BACKGROUND:Higher consumption of some polyphenols has been associated with a reduced risk of diabetes. However, no studies have evaluated the relation between all polyphenol subclasses and the incidence of diabetes. OBJECTIVE:We aimed to prospectively examine the associations between the intake of total polyphenols and different groups of polyphenols (flavonoids, phenolic acids, stilbenes, lignans, and others) on the risk of incident diabetes in the PREDIMED (Prevención con Dieta Mediterránea) trial. METHODS:This was an observational cohort analysis of the nondiabetic participants in the PREDIMED trial. This study was a multicenter, controlled, randomized, parallel-group feeding trial to assess the effects of either a Mediterranean diet that was supplemented with extra-virgin olive oil or nuts or advice to adhere to a low-fat control diet on cardiovascular outcomes in elderly men and women at high cardiovascular disease risk. From the 7447 randomly assigned participants, 3430 were selected because they were free of diabetes at baseline and filled out the food-frequency questionnaires (FFQs). Polyphenol intake was calculated by matching food consumption data from repeated FFQs with the Phenol-Explorer database on the polyphenol content of each reported food. HRs and 95% CIs for diabetes according to tertiles of polyphenol intake were estimated with the use of time-dependent Cox proportional hazards models. RESULTS:Over a mean of 5.51 y of follow-up (18,900 person-years), there were 314 new cases of diabetes. After multivariable adjustment, we observed a 28% reduction in new-onset diabetes in the highest compared with the lowest tertile of total polyphenol intake (HR: 0.72; 95% CI: 0.52, 0.99; P-trend = 0.05). The intake of subclasses of polyphenols also was inversely associated with diabetes risk, including for total flavonoids (HR: 0.67; 95% CI: 0.48, 0.93; P-trend = 0.02), stilbenes (HR: 0.57; 95% CI: 0.38, 0.84; P-trend = 0.003), dihydroflavonols (HR: 0.59; 95% CI: 0.40, 0.88; P-trend = 0.003), and flavanones (HR: 0.69; 95% CI: 0.49, 0.97; P-trend = 0.03). CONCLUSIONS:A high intake of total polyphenols, total flavonoids (specifically flavanones and dihydroflavonols), and stilbenes is associated with a reduced risk of diabetes in elderly persons at high risk of cardiovascular disease. This trial was registered at http://www.controlled-trials.com as ISRCTN35739639.