Mediterranean diet and diabetes risk in a cohort study of individuals with prediabetes: propensity score analyses

Aims Randomized controlled trials have demonstrated the efficacy of several dietary patterns plus physical activity to reduce diabetes onset in people with prediabetes. However, there is no evidence on the effect from the Mediterranean diet on the progression from prediabetes to diabetes. We aimed to evaluate the effect from high adherence to Mediterranean diet on the risk of diabetes in individuals with prediabetes. Methods Prospective cohort study in Spanish Primary Care setting. A total of 1184 participants with prediabetes based on levels of fasting plasma glucose and/or glycated hemoglobin were followed up for a mean of 4.2 years. A total of 210 participants developed diabetes type 2 during the follow up. Hazard ratios of diabetes onset were estimated by Cox proportional regression models associated to high versus low/medium adherence to Mediterranean diet. Different propensity score methods were used to control for potential confounders. Results Incidence rate of diabetes in participants with high versus low/medium adherence to Mediterranean diet was 2.9 versus 4.8 per 100 persons-years. The hazard ratios adjusted for propensity score and by inverse probability weighting (IPW) had identical magnitude: 0.63 (95% confidence interval, 0.43–0.93). The hazard ratio in the adjusted model using propensity score matching 1:2 was 0.56 (95% confidence interval, 0.37–0.84). Conclusions These propensity score analyses suggest that high adherence to Mediterranean diet reduces diabetes risk in people with prediabetes.