AIMS: To assess the effect of fibric acid derivative bezafibrate on incidence of type 2 diabetes in obese patients over a median 6.3 years follow-up period. METHODS AND RESULTS: The study sample comprised 339 non-diabetic obese patients (body mass index >/=30.0 kg/m(2)) aged 42-74. Patients received either bezafibrate retard 400 mg (178 patients) or placebo (161 patients) once daily. Development of new diabetes was recorded in 98 patients: in 56 (37.0%) from the placebo group vs. 42 (27.1%) from the bezafibrate group, (P log-rank=0.01). The median time (interquartile range) until onset of new diabetes was significantly delayed in patients on bezafibrate when compared with those on placebo: 4.0 (2.1-5.0) vs. 2.0 (0.5-3.5) years, P=0.002. Multivariable analysis identified bezafibrate treatment as an independent predictor of reduced risk of new diabetes with hazard ratio (HR) 0.59 [95% confidence interval (CI) 0.39-0.91]. Other significant variables associated with future overt type 2 diabetes in obese patients were triglycerides (50 mg/dL increment) with HR 1.15 (95% CI 1.02-1.28) and fasting glucose (10 mg/dL increment) with HR 2.27 (95% CI 1.83-2.81). CONCLUSION: Bezafibrate, when compared with placebo, reduced the incidence and delayed the onset of type 2 diabetes in obese patients over a long-term follow-up period.