Abstract P290: Predicting the Change in the Omega-3 Index with Data from 9 Supplementation Studies
The omega-3 index (O3I; erythrocyte EPA+DHA % of total fatty acids) is directly associated with favorable cardiovascular and cognitive outcomes. In spite of its utility as a biomarker of omega-3 status, there remains unexplained variability in its response to EPA+DHA supplementation. De-identified data (n=841) from 9 published intervention studies were used to predict changes in the O3I. All studies provided data on EPA+DHA supplementation dose and duration, pre- and post-O3I values, and standard demographic factors. Stepwise regression analysis was performed to select variables and create a predictive model, using % change as the outcome variable. The mean age and weight of the participants was 48 ± 16 years and 83 ± 18 kg (BMI: 29 ± 5). The cohort was 49% (n=409/841) male and 86% (n=722/841) white. Treatment groups took an average of 2115 ± 1048 mg/day of EPA+DHA (range: 300 - 3600 mg/day; placebo groups took 0 mg/day) for 11.5 ± 4.0 weeks (range: 4 - 20 weeks). The O3I increased from 5.28 ± 1.58% to 9.27 ± 2.59% in the supplemented individuals (n=434), and was unchanged in the placebo groups (5.30 ± 1.52% to 5.30 ± 1.50%; n=407; P<0.0001). Variables that were selected into the model were age, sex, weight, dose of DHA, dose of EPA, duration, baseline O3I and an interaction between weight and age. The model R2 was 0.75 and the root mean square error was 0.18. The correlation between the predicted and observed O3I changes was 0.84 (P<0.0001; Figure). The model predicts individual changes in the O3I with reasonable accuracy, and may be useful for selecting the appropriate supplemental EPA+DHA doses and duration for clinical research studies that aim to reach a target O3I level.
Author Disclosures: K.H. Jackson: A. Employment; Significant; OmegaQuant Analytics, LLC. R. Walker: None. G.C. Shearer: None. W.S. Harris: F. Ownership Interest; Significant; OmegaQuant Analytics, LLC.
- © 2017 by American Heart Association, Inc.