Patterns of recommended dietary behaviors predict subsequent risk of mortality in a large cohort of men and women in the United States.

Recommendations for intake of fruits and vegetables, whole grains, lean meats, and low-fat dairy form the underpinning of dietary guidance for health promotion. We examined the association of a summary index of food consumption behaviors compatible with the spirit of prevailing dietary guidance and mortality. We used data from the NIH-American Association of Retired Persons cohort (n = 350,886), aged 50-71 y and disease free at baseline in 1995-1996, to examine the association of a dietary behavior score (DBS) with mortality after 10.5 y of follow-up (deaths, n = 29,838). The DBS included 6 equally weighted components derived from responses to questions on usual dietary behaviors related to consumption of fruits, vegetables, low-fat dairy, whole grains, lean meat and poultry, and discretionary fat. The covariate-adjusted association of DBS and mortality from all causes, cancer, and coronary heart disease was examined using Cox proportional hazards regression methods. Compared with those in the lowest one-fifth of DBS, the multivariate-adjusted relative risk of mortality in the highest one-fifth of the DBS was 0.75 (95% CI, 0.70-0.80) in women and 0.79 (95% CI, 0.75-0.83) in men (P-trend < 0.0001). The inverse association of DBS and mortality was significant in both genders in nearly all categories of covariates. Similar trends were observed for DBS associations with mortality from cancer and heart disease. Nearly 12% of the covariate-adjusted population risk of mortality was attributable to nonconformity with dietary recommendations. Adoption of recommended dietary behaviors was associated with lower mortality in both men and women independent of other lifestyle risk factors.