Racial and Ethnic Disparities in Regulatory Air Quality Monitor Locations in the US.

Importance: Understanding exposure to air pollution is important to public health, and disparities in the spatial distribution of regulatory air quality monitors could lead to exposure misclassification bias. Objective: To determine whether racial and ethnic disparities exist in Environmental Protection Agency (EPA) regulatory air quality monitor locations in the US. Design, Setting, and Participants: This national cross-sectional study included air quality monitors in the EPA Air Quality System regulatory monitoring repository, as well as 2022 American Community Survey Census block group estimates for racial and ethnic composition and population size. Bayesian mixed-effects models of the count of criteria pollutant monitors measuring an area were used, adjusting for population size and accounting for spatial autocorrelation. Data were analyzed from March to June 2024. Exposure: Census block group-level racial and ethnic composition. Main Outcome and Measures: Number of regulatory monitors measuring a census block group by criteria pollutant (particulate matter [PM], ozone [O3], nitrogen dioxide [NO2], sulfur dioxide [SO2], lead [Pb], and carbon monoxide [CO]). Results: This analysis included 329 725 481 individuals living in 237 631 block groups in the US (1 936 842 [0.6%] American Indian and Alaska Native, 18 554 697 [5.6%] Asian, 40 196 302 [12.2%] Black, 60 806 969 [18.4%] Hispanic, 555 712 [0.2%] Native Hawaiian and Other Pacific Islander, 196 010 370 [59.4%] White, 1 208 267 [0.3%] some other race, and 10 456 322 [0.4%] 2 or more races). Adjusting for population size, monitoring disparities were identified for each criteria pollutant. Relative to the White non-Latino population, all groups were associated with fewer NO2, O3, Pb, and PM monitors. Disparities were consistently largest for Native Hawaiian and Other Pacific Islander populations, followed by American Indian and Alaska Native populations and those of 2 or more races. An increase in percentage of Native Hawaiian and Other Pacific Islander race was associated with fewer monitors for SO2 (adjusted odds ratio [aOR], 0.91; 95% BCI, 0.90-0.91), CO (aOR, 0.95; 95% BCI, 0.94-0.95), O3 (aOR, 0.95; 95% BCI, 0.94-0.95), NO2 (aOR, 0.97; 95% BCI, 0.91-0.94), and PM (aOR, 0.96; 95% BCI, 0.95-0.96). An increase in the percentage of those of Asian race was associated with slightly more SO2 (aOR, 1.04; 95% BCI, 1.03-1.04) monitors. Conclusions and Relevance: This cross-sectional study of racial and ethnic disparities in the location of EPA regulatory monitors determined that data may not be equitably representative of air quality, particularly for areas with predominantly Native Hawaiian and Other Pacific Islander or American Indian or Alaska Native populations. Integration of multiple data sources may aid in filling monitoring gaps across race and ethnicity. Where possible, researchers should quantify uncertainty in exposure estimates.