Skip to content

Abstract Search

Environment/Climate Change

Long-term Exposure to Traffic-related Air Pollution and Cardiovascular Hospitalizations: Examining Health Disparities in 14 U.S. States Adjani A. Peralta* Adjani Peralta Yaguang Wei Joel Schwartz

Background

Cardiovascular diseases are a leading cause of morbidity and mortality in the U.S. Understanding the relationship between exposure to traffic-related air pollution and cardiovascular hospitalizations could help identify vulnerable populations and provide evidence for a revised EPA annual NO2 standard.

Methods

Using generalized linear models, we studied the impact of nitrogen dioxide ( NO2) and elemental carbon (EC) exposure, specifically from traffic-related air pollution exposures, on the rate of all-cardiovascular (CVD) hospitalization in 14 U.S. states between 2000-2016. We aggregated hospitalization rates from the Health Cost and Utilization Project State Inpatient Databases and estimated annual PM2.5 exposures from a validated ensemble exposure model to the ZIP code level within each state. Covariates were attained from the decennial Census and the American Community Surveys. Multiplicative interaction terms were used to assess effect measure modification by race and poverty.

Results

For each 1-ppb increase in annual NO2 and 1-µg/m3 in EC, we found positive associations with all-cardiovascular hospitalization rates: NO2 1.0072% (95% CI: 1.0067,1.0077) and EC 1.27% (95% CI: 1.24, 1.29). Significant positive associations persisted at <58 ppb NO2 levels, the current annual US standard: NO2 1.0097% (95% CI: 1.0091, 1.010) and EC 1.22% (95% CI: 1.20, 1.25). We found evidence that both race and poverty modify the associations.

Conclusion

Prolonged exposure to traffic-related air pollution contributes to increased rates of all-cardiovascular hospitalization in 14 states, affecting populations even in areas with low NO2 levels. Vulnerability to CVD hospitalization persisted among Black individuals and in regions characterized by elevated poverty levels.