The figure has 6 panels, arranged in 3 columns and 2 rows. Each panel is a meta-regression plot showing the relationship between the probability of dying and the mean EPA or DHA intake in grams per day in the various intake groups of prospective cohort studies. The three columns of panels correspond to alternative definitions of mortality: Cardiac or cardiovascular mortality on the left; cardiac, cardiovascular or sudden cardiac death mortality in the middle; and all-cause mortality on the right. The upper row shows dose-response meta-regressions assuming a linear dose response relationship throughout the range of mean intakes of EPA or DHA. The lower row shows corresponding meta-regressions assuming different linear relationships with mean dose above or below a threshold of 0.20 grams per day. The linear models in the top row and the spline model for all cause mortality (bottom row on the right) show no evidence of association between mean EPA or DHA intake and the probability of death. However, the splined models for cardiac or cardiovascular mortality (bottom row on the left), and cardiac, cardiovascular mortality or sudden cardiac death mortality (bottom row on the right) are consistent with a decreasing probability of death for mean intakes of EPA or DHA up to 0.20 grams per day, with no further effect on the risk of mortality for higher mean intakes.

Figure 4Meta-regression analyses

The three columns display three alternative definitions of mortality. The upper row (A) shows dose-response meta-regressions assuming a linear dose response relationship throughout the range of mean intakes of EPA and DHA. The lower row (B) shows corresponding meta-regressions assuming different linear relationships with mean dose above or below a threshold of 0.20 grams per day. See Table 4 for results using alternative thresholds.

From: Results

Cover of Effects of Eicosapentanoic Acid and Docosahexanoic Acid on Mortality Across Diverse Settings: Systematic Review and Meta-Analysis of Randomized Trials and Prospective Cohorts
Effects of Eicosapentanoic Acid and Docosahexanoic Acid on Mortality Across Diverse Settings: Systematic Review and Meta-Analysis of Randomized Trials and Prospective Cohorts: Nutritional Research Series, Vol. 4.
Technical Reviews, No. 17.4.
Trikalinos TA, Lee J, Moorthy D, et al.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.