Associations between dietary patterns and the incidence of total and fatal cardiovascular disease and all-cause mortality in 116,806 individuals from the UK Biobank: a prospective cohort study.
Gao M., Jebb SA., Aveyard P., Ambrosini GL., Perez-Cornago A., Carter J., Sun X., Piernas C.
BACKGROUND: Traditionally, studies investigating diet and health associations have focused on single nutrients. However, key nutrients co-exist in many common foods, and studies focusing solely on individual nutrients may obscure their combined effects on cardiovascular disease (CVD) and all-cause mortality. We aimed to identify food-based dietary patterns which operate through excess energy intake and explain high variability in energy density, free sugars, saturated fat, and fiber intakes and to investigate their association with total and fatal CVD and all-cause mortality. METHODS: Detailed dietary data was collected using a 24-h online dietary assessment on two or more occasions (n = 116,806). We used reduced rank regression to derive dietary patterns explaining the maximum variance. Multivariable Cox-proportional hazards models were used to investigate prospective associations with all-cause mortality and fatal and non-fatal CVD. RESULTS: Over an average of 4.9 years of follow-up, 4245 cases of total CVD, 838 cases of fatal CVD, and 3629 cases of all-cause mortality occurred. Two dietary patterns were retained that jointly explained 63% of variation in energy density, free sugars, saturated fat, and fiber intakes in total. The main dietary pattern was characterized by high intakes of chocolate and confectionery, butter and low-fiber bread, and low intakes of fresh fruit and vegetables. There was a positive linear association between the dietary pattern and total CVD [hazard ratio (HR) per z-score 1.07, 95% confidence interval (CI) 1.04-1.09; HRtotal CVD 1.40, 95% CI 1.31-1.50, and HRall-cause mortality 1.37, 95% CI 1.27-1.47 in highest quintile]. A second dietary pattern was characterized by a higher intakes of sugar-sweetened beverages, fruit juice, and table sugar/preserves. There was a non-linear association with total CVD risk and all-cause mortality, with increased risk in the highest quintile [HRtotal CVD 1.14, 95% CI 1.07-1.22; HRall-cause mortality 1.11, 95% CI 1.03-1.19]. CONCLUSIONS: We identified dietary patterns which are associated with increased risk of CVD and all-cause mortality. These results help identify specific foods and beverages which are major contributors to unhealthy dietary patterns and provide evidence to underpin food-based dietary advice to reduce health risks.