Abstract WE508: Ultra Processed Foods (UPF) and Risk of Incident Coronary Heart Disease (CHD) REasons for Geographic and Racial Disparities in Stroke (REGARDS)

Abstract

Background: Diet plays a critical role in coronary heart disease (CHD) risk, and evidence links intake of ultra-processed foods (UPFs) to increased cardiovascular risk. CHD mortality remains higher among men, and overall diet quality in the United States is poor, with men exhibiting lower Healthy Eating Index (HEI) scores than women. However, few studies have examined whether the association between UPF intake and CHD risk differs by sex. We investigated the longitudinal relationship between UPF consumption and incident CHD among both sexes, hypothesizing that UPF intake varies by sex and that higher consumption is associated with greater CHD risk. Methods: We analyzed data from participants in the REasons for Geographic and Racial Disparities in Stroke (REGARDS) cohort free from CHD with complete covariate and diet data. UPF consumption was assessed using the NOVA classification system and quantified as both a percentage of total energy and total grams. Associations between UPF intake and incident CHD, defined as fatal and non-fatal myocardial infarction, were evaluated using Cox proportional hazards models. Results: Among 15,997 participants (mean [SD] age 64 (9), 60% female), males consumed significantly more UPF, both as a percentage of total energy intake (mean ± SD 49.5% ± 11.8% vs. 47.0% ± 12.3%, p < 0.0001) and as a percentage of total grams consumed (median IQR: 19.0% [12.8–27.0%] vs. 14.8% [9.7–23.1%], p < 0.0001). Males also had a higher rate of CHD events across quartiles of UPF intake by energy and grams (Figure 1, Figure 2). In unadjusted models, higher UPF intake (% of total energy) was associated with significantly greater risk of incident CHD in both sexes. However, these associations were no longer statistically significant in the fully adjusted model (Table 1). Conclusions: Higher UPF intake was associated with increased CHD risk in both male and females in unadjusted models; however, these relationships attenuated to null in fully adjusted models. Although the association between UPF intake and CHD was attenuated, effect sizes remained large, suggesting a potentially meaningful relationship.