We determined whether longitudinal changes in MetS were influenced by interaction between a genetic risk score (GRS) and habitual dietary patterns among whites and African Americans. We used the Atherosclerosis Risk in Communities study data from 4-visits (1987-1998) for 9,778 whites and 2,922 African Americans aged 45-65y at baseline. MetS was defined according to the NCEP, Adult Treatment Panel III criteria. Dietary variables were obtained from the Atherosclerosis Risk in Communities study food frequency questionnaire. Generalized linear models using risk ratios (RR) and 95% confidence intervals (CI) with p < 0.05 were used to judge model statistical significance. For every 5-allele increase, the GRS was associated with increased risk for MetS in whites (RR=1.12;95% CI=1.11-1.13) and African Americans (RR=1.16;1.14-1.19), respectively. Those in the upper half of the GRS had 1.20-fold (RR=1.20;1.18-1.23) and 1.25-fold (RR=1.25;1.20-1.29) increased risk for MetS among whites and African Americans, respectively. For every 5-unit increase in factor scores for dietary patterns: Red/processed meat (RR= 1.13;1.09-1.18), skim milk/lowfat dairy (RR=1.11;1.06-1.17), high-fat dairy/butter (RR=0.95;0.90-100), and high carbohydrate foods (RR=1.20;1.14-1.27) were associated with longitudinal changes in MetS among whites. Dietary pattern with whole milk/high fat dairy was protective in MetS among African Americans (RR=0.81;0.72-0.90) but the dietary pattern characterized by red processed meat was harmful (RR=1.16;1.08-1.25). In interaction models, among whites, a protective risk remained for high-fat dairy dietary pattern after interaction with the risk-causing GRS (RR=0.85;0.75-0.96). Certain dietary patterns may be associated with MetS. However, the effect of a higher fat intake dietary pattern may be protective of MetS among whites and African Americans even with a high burden of risk-causing alleles among whites.


D.S. Hardy: None.


National Heart, Lung, and Blood Institute (K01HL127278)

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.