We sought to evaluate the contribution of various modifiable risk factors to the partial population attributable risk (PARp) for diabetes in an Asian Indian population. Of a cohort of 3589 individuals, representative of Chennai, India, followed up after a period of ten years, we analyzed data from 1376 individuals who were free of diabetes at baseline. A diet risk score was computed incorporating intake of refined cereals, fruits and vegetables, dairy products, and monounsaturated fatty acid. Abdominal obesity was found to contribute the most to incident diabetes [Relative Risk (RR) 1.63(95%CI 1.21–2.20)]; (PARp 41.1% (95%CI 28.1–52.6)]. The risk for diabetes increased with increasing quartiles of the diet risk score [highest quartile RR 2.14(95% CI 1.26–3.63)] and time spent viewing television [(RR 1.84(95%CI 1.36–2.49] and sitting [(RR 2.09(95%CI 1.42–3.05)]. The combination of five risk factors (obesity, physical inactivity, unfavorable diet risk score, hypertriglyceridemia and low HDL cholesterol) could explain 80.7% of all incident diabetes (95%CI 53.8–92.7). Modifying these easily identifiable risk factors could therefore prevent the majority of cases of incident diabetes in the Asian Indian population. Translation of these findings into public health practice will go a long way in arresting the progress of the diabetes epidemic in this region.