ACCURATE prediction of Type 2 diabetes (T2D) risk is essential to enable early and effective prevention. However, existing models fail to tailor risk predictions based on the potential benefits of specific preventive interventions. To address this gap, researchers have developed and validated a novel risk prediction model that not only estimates the likelihood of developing T2D, but also identifies the most effective preventive strategy for each individual. Crucially, this is the first model of its kind to incorporate personalised intervention effects, offering a potential shift in how diabetes prevention is approached in clinical settings.
The study used data from 2,640 participants in the Diabetes Prevention Program (DPP) randomised to placebo, metformin, or intensive lifestyle intervention. A Cox proportional hazards regression model was created using common clinical predictors: sex, HbA1c, fasting plasma glucose (FPG), BMI, triglycerides, and intervention group. To individualise preventive strategies, the model included interactions between intervention type and patient-specific variables such as age, FPG, and BMI. Validation was conducted internally within the DPP cohort and externally using data from 2,104 individuals with prediabetes in the Multi-Ethnic Study of Atherosclerosis (MESA).
In the DPP cohort, the model achieved a C-statistic of 0.71 (95% CI: 0.68–0.74), while in the external MESA cohort, discrimination improved substantially with a C-statistic of 0.86 (95% CI: 0.83–0.88). The model suggested that intensive lifestyle intervention would be the optimal preventive intervention (giving the lowest 3 year risk) for 86% of DPP participants and 97% of those in MESA, while metformin was preferred for the remainder. Notably, the model demonstrated consistent performance across racial and ethnic groups, supporting its generalisability.
This study marks a significant step forward in the personalised prevention of T2D. By identifying which individuals are most likely to benefit from lifestyle changes versus metformin, the model could refine clinical decision-making and enhance patient outcomes. Limitations include reliance on trial and observational data, and the need for further validation in real-world settings. Nonetheless, the integration of individualised intervention recommendations represents a valuable addition to diabetes prevention strategies and offers a practical tool for use in routine primary care.
Reference
Jaeger et al. Development and validation of a diabetes risk prediction model with individualized preventive intervention effects. J Clin Endocrinol Metab. 2025;DOI: 10.1210/clinem/dgaf250.
