Dietary patterns and total mortality in a Mediterranean cohort: the SUN project.

BACKGROUND Different dietary patterns have been associated with several health outcomes, including morbidity and mortality. There is little evidence on the association between empirically derived dietary patterns and all-cause mortality in Southern European populations. OBJECTIVE The aim of our study was to prospectively evaluate the association between an empirically derived dietary pattern and all-cause mortality. DESIGN The Seguimiento Universidad de Navarra (SUN) Project is an ongoing, multipurpose, prospective and dynamic Spanish cohort. PARTICIPANTS/SETTING Participants were a prospective cohort of 16,008 middle-aged Spanish adults. All of them were university graduates (alumni) (59.6% women, mean age 38 years). Usual diet was assessed at baseline with a validated semiquantitative food frequency questionnaire. MAIN OUTCOME MEASURES Deaths were confirmed by review of medical records and of the National Death Index. STATISTICAL ANALYSIS Dietary patterns were ascertained through a factor (principal component) analysis based on 30 predefined food groups. Participants were classified according to tertiles of adherence to the three main dietary patterns identified with factor analysis. Cox regression models were fitted to estimate multivariable-adjusted hazard ratios and 95% CIs for mortality. RESULTS Three major dietary patterns were identified. They were labelled Western dietary pattern (rich in red and processed meat, potatoes, and fast food), Mediterranean dietary pattern (rich in vegetables, fish and seafood, fruits, and olive oil), and alcoholic beverages dietary pattern. During follow-up, 148 deaths were reported (mean age at death 54.5 years). After adjustment for potential confounders, the lowest risk of all-cause mortality was found in the tertile of highest adherence to the Mediterranean dietary pattern (adjusted hazard ratio for third vs first tertile 0.53, 95% CI 0.34 to 0.84) (P for trend <0.01). The Western dietary pattern and alcoholic beverages dietary pattern showed no significant association with mortality. CONCLUSIONS Greater adherence to an empirically derived dietary pattern approaching the traditional Mediterranean diet was associated with a reduction in the risk of all-cause mortality among middle-aged Mediterranean adults.