Abstract

BACKGROUND: The remarkable similarity of lifestyle and environmental risk factors for type 2 (non-insulin-dependent) diabetes mellitus and colon cancer has led to the hypothesis that diabetes may increase the risk of this cancer. We prospectively examined the relationship between diabetes and risk of colorectal cancer in a cohort of 118 403 women aged 30 through 55 years who were without previously diagnosed cancer at baseline in 1976. METHODS: The women, who were enrolled in the Nurses' Health Study, were assessed for history of diabetes at baseline and during follow-up by use of biennial questionnaires. Self-reported diabetes was validated by information obtained from a supplemental questionnaire on symptoms and treatment and was confirmed by medical record review in a sample of the participants. Incident cases of colorectal cancer were ascertained through medical record review. All reported P values are two-sided. RESULTS: During 18 years of follow-up (2 001 061 person-years), we documented 892 new cases of colorectal cancer. After adjustment for age, body mass index (weight in kg/height in m 2 ), physical activity, and other covariates, relative risks (RRs) were 1.43 (95% confidence interval [ CI] = 1.10-1.87; P = .009) for colorectal cancer, 1.49 (95% CI = 1.09-2.06; P = .01) for colon cancer, 1.11 (95% CI = 0.56-2.21; P = .76) for rectal cancer, 1.56 (95% CI = 1.07-2.28; P = .02) for advanced colorectal cancer, and 2.39 (95% CI = 1.46-3.92; P = .0005) for fatal colorectal cancer. CONCLUSION: Our data provide support for the hypothesis that diabetes is associated with an increased risk of colorectal cancer in women.