Variation in Red Meat Categories Meaningfully Influences Red Meat Intake Estimates of the U.S. Population Aged 2 + Years

Limitations in dietary assessment methodology lead to broad and inconsistent red meat intake categories. For example, red meat is often combined with processed meat, which includes processed poultry. Our objective was to quantitatively assess how these limitations influence red meat intake estimates of the U.S. population. Three distinct red meat categories were identified from a series of systematic literature reviews. By systematically disaggregating and reaggregating meat variables (lean meat oz-eq) from the USDA's Food Pattern Equivalents Database, the following categories were created: 1) ‘unprocessed red meat’, 2) ‘total red meat’ [unprocessed red meat + processed red meat], and 3) ‘mixed red and processed meat’ [unprocessed red meat + total processed meat which includes processed poultry]. The 2015–16 and 2017–18 cycles of NHANES (n = 15,039) were combined to estimate and compare mean intakes across these three categories as well as the proportion of the population that exceeded age- and sex-specific red meat allotments in the Dietary Guidelines for Americans recommended eating patterns for those aged 2 + years. The two-part correlated model of the NCI Usual Intake method in SAS was used. Estimates were adjusted for age, sex, race, and ethnicity and pairwise comparisons were made across categories; P < 0.05. Mean intake estimates (lean meat oz-eq/day) differed by red meat category (pairwise comparisons P < 0.00,001): 1.5 ± 0.04 for unprocessed red meat, 2.2 ± 0.05 for total red meat, 2.4 ± 0.05 for mixed red and processed meat. The % of the population exceeding their respective age-sex red meat allotment in the DGA eating patterns also differed by category (pairwise comparisons P < 0.00,001): 18.3 ± 1.9% for unprocessed red meat, 45.1 ± 1.8% for total red meat, 55.2 ± 1.8% for mixed red and processed meat. Our results emphasize the importance of accurate and transparent red meat categories in nutrition epidemiology when estimating population intakes. Future research should examine how differences in these intake categories can influence associations between red meat intake and chronic disease. Not applicable.