AIM To determine the predictive role of body mass index(BMI) and waist circumference(WC) for diabetes and prediabetes risk in future in total sample as well as in men and women separately. METHODS In a population based cohort study, 1765 with mean ± SD age: 42.32 ± 6.18 healthy participants were followed up from 2003 till 2013(n = 960). Anthropometric and biochemical measures of participants were evaluated regularly during the follow up period. BMI and WC measures at baseline and diabetes and prediabetes status of participants at 2013 were determined. Multivariable logistic regression analysis was used for determining the risk of diabetes and prediabetes considering important potential confounding variables. Receiver operatingcharacteristic curve analysis was conducted to determine the best cut of values of BMI and WC for diabetes and prediabetes. RESULTS At 2013, among participants who had complete data, 45 and 307 people were diabetic and prediabetic, respectively. In final fully adjusted model, BMI value was a significant predictor of diabetes(RR = 1.39, 95%CI: 1.06-1.82 and AUC = 0.68, 95%CI: 0.59-0.75; P < 0.001) however not a significant risk factor for prediabetes. Also, WC was a significant predictor for diabetes(RR = 1.2, 95%CI: 1.05-1.38 and AUC = 0.67, 95%CI: 0.6-0.75) but not significant risk factor for prediabetes. Similar results were observed in both genders.CONCLUSION General and abdominal obesity are significant risk factors for diabetes in future. AIM To determine the predictive role of body mass index (BMI) and waist circumference (WC) for diabetes and prediabetes risk in the future in total sample as well as in men and women separately. METHODS In a population based cohort study, 1765 with mean ± SD age: 42.32 ± 6.18 healthy participants were from 2003 till 2013 (n = 960). Anthropometric and biochemical measures of participants were regularly during the follow up period. BMI and WC measures at baseline and diabetes and prediabetes status of participants at Multivariable logistic regression analysis was used for determining the risk of diabetes and prediabetes considering important potential confounding variables. Receiver operatingcharacteristic curve analysis was conducted to determine the best cut of values ​​of BMI and WC for diabetes and prediabetes. RESULTS At 2013, among participants who had complete data, 45 and 307 people were diabetic and prediabetic, respectively djusted model, BMI value was a significant predictor of diabetes (RR = 1.39, 95% CI: 1.06-1.82 and AUC = 0.68, 95% CI: 0.59-0.75; P <0.001) , WC was a significant predictor for diabetes (RR = 1.2, 95% CI: 1.05-1.38 and AUC = 0.67, 95% CI: 0.6-0.75) but not significant risk factor for prediabetes. Similar results were observed in both genders.CONCLUSION General and abdominal obesity are significant risk factors for diabetes in future.