Colorectal cancer and diabetes are common diseases that share many risk factors. It has been hypothesized that diabetes is a risk factor for colorectal cancer. We used two large population-based case-control studies from North Carolina to determine whether diabetes and/or insulin therapy was associated with colon cancer and/or rectal cancer (defined as cancer of the sigmoid colon, rectosigmoid, or rectum) and whether this association differed by race. Cases and matched controls from the North Carolina Colon Cancer Studies I and II were interviewed about demographics, dietary factors, diagnosis of diabetes, and use of medications to treat diabetes. Odds ratios (OR) and 95% confidence intervals (95% CI) were estimated using unconditional logistic regression. Colon and rectal cancer cases reported a higher prevalence of diabetes than their respective control groups. Compared with Whites without diabetes, Whites with diabetes had adjusted ORs of 1.40 (95% CI, 0.93-2.12) for colon cancer and 1.38 (95% CI, 1.00-1.90) for rectal cancer. Diabetes was not associated with colon or rectal cancer among African Americans [OR, 1.17 (95% CI, 0.81-1.70) and 0.75 (95% CI, 0.44-1.28), respectively]. Among Whites with diabetes, insulin use was positively associated with rectal cancer. The same association was not seen for African American diabetics using insulin; however, the number of African Americans using insulin was small. In sum, diabetes was positively associated with rectal cancer and approached a positive association with colon cancer among Whites. No association was present among African Americans. Insulin use was also positively associated with rectal cancer among Whites.