Estimated Optimal Individualized Diabetes Risk Prediction From Preventive Interventions in the U.S. General Population

OBJECTIVE To calculate 3-year predicted risk for diabetes with a model including individualized preventive intervention effects for metformin therapy and intensive lifestyle among U.S. adults with prediabetes. RESEARCH DESIGN AND METHODS We included 2,778 participants with prediabetes from the National Health and Nutrition Examination Survey cycles 2015–2020. Using a validated type 2 diabetes risk prediction model, we calculated predicted risk and summarized the optimal prevention strategy (lowest predicted risk). RESULTS Mean predicted risk in this sample was 18.4% (95% CI 17.6, 19.3) for standard lifestyle recommendations (placebo), 14.4% (95% CI 13.9, 14.8) for metformin, 8.0% (95% CI 7.6, 8.4) for intensive lifestyle, and 7.6% (95% CI 7.2, 7.9) for participants’ respective optimal intervention. The optimal intervention strategy was intensive lifestyle for 91% of the analytic sample. CONCLUSIONS We demonstrate the potential population health benefits for diabetes prevention with use of a clinical decision tool, a diabetes risk prediction model with individualized preventive intervention effects.