An AI-Powered Lifestyle Intervention vs Human Coaching in the Diabetes Prevention Program: A Randomized Clinical Trial
Primary result · Composite primary outcome achievedResult — go to the resultRisk difference −0.2% 1-sided 95% CI −8.2%Authors' conclusion
Among adults with prediabetes and overweight or obesity, referral to a fully automated AI-led DPP was noninferior to referral to a human-led DPP in achieving a composite outcome based on weight reduction, physical activity, and HbA1c.
Study context
- Population
- Adults ≥18 years with prediabetes and overweight or obesity; median age 58 years, 71% female, median BMI 32.3randomized 1:1
- Participants
- 368 adults
- Comparison
- Referral to artificial intelligence (AI)-led Diabetes Prevention Program (DPP) vs referral to human-led DPP
- AI-led DPPn = 183
- Human-led DPPn = 185reference
- Primary endpoint
- Composite:
- Hemoglobin A1c (HbA1c) <6.5% throughout and ≥5% weight loss
- ≥4% weight loss plus ≥150 min weekly physical activity (actigraphy)
- Absolute HbA1c reduction ≥0.2 percentage points
- Follow-up
- 12 months
Methodological detailsphase 3, parallel-group, pragmatic, noninferiority, 2 US clinical sites, Conducted October 11, 2021, to December 16, 2024 at 2 US sites (Baltimore, Maryland; Reading, Pennsylvania); interventions delivered independently of the study team
- Phase 3
- Parallel-group
- Pragmatic
- Noninferiority
- 2 US clinical sites
- Conducted October 11, 2021, to December 16, 2024 at 2 US sites (Baltimore, Maryland; Reading, Pennsylvania); interventions delivered independently of the study team
Primary result
Composite primary outcome achieved
Risk difference
−0.2%
1-sided 95% CI −8.2%
Events by treatment group
AI-led DPPvs Human-led DPP
risk difference −0.2% (1-sided 95% CI −8.2%)
As reported in the abstract
Further findings
Initiation of the referred DPP
Individual composite components and sensitivity analyses
- Findings consistent across individual components of the composite and in sensitivity analyses
Original publication
An AI-Powered Lifestyle Intervention vs Human Coaching in the Diabetes Prevention Program: A Randomized Clinical Trial. JAMA. Published 16 December 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →