Most AI medical devices cleared for use were not tested on patient outcomes
A new analysis shows that, of 1,357 artificial intelligence (AI)-based medical devices authorized by the U.S. Food and Drug Administration (FDA) for use in patient care, only three had been tested on whether they actually improve patients' health.
A recent study published in PLOS Digital Health reveals that only three out of 1,357 AI-based medical devices cleared for use by the U.S. FDA have been tested to assess their impact on patient health outcomes. The research, conducted by Rawan Abulibdeh and colleagues from the University of Toronto, highlights significant concerns about the effectiveness and equity of AI devices in healthcare.
Currently, AI devices typically only need to demonstrate substantial equivalence to an existing device, without requiring proof that they improve patient health or benefit diverse populations. The study found that while 34 devices underwent registered clinical trials, only three tested patient-centered outcomes like mortality, stroke rates, hospitalizations, and quality of life.
Most studies were conducted in affluent healthcare systems and excluded crucial patient groups such as pregnant women, adults over 75, and non-English speakers.
The researchers argue that structural barriers, including financial incentives and logistical challenges, often prioritize rapid development over rigorous testing. They warn that this approach could exacerbate healthcare disparities, potentially exposing low- and middle-income countries to untested AI tools. The study concludes that current policies for AI medical device authorization must be revised to ensure devices are not just similar to existing ones but genuinely enhance patient health across various subgroups and settings.
Written by urgent.news from Medical Xpress's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
