Low-Fat Dietary Pattern and Risk of Colorectal Cancer: The Women???s Health Initiative Randomized Controlled Dietary Modification Trial

Abstract

Previous studies have failed to document positive effects of a low-fat diet on the risk of colorectal cancer. The risk was examined in relation to the dietary pattern in the Women's Health Initiative Dietary Modification Trial. This randomized, controlled trial enrolled 48,835 postmenopausal women 50 to 79 years of age seen at 40 clinical centers throughout the United States in the years 1993-1998. The interventional diet called for at least 5 daily servings of vegetables and fruits and at least 6 servings of grains with the goal of lowering total fat to 20% of energy intake. An intensive behavioral intervention entailed 18 group sessions in the first year and, subsequently, quarterly sessions led by specially trained and certified nutritionists. Dietary goals were individualized on the basis of height. Motivational interviewing was part of the program. Forty percent of participants were assigned to the intervention and 60% to a comparison group of women who were given dietary guidelines but were not asked to change their diets. The 2 groups had similar risk profiles for colorectal cancer. During a mean follow-up interval of 8.1 years, the reduction in percentage of energy from fat was approximately 70% of that planned. Fewer than one third of women met the goal after 12 months and only 14%' at year 6. Consumption of saturated fat did decline, and women in the intervention group did eat more fruits, vegetables, and grains. Dietary intake of folate and plasma carotenoid levels increased significantly, whereas serum cholesterol levels declined. Rates of invasive colorectal cancer were 0.13% per year in the intervention group and 0.12% in the comparison group. The hazard ratio (HR) was 1.08, with a 95% confidence interval (CI) of 0.90-1.29. The cumulative hazard of colorectal cancer was very similar in the 2 management groups, and no time trend for invasive colorectal cancer was evident. Outcomes were similar in the 2 groups, and there were no significant differences in total cancer incidence, total cancer mortality, or total mortality. Tumor characteristics were similar in the 2 groups. Annualized incidence rates of self-reported colonic polyps or adenomas were lower in the intervention group (HR, 0.91; 95% CI, 0.87-0.95). The risk of colon cancer increased with advancing age, but intervention did not interact with age at baseline. This large trial gave no evidence that, when taken by midlife to late-life postmenopausal women, a low-fat diet decreases the risk of colorectal cancer over 8 years of observation.