and the Mayo Clinic — has said the labs would not be a part of the government regulatory processes.
“We recognize that you were a board observer for CHAI and ask that you help us understand how putting the organization directly in control of market entry for innovative technologies does not represent a significant conflict of interest,” the lawmakers wrote in a letter to HHS. “These types of blurred lines are exactly what foster distrust in government and big tech.”
Under CHAI’s plan, labs would adhere to standards for testing laboratories as defined by the International Organization for Standardization, a Switzerland-based nongovernmental group that promotes tech platforms working across borders. CHAI also unveiled standards for AI transparency in line with HHS’ requirements for firms seeking government certification. HHS has eyed a network of assurance labs and an AI safety program as part of its broader strategy. HHS and CHAI didn’t respond to requests for comment.
The bigger picture: Deregulation is a top priority for the incoming Trump administration and House GOP leaders have been skeptical of new legislation that would further regulate AI. Trump is expected to rescind President Joe Biden’s executive order promising a coordinated approach to AI.
| | | |
Saratoga Springs, N.Y. | Erin Schumaker/POLITICO |
This is where we explore the ideas and innovators shaping health care. Scientists are recruiting and tagging dogs, goats and other farmyard animals to research whether they can predict earthquakes, The Guardian reports. Share any thoughts, news, tips and feedback with Carmen Paun at cpaun@politico.com, Daniel Payne at dpayne@politico.com, Ruth Reader at rreader@politico.com, or Erin Schumaker at eschumaker@politico.com. Send tips securely through SecureDrop, Signal, Telegram or WhatsApp.
| | | | REGISTER NOW: As the 118th Congress ends, major decisions loom, including healthcare appropriations. Key focus: site neutrality. Can aligning hospital and clinic costs cut federal spending, reflect physician costs, and lower patient expenses? Join policymakers and providers to discuss. | | | | | | |
| |
 AI tools work best for doctors who use the tools frequently or don't see many patients. | Getty Images |
Clinicians who use tools with artificial intelligence might not perform more efficiently than those who don’t use AI tools. At least that’s the takeaway from researchers who followed more than 200 primary care clinicians for 180 days. Half of the clinicians used AI-driven software to help write their clinical documentation, while the other half didn’t. The study, published in NEJM AI and conducted at Atrium Health, a large academic health system, found some modest gains in efficiency. The AI tool typically worked best for doctors who used the tool frequently or saw a low volume of patients. Family doctors also saw some gains from the tool. Small or no efficiency gains from AI tools could pose significant problems for providers who invest heavily in the technology, believing it will save time for clinicians and labor costs for health systems. Many health executives prioritize potential cost savings when evaluating AI tools to invest in. But many variables could change outcomes. The market offers a large number of AI tools, each with different capabilities, levels of integration into existing workflows and clinician acceptance. Some health system leaders have said AI tools are, for now, more about reducing the clinicians’ mental loads, which could reduce burnout, though some early evidence suggests AI might not reduce burnout. Further research is recommended to explore the findings and improve clinical adoption.
| | | | Want to know what's really happening with Congress's make-or-break spending fights? Get daily insider analysis of Hill negotiations, funding deadlines, and breaking developments—free in your inbox with Inside Congress. Subscribe now. | | | | | | |
| |
 Plummeting cervical cancer mortality rates offer a compelling case for HPV vaccination efforts. | AFP via Getty Images | The mortality rate for young women with cervical cancer has dropped significantly in the U.S., according to new research in JAMA. It’s dropping even faster than projected for women under 25 in recent years, the researchers said, thanks to vaccines. Those who were part of that group from 2016 to 2021 — when the death rate dropped below the projections — were the first to have broad vaccination against human papillomavirus, which can cause cervical cancer. And it’s not just mortality rates. The incidence of cervical cancer itself has dropped among young women in recent years. Why it matters: The reduction in cancer cases and deaths offers a compelling case for wide-reaching vaccination efforts, even as some health leaders for the incoming Trump administration present new challenges to those efforts. | | |
| Follow us on Twitter | | | | Follow us
| | | | |
