OBJECTIVES: Prostate cancer has few known risk factors. As part of a population-based case-control study conducted in four health maintenance organizations, the authors examined the associations between fatal prostate cancer and several medical and behavioral characteristics. METHODS: Cases were 768 health plan members who died of prostate adenocarcinoma during the period 1997-2001. We randomly selected controls (929) from the health plan membership and matched them to cases on health plan, age, race, and pattern of health plan membership. We examined medical records to obtain information on potential risk factors during the 10 years before the date on which prostate cancer was first suspected; the same reference date was used for the matched controls. RESULTS: Anthropometric characteristics, as well as personal histories of benign prostatic hypertrophy, transurethral prostatectomy, cancer, diabetes, prostatitis, hypertension, and vasectomy were largely similar for cases and controls. Men who died from prostate cancer were more likely than controls to have been cigarette smokers according to the most recent smoking notation before the reference date (odds ratio 1.5, 95% confidence interval 1.1-2.0). CONCLUSIONS: The observed increase in risk associated with recent cigarette smoking is consistent with the findings of several other studies. However, in contrast with some reports, we observed no connection between fatal prostate cancer and some prior health conditions or measures of body size.