Finer Particulate Matter Exposure Disparities Exist but Vary across Pollution Concentrations

Finer particulate matter plays a critical role in human health. Existing studies generally used annual means to investigate exposure disparities, overlooking intra-annual disparity variations. This study evaluated the distributions of hourly PM_2.5 and PM₁ related to population compositions (i.e., Black, Hispanic, education attainment, and poverty) in the contiguous United States. We observed disparities throughout pollution concentrations, which were not uniformly distributed. Compared to counties with low shares of subpopulations (i.e., the 10th percentile of shares), counties with high shares (i.e., the 90th percentile) experienced changes in PM_2.5 by 11.22 μg/m³ (95% confidence interval: 9.99, 12.48), -10.63 μg/m³ (-11.68, -9.52), 8.81 μg/m³ (7.56, 10.06), and 2.27 μg/m³ (0.94, 3.53) when pollution was extremely high (i.e., the 99th percentile of pollution) for Black, Hispanic, low-education, and poor populations, respectively, and smaller disparities when pollution was at the median 3.22 μg/m³, -1.73 μg/m³, 1.71 μg/m³, and -0.30 μg/m³, respectively. Generally, higher shares of Black or low-education populations or lower shares of Hispanic populations experienced higher exposures, while the relation with poverty was not unidirectional. For comparison, using annual means with single-variable linear regression may underestimate the hourly disparities by more than 70% when pollution was extremely high or even in the opposite direction for poor populations. Such nonuniform exposure disparities should be considered to accurately evaluate pollution-related health disparities and facilitate pollution regulation policies.