Fig. 2

Prospective associations between dietary patterns and the risk of total CVD events and all-cause mortality (n = 116,806). Notes: All the models were stratified by sex and regions (England, Scotland, and Wales) and adjusted for ethnicity, socioeconomic status, behavioral risk factors, energy intake, and menopause in women. Z-scores for DP1 and DP2 were analyzed in mutually adjusted models to examine their independent associations with health outcomes. Adjusted HRs (hazard ratio) and confidence intervals (CI) of DP scores quintiles obtained using the floated absolute risk method of Cox proportional hazards regression, which enabled the comparisons across different quintiles of z-score. Trend tests were conducted by including the median score of each pattern quintile as a continuous variable in the models