#2-40E. AI vs. Human Coaching: Noninferiority of a Fully Automated Digital Diabetes Prevention Program (JAMA, 2025)

Prediabetes affects approximately one-third of US adults, yet adoption of the highly effective Diabetes Prevention Program (DPP) remains critically low due to barriers in accessibility and availability. Can fully automated technology offer a scalable solution comparable to traditional coaching? This phase 3, pragmatic, noninferiority randomized clinical trial (RCT) compared an exclusively AI-led DPP, delivered via mobile app, against referral to a human coach-led DPP over 12 months among adults with prediabetes and overweight/obesity. The study found that referral to the AI-led DPP was noninferior to the human-led DPP in achieving a composite outcome based on recommended thresholds for weight loss, physical activity, and HbA1c reduction. This result is significant as it validates a scalable, low-touch, on-demand model for prevention. However, its clinical significance must be critically weighed against the limitation that the trial relied on surrogate outcomes (weight and HbA1c metrics) rather than the long-term incidence of type 2 diabetes1011. For busy clinicians, this research is crucial: the AI-led approach offers a validated, accessible pathway to expand evidence-based lifestyle intervention to the large population of patients who currently receive no structured prevention care.

Citation: Mathioudakis N, Lalani B, Abusamaan MS, et al. An AI-Powered Lifestyle Intervention vs Human Coaching in the Diabetes Prevention Program. JAMA. 2025. DOI: 10.1001/jama.2025.19563

Disclaimer: This audio summary is based on personal interpretation and does not guarantee the exact content of the original paper. Please refer to the original article for details.