Although diabetes mellitus is associated with an increased risk of several malignancies, a negative association with prostate cancer is biologically most plausible. The epidemiologic evidence is, however, inconsistent, limited and based mostly on small studies. We present results from a large, population-based cohort study in Sweden, where we assessed prostate cancer risk among patients hospitalized for diabetes mellitus. The cohort was composed of patients identified in the Swedish In-Patient Register as having a hospital discharge diagnosis of diabetes mellitus in 1965-1994. The follow-up was done by linkages with the national cancer register and other population-based registers. Standardized incidence ratios (SIRs), with 95% confidence interval (CI), were used as a measure of relative risk. After complete exclusion of the first year of follow-up (to avoid selection bias), 135,950 men remained in the cohort, contributing 827,099 years of follow-up to the study. A total of 2,455 incident cases of primary prostate cancer were identified during 1-31 years of follow-up, yielding an overall SIR of 0.91 (95% CI 0.87-0.94); this risk reduction was more pronounced among patients who have been hospitalized for diabetic complications (SIR = 0.82; 95% CI 0.74-0.91). We found no consistent trends in risk related to age at first hospitalization or to duration of follow-up. We did find a small, but significantly decreased risk of prostate cancer among men who had been hospitalized for diabetes mellitus.
Copyright 2002 Wiley-Liss, Inc.