Epidemiologic evidence suggests diabetic men have a slightly lower prostate cancer risk than non-diabetic men. We examined this association in a prospective cohort study of 35 239 men, 50-76 years old, in Washington State who completed a baseline questionnaire between 2000 and 2002. Incident prostate cancers as of 31 December 2004 were identified through the SEER registry. Diabetic men had a slightly lower risk of prostate cancer than non-diabetic men (hazard ratio (HR) 0.83, 95% confidence interval (CI) 0.64-1.07). Insulin users overall and insulin users with diabetic complications had decreased risks, compared to non-diabetic men (HR 0.49, 95% CI 0.26-0.92) and (HR 0.36, 95% CI 0.15-0.87), respectively. Oral medication use for diabetes was not associated with prostate cancer. Insulin is likely a marker of severity of diabetes. Future studies of this association should consider diabetes type, treatment, severity, complications and biomarkers.