Diet index-based and empirically derived dietary patterns are associated with colorectal cancer risk.

Previous studies have derived patterns by measuring compliance with preestablished dietary guidance or empirical methods, such as principal components analysis (PCA). Our objective was to examine colorectal cancer risk associated with patterns identified by both methods. The study included 431 incident colorectal cancer cases (225 men, 206 women) and 726 healthy controls (330 men, 396 women) participating in a population-based, case-control study. PCA identified sex-specific dietary patterns and the Healthy Eating Index-2005 (HEI-05) assessed adherence to the 2005 Dietary Guidelines for Americans. A fruits and vegetables pattern and a meat, potatoes, and refined grains pattern were identified among men and women; a third pattern (alcohol and sweetened beverages) was identified in men. The fruits and vegetables pattern was inversely associated with risk among men [odds ratio (OR) = 0.38, 95% CI = 0.21-0.69 for the highest compared with the lowest quartile] and women (OR = 0.35, 95% CI = 0.19-0.65). The meat, potatoes, and refined grains pattern was positively associated with risk in women (OR = 2.20, 95% CI = 1.08-4.50) and there was a suggestion of a positive association among men (OR = 1.56, 95% CI = 0.84-2.90; P-trend = 0.070). Men and women with greater HEI-05 scores had a significantly reduced risk of colorectal cancer (OR = 0.56, 95% CI = 0.31-0.99; OR = 0.44, 95% CI = 0.24-0.77, respectively). Following the Dietary Guidelines or a dietary pattern lower in meat, potatoes, high fat, and refined foods and higher in fruits and vegetables may reduce colorectal cancer risk.