The main goal of the study was to assess whether referral to an entirely AI-run DPP lifestyle program was noninferior to referral to an entirely human-run DPP in terms of improving weight loss, glycemia, and physical activity in adults with prediabetes.
This was a parallel group, pragmatic, noninferiority randomized controlled clinical trial performed from October 11, 2021, to December 16, 2024, at two research sites in the United States: in Baltimore, Maryland, and in Reading, Pennsylvania. Patients were recruited for this study when they were at least 18 years of age with either prediabetes and overweight or obese.
A total of 368 participants were included in the analysis. The median age was 58 years (interquartile range [IQR], 50-65 years), and the participants were comprised of 71% female, 27% Black, 6% Hispanic, and 61% White individuals. The median body mass index (BMI) was 32.3 (IQR, 28.5-37.1). The participants were equally randomly assigned to a study group undergoing an AI-driven DPP through a mobile app using a digital scale connected by Bluetooth or a remote human coach-led DPP. The study groups were both independent of the research group.
Key Findings
The rate of initiation of the AI-led DPP was 93.4% (171 of 183) compared with 82.7% (153 of 185) for the human-led DPP.
The rate of achieving the composite primary outcome was 31.7% (58 of 183) for the AI-led DPP and 31.9% (59 of 185) for the human-led DPP.
The risk difference was -0.2% with a one-sided 95% CI of -8.2%, which met the criterion for noninferiority.
A composite endpoint measured at 12 months and defined as the maintenance of hemoglobin A1c (HbA1c) values of less than 6.5% during the entire study period and either of the following: at least 5% of absolute weight loss; at least 4% of absolute weight loss plus at least 150 minutes of physical activity per week; and/or an absolute decrease of at least 0.2% of HbA1c values in comparison with baseline was the primary outcome.
Noninferiority had been prespecified with the lower bound of the one-sided 95% confidence limit of the risk difference no longer than −15%.
In a group of adults with prediabetes and overweight or obesity, referral to a completely automated AI-powered Diabetes Prevention Program was found to be noninferior to a human-conducting program for achieving significant improvement in weight, physical activity, and glycemic control. These findings establish the inclusion of AI-driven lifestyle interventions in the panoply of strategies for the prevention of diabetes in the general public.
Reference:
Mathioudakis N, Lalani B, Abusamaan MS, et al. An AI-Powered Lifestyle Intervention vs Human Coaching in the Diabetes Prevention Program: A Randomized Clinical Trial. JAMA. 2025;334(23):2079–2089. doi:10.1001/jama.2025.19563
Disclaimer: This website is primarily for healthcare professionals. The content here does not replace medical advice and should not be used as medical, diagnostic, endorsement, treatment, or prescription advice. Medical science evolves rapidly, and we strive to keep our information current. If you find any discrepancies, please contact us at corrections@medicaldialogues.in. Read our Correction Policy here. Nothing here should be used as a substitute for medical advice, diagnosis, or treatment. We do not endorse any healthcare advice that contradicts a physician's guidance. Use of this site is subject to our Terms of Use, Privacy Policy, and Advertisement Policy. For more details, read our Full Disclaimer here.
NOTE: Join us in combating medical misinformation. If you encounter a questionable health, medical, or medical education claim, email us at factcheck@medicaldialogues.in for evaluation.