Skip to main content
. 2021 Feb 1;190(7):1353–1365. doi: 10.1093/aje/kwab004

Figure 2.

Figure 2

Associations of single dietary components with predicted risk of atherosclerotic cardiovascular disease (ASCVD) among women (A) and men (B) in the CARDIA Study, 1985–2006. The figure shows the estimated change in predicted ASCVD risk for an interquartile range change (25th–75th percentiles) in a single dietary component while all of the other components are fixed at their 25th, 50th (median), or 75th percentiles. The right side of the vertical dotted line indicates a more harmful association, whereas the left side indicates a more protective association. Only dietary components that passed Bayesian kernel machine regression variable selection were plotted. Bars, 95% credible intervals (CrIs). CARDIA, Coronary Artery Risk Development in Young Adults; SSB, sugar-sweetened beverages.