Background and Aims: Alcohol consumption correlates with type 2 diabetes through its effects on insulin resistance, changes in alcohol metabolite levels, and anti-inflammatory effects. We aim to clarify association between alcohol consumption and risk of diabetes in Taiwanese population. Methods: The National Health Interview Survey (NHIS) in 2001, 2005, and 2009 selected a representative sample of Taiwan population using a multistage sampling design. Information was collected by standardized face to face interview. Study subjects were connected to the Taiwan National Health Insurance claims dataset and National Register of Deaths Dataset from 2000 to 2013. Kaplan-Meier curve with log rank test was employed to assess the influence of alcohol drinking on incidence of diabetes. Univariate and multivariate Cox proportional regression were used to recognize risk factors of diabetes. Results: A total of 43,000 participants were included (49.65% male; mean age, 41.79±16.31 years). During the 9-year follow-up period, 3,650 incident diabetes cases were recognized. Kaplan-Meier curves comparing the four groups of alcohol consumption showed significant differences (p<0.01). After adjustment for potentially confounding variables, compared to social drinkers, the risks of diabetes were significantly higher for non-drinkers (adjusted hazard ratio [AHR]=1.21; 95% confidence interval [CI], 1.09-1.34; p<0.01), regular drinkers (AHR=1.19; 95% CI, 1.06-1.35; p <0.01), and heavy drinkers (AHR=2.21, 95% CI, 1.56-3.13, p<0.01). Conclusions: Social drinkers have a significantly decreased risk of new-onset diabetes compared with non-, regular, and heavy drinkers. Disclosure Y. Lai: None.