Estimation and predictors of the Omega-3 Index in the UK Biobank

Abstract

Information on the Omega-3 Index (O3I) in the United Kingdom (UK) is scarce. The UK-Biobank (UKBB) contains data on total plasma <i>n</i>3-PUFA% and DHA% measured by NMR. The aim of our study was to create an equation to estimate the O3I (eO3I) from these data. We first performed an inter-laboratory experiment with 250 random blood samples in which the O3I was measured in erythrocytes by GC, and total <i>n</i>3 % and DHA% were measured in plasma by NMR. The best predictor of eO3I included both DHA% and a derived metric, the total <i>n</i>3 %-DHA%. Together these explained 65 % of the variability (<i>r</i> = 0·832, <i>P</i> < 0·0001). We then estimated the O3I in 117 108 UKBB subjects and correlated it with demographic and lifestyle variables in multivariable-adjusted models. The mean eO3I was 5·58 % (sd 2·35 %) in this UKBB cohort. Several predictors were significantly correlated with eO3I (all <i>P</i> < 0·0001). In general order of impact and with directionality (-, inverse and +, direct): oily-fish consumption (+), fish oil supplement use (+), female sex (+), older age (+), alcohol use (+), smoking (-), higher waist circumference and BMI (-), lower socioeconomic status and less education (-). Only 20·5 % of eO3I variability could be explained by predictors investigated, and oily fish consumption accounted for 7·0 % of that. With the availability of the eO3I in the UKBB cohort, we will be in a position to link risk for a variety of diseases with this commonly used and well-documented marker of <i>n</i>3-PUFA biostatus.