The authors investigated the association of body mass index (BMI) with suicide in a record linkage study based on the Swedish Military Service Conscription Register, the Population and Housing Censuses, and the Cause of Death Register. The cohort studied consisted of 1,299,177 Swedish men who were conscripted in 1968-1999, had their BMI measured at age 18-19 years, and were followed up for as long as 31 years. A strong inverse association was found between BMI and suicide. For each 5-kg/m2 increase in BMI, the risk of suicide decreased by 15% (95% confidence interval: 9, 21). The association was similar when subjects with mental disorder at baseline were excluded from the analysis. BMI-suicide associations were similar in relation to suicide deaths occurring in the first 5 years of follow-up (hazard ratio for each 5-kg/m2 increase in BMI = 0.84, 95% confidence interval: 0.73, 0.96) compared with associations > or = 10 years after baseline (hazard ratio = 0.87, 95% confidence interval: 0.79, 0.96), indicating that weight loss as a consequence of mental illness does not explain the BMI-suicide association and that factors influencing BMI may be causally implicated in the etiology of mental disorders leading to suicide.