Dietary patterns predict mortality in the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial (PLCO) cohort

We examined dietary patterns in relation to all‐cause, cancer, and cardiovascular mortality in the PLCO cohort. Principal components analysis on dietary data from 111,525 participants aged 55–74 y identified 3 dietary patterns across US geographic regions and ethnic groups: ‘fruit‐vegetable’ (FV), ‘red meat‐starch’ (RMS), and ‘reduced‐fat food’ (RFF) patterns. Among 64,232 participants free of major chronic disease at baseline (mean (SD) follow‐up 6.2 (1.4) y), 3168 deaths (1478 cancer, 716 cardiovascular) occurred. In adjusted proportional hazards models, the RMS pattern was associated with increased all‐cause mortality (relative risk (RR)=1.18 comparing extreme quintiles, 95% confidence interval (CI) 1.03–1.35; trend p=0.002). The FV pattern was associated with lower cancer (RR=0.68, 95% CI 0.56–0.81; trend p<0.0001) and all‐cause mortality (RR=0.68, 95% CI 0.60–0.77; trend p<0.0001). The RFF pattern was associated with lower cardiovascular (RR=0.73, 95% CI 0.58–0.92; trend p=0.049) and all‐cause mortality (RR=0.79, 95% CI 0.71–0.88; trend p=0.0006). These findings indicate that FV and RFF dietary patterns reduce risk of premature death in healthy adults, while a RMS pattern increases risk.