The Role of Community Food Environments to Reduce Social Disparities in Cardiovascular Disease

Background/Purpose: Cardiovascular diseases (CVDs) account for 32% of all deaths worldwide. Residential food environments are one of the important social determinants of CVD incidence, mortality, and risk factors. However, past literature has been limited by short-term follow-ups, time-invariant environmental measurements at baseline, and/or not investigating both healthy and unhealthy aspects of the food environment. In addition, little is known about whether the effect of food environments on CVD differs by individual’s sociodemographic characteristics. We examined the effects of time-varying healthy and unhealthy food environments on CVD incidence over 10 years of follow-up. In addition, we examined whether race/ethnicity and educational level of individuals moderate the association between food environments and CVD incidence.

Methods: Data were extracted the Health and Retirement Study. Our analytic sample includes respondents aged 50 years and older who had no recorded history of CVD as of 2006 (N=11,254). We used Cox proportional hazards modeling by including CVD incidence from 2006 to 2016 as the outcome variable and four time-varying food environmental measures at the county level (densities of grocery stores, supercenters/club stores, full-service restaurants, and fast-food restaurants) as exposures, after controlling for sociodemographic factors (age, sex, education, and family income), biomedical risk factors at the baseline (hypertension and diabetes), behavioral risk factors at the baseline (frequency of alcohol consumption, smoking status, and frequency of moderate to vigorous physical activity), and neighborhood factors (neighborhood socioeconomic status and neighborhood walkability). We also tested whether educational level and race/ethnicity moderate the association between food environments and CVD incidence.

Results: We observed that 1,790 out of 11,254 adults experienced incident CVD event over the observation period. Density of supercenters/club stores decreased the risk of CVD incidence (HR=0.96, 95% CI=0.92-1.00). In addition, we found a significant interaction effect of education levels in (1) the association between density of full-service restaurants and CVD incidence (interaction term: HR=0.98, 95% CI=0.97-1.00) and (2) the association between density of fast-food restaurants and CVD incidence (interaction term: HR=0.99, 95% CI=0.97-1.00). Specifically, a greater density of full-service restaurants increased risk of incident CVD for individuals with 0-11 years of schooling but decreased CVD risk for those with 12-15 and 16+ years of schooling. Also, while a greater density of fast-food restaurants increased risk of CVD incidence for individuals with 0-11 years of schooling, density of fast-food restaurants did not significantly increase risk of CVD incidence for those with 12-15 years of schooling and 16+ years of schooling. Race/ethnicity was not a significant moderator.

Conclusions/Implications: This study provides a longitudinal analysis of the long-term relationship between healthy and unhealthy food environments and CVD incidence. Public policymakers should be aware of the benefits and nuances of varying food environment components as they can either contribute to positive or negative cardiovascular health. Further research is warranted to examine the mediating mechanisms by which food environments impact CVD.