Nestoras Mathioudakis, MD, MHS, a BIDS faculty member, is the first author of an article recently published in JAMA: “An AI-Powered Lifestyle Intervention vs. Human Coaching in the Diabetes Prevention Program.” This study examined how referral to a lifestyle intervention exclusively driven by AI compares with referral to a human coach-led diabetes prevention program. The authors concluded that among adults with prediabetes and overweight or obesity, referral to a fully automated AI-led diabetes prevention program was noninferior to referral to a human-led diabetes prevention program in achieving a composite outcome based on weight reduction, physical activity, and hemoglobin A1C.
When asked about AI and clinical care, Dr. Mathioudakis responded, “Patient-facing AI tools, from digital health coaches to symptom checkers and clinical question-and-answer resources, are rapidly expanding. Our randomized trial is among the first to directly compare a fully automated, patient-facing coach for diabetes prevention with the established, human coach-led Diabetes Prevention Program (DPP). Given that more than one in three U.S. adults have prediabetes and fewer than 1% currently access human-led DPP, our findings suggest AI-based digital coaches could offer a scalable, lower-cost way to help address this burden. Still, larger dissemination studies are needed to understand patient experience, trust, and engagement with AI-based programs in real-world settings and to determine their impact on diabetes risk reduction.”
Dr. Mathioudakis’ studies like this one offer valuable insights into using health informatics for diabetes prevention and management and integrating machine learning and AI in diabetes research.