Abstract
PURPOSE: To assess the association between diabetes mellitus (DM) and the incidence of cancer at different sites. METHODS: Data from the baseline and first three follow-up visits of the Atherosclerosis Risk in Communities (ARIC) study, an ongoing cohort study of adults from four American communities, were used in this study. Of 15,792 persons aged 45-64 years old who participated in the baseline visit, the data of 15,118 participants were available for this study. For each cancer site, a conditional stratified Poisson regression model was fitted to estimate the adjusted relative rate and 95% confidence interval (adj. RR, 95% CI) of its incidence in diabetics compared to non-diabetics. RESULTS: We excluded 850 participants with a history of cancer at baseline and 149 participants who developed cancer during 2 years after enrollment, leaving a total of 14,119 participants of whom 1721 were diabetics. Independent of age, body mass index, alcohol consumption, and physical activity, DM decreased the risk of all cancers combined (adj. RR: 0.77, 95% CI: 0.60, 0.98) and the risk of prostate cancer (adj. RR: 0.51, 95% CI: 0.27, 0.97) and increased the risk of colorectal cancer in non-menopausal women (adj. RR: 12.08, 95% CI: 2.06, 70.94). CONCLUSIONS: In conclusion, DM may be associated with an increased risk of colorectal cancer in non-menopausal women and a decreased risk of prostate cancer and all cancers combined.