Abstract
A plant-based diet is associated with all-cause and cardiovascular disease (CVD) mortality, but the associations might differ by plant qualities. We examined the associations between the plant-based diet index (PDI) and mortality. Adults with 24-hour dietary recall from the 2007-2015 Korean National Health and Nutrition Examination Surveys (KNHANES), who were eligible for and linked to national mortality records through 2019, were included (n = 44,291). Three PDIs were calculated: overall PDI (all plant foods), healthful PDI (hPDI, healthful plant foods), and unhealthful PDI (uPDI, unhealthful plant foods), with higher scores for respective plant foods and animal foods reverse-scored. A weighted Cox proportional hazards model estimated hazard ratios (HRs) and 95% confidence intervals (CIs) for mortality. During follow-up (median 8.3-year), 2,302 deaths were identified. For all-cause mortaltiy, HRs (95% CIs) were 0.77 (0.64-0.93) for overall PDI, 0.82 (0.68-0.99) for hPDI, and 1.17 (0.98-1.39) for uPDI in the fifth versus first quintile (P-trend: <0.001, 0.035, 0.031). For CVD mortality, corresponding HRs were 0.60 (0.40-0.90), 0.63 (0.42-0.94), and 1.90 (1.32-2.75). Adherence to an overall PDI and hPDI was associated with a lower risk of all-cause and CVD mortality, while uPDI was associated with a higher risk.