Dietary Patterns, Plasma Fatty Acid Profiles and Risk of Coronary Heart Disease and Type 2 Diabetes (P18-092-19)

Abstract

ObjectivesReduced rank regression (RRR) can incorporate a priori biological hypotheses into exploratory techniques used to generate dietary patterns. No previous studies have used nutrition biomarkers including plasma fatty acids as response variables in RRR. We aimed to derive dietary patterns that explain variation in plasma fatty acid concentrations using RRR and evaluate these in relation to risk of coronary heart disease (CHD) and type 2 diabetes (T2D). MethodsWe derived dietary patterns in a subsample of 711 participants with fatty acid concentrations in the Singapore Chinese Health Study using RRR with 31 food groups/items as predictors and 10 plasma fatty acid biomarkers as response variables. Scores for the dietary patterns derived in the subset were then calculated among the full cohort. We followed up 58,065 and 45,411 men and women for CHD mortality and incident T2D respectively. ResultsWe identified a ‘prudent pattern’ high in green vegetables, fruits and fish and low in rice, eggs and red meat, and a ‘low-meat pattern’ high in bread, margarine and fruits, and low in red meat, seafood and poultry. During 1077,170 and 494,741 person-years of follow-up, 3016 CHD mortality events and 5207 cases of T2D respectively were identified. Both the ‘prudent pattern’ [all adjusted HRs for extreme quintiles, 0.68 (95% CI: 0.60, 0.77); P-trend < 0.001] and ‘low-meat pattern’ [HR, 0.86 (95% CI: 0.76, 0.96); P-trend = 0.010] were associated with lower risk of CHD mortality. The ‘prudent pattern’ was not associated with T2D whereas the ‘low-meat pattern’ was inversely associated with T2D but appeared restricted to women [HR, 0.69 (95% CI: 0.61, 0.78); P-trend < 0.001; P-interaction for sex = 0.001]. ConclusionsUsing nutrition biomarkers as response variables in RRR may be a promising approach to generating dietary patterns predictive of noncommunicable chronic disease risk. Funding SourcesThis study was supported by the National Institutes of Health, USA. JYHS is supported by the NGS Scholarship. W-PK is supported by the National Medical Research Council, Singapore. Supporting Tables, Images and/or Graphs▪▪▪