Associations of animal source foods, cardiovascular disease history, and health behaviors from the national health and nutrition examination survey: 2013–2016

Document Type


Publication Date



Background: Some individuals adopt vegetarian or plant-based diets to improve their health. Observational evidence suggests diets composed of higher amounts of animal-source foods (ASFs) are associated with increased risk for disease and early mortality. In many of these studies, those who ate fewer animal-source foods reported fewer disease risk factors and unhealthy behaviors, which could indicate bias. Purpose: This study aims to examine the relationships between ASF consumption, health behaviors, and cardiovascular disease (CVD) prevalence in a population-representative sample of U.S. civilians controlling for confounders. Methods: Respondent data were collected from the National Health and Nutrition Examination Survey (NHANES) 2013–2016 collection years. Collected data included demographics, ASF intake, healthy lifestyle variables, body mass index, and blood lipids. Results: There was a higher proportion of those with CVD history who consumed red meat (61.3%; C.I. 41.7%–77.8%), but the proportion was lower for white (23.3%; C.I. 12.6%–39.0%) and processed meat (15.4%; C.I. 6.5%–32.3%). When adjusted for sex, the odds of CVD history increased for red meat compared to processed meat consumption (OR 2.95; C.I. 1.14–7.66). Unhealthy lifestyle increased the odds of CVD history by nearly 8-fold (OR 7.8; C.I. 3.44–17.7). Individual factors including age, smoking history, body mass index, and blood lipids, and demographic factors, including education level, race, and income, were also associated with increased odds for CVD history. ROC analysis revealed 77.2% AUC for CVD history classified by individual factors (BMI ≥30 kg/m**2, ≤ 30 min moderate physical activity, smoker, fiber intake ≤25 g, dental visit more than two years ago, and age above 60 years). Three or more factors moderately predicted CVD history when optimized for sensitivity (73.4%) and specificity (71%). Adjusted for sex, the relationship between CVD and moderate physical activity became stronger possibly reflecting lifestyle changes. Despite evidence of lifestyle changes, modifiable risk factors persisted in the CVD group. CVD diagnosis in males was substantially delayed compared to females concerning the sex-specific age cutoff associated with higher risk. The healthy lifestyle group was characterized by earlier CVD diagnosis and fewer overall risk factors compared to the unhealthy lifestyle group. Conclusion: CVD history was strongly associated with demographic, lifestyle, and dietary factors. Future research should focus on multidimensional models for disease risk stratification and prevention, including individual, behavioral, and sociodemographic factors.

Publication Title

Global Epidemiology



This document is currently not available here.