Dietary Patterns Derived Using Exploratory and Confirmatory Factor Analysis are Stable and Generalizable Across Race, Region, and Gender Subgroups in the REGARDS Study

Suzanne E. Judd, Abraham J. Letter, James M. Shikany, David L. Roth, P. K. Newby

Examining diet as a whole using dietary patterns as exposures is a complementary method to using single food or nutrients in studies of diet and disease, but the generalizability of intake patterns across race, region, and gender in the United States has not been established. To employ rigorous statistical analysis to empirically derive dietary patterns in a large bi-racial, geographically diverse population and examine whether results are stable across population subgroups. The present analysis utilized data from 21,636 participants in the Reasons for Geographic and Racial Differences in Stroke (REGARDS) study who completed the Block 98 food frequency questionnaire. We employed exploratory factor analysis and confirmatory factor analyses on 56 different food groups iteratively and examined differences by race, region, and sex to determine the optimal factor solution in our sample. Five dietary patterns emerged: the "Convenience" pattern was characterized by mixed dishes; the "Plant-based" pattern by fruits, vegetables, and fish; the "Sweets/Fats" pattern by sweet snacks, desserts, and fats and oils; the "Southern" pattern by fried foods, organ meat, and sweetened beverages; and the "Alcohol/Salads" pattern by beer, wine, liquor, and salads. Differences were most pronounced in the Southern pattern with black participants, those residing in the Southeast, and participants not completing high school having the highest scores. Five meaningful dietary patterns emerged in the REGARDS study and showed strong congruence across race, sex, and region. Future research will examine associations between these patterns and health outcomes to better understand racial disparities in disease and inform prevention efforts.