Division of Epidemiology and Community Health, School of Public Health, University of Minnesota, 1300 South Second Street, Suite 300, Minneapolis, MN 55454, USA.
BACKGROUND: There is considerable support for associations between insulin and IGF-I levels and colorectal cancer. Diet may relate to colorectal cancer through this mechanism, for example, diets high in glycemic index, glycemic load and/or carbohydrate are hypothesized to increase insulin load and the risk of insulin resistance, hyperinsulinemia. Case-control studies support this hypothesis, but prospective cohorts have had mixed results. METHODS: In the Breast Cancer Detection Demonstration Project (BCDDP) follow-up cohort of 45,561 women, we used Cox proportional hazards regression to assess the distribution of 490 incident cases of colorectal cancer ascertained during 8.5 years of follow-up across quintiles of carbohydrate intake, glycemic index, and glycemic load. We also stratified by combined BMI and physical activity levels. RESULTS: We found reductions in colorectal cancer risk for diets high in carbohydrate (RR for Q5 vs. Q1 = 0.70, 95% CI: 0.50-0.97) and glycemic index (0.75, 95% CI: 0.56-1.00), and no significant association for glycemic load (0.91, 95% CI: 0.70-1.20). Inverse associations were weakest in normal weight active persons. The inverse association for glycemic index was strongest for the portion from dairy food. CONCLUSIONS: These results do not support an association between diets high in carbohydrate, glycemic index or glycemic load and colorectal cancer.