Recent studies in Western nations have shown inverse associations between body mass index (BMI, measured as weight (kg)/height (m)(2)) and suicide. However, it is uncertain whether the association is similar in non-Western settings, and the biologic pathways underlying the association are unclear. The authors investigated these issues in a cohort of 542,088 Taiwanese people 20 years of age or older who participated in a health check-up program (1994-2008); there were 573 suicides over a mean 8.1 years of follow up. There was a J-shaped association between BMI and suicide risk (P for the quadratic term = 0.033) but limited evidence of a linear association (adjusted hazard ratio per 1-standard-deviation increase = 0.95 (95% confidence interval: 0.85, 1.06)); compared with individuals whose BMI was 18.5-22.9, adjusted hazard ratios for those with a BMI <18.5 or ≥35 were 1.56 (95% confidence interval: 1.07, 2.28) and 3.62 (95% confidence interval: 1.59, 8.22), respectively. A high waist-to-hip ratio was associated with an increased risk of suicide. There was some evidence for a reverse J-shaped association of systolic blood pressure and high density lipoprotein cholesterol with suicide and an association of higher triglyceride level with increased suicide risk; these associations did not appear to mediate the associations of BMI and waist-to-hip ratio with suicide.