Skip to main content
Skip to main content

Abstract

Background:

Ambient nitrogen dioxide (NO2) and fine particulate matter with aerodynamic diameter 2.5μm (PM2.5) threaten public health in the US, and systemic racism has led to modern-day disparities in the distribution and associated health impacts of these pollutants.

Objectives:

Many studies on environmental injustices related to ambient air pollution focus only on disparities in pollutant concentrations or provide only an assessment of pollution or health disparities at a snapshot in time. In this study, we compare injustices in NO2- and PM2.5-attributable health burdens, considering NO2-attributable health impacts across the entire US; document changing disparities in these health burdens over time (2010–2019); and evaluate how more stringent air quality standards would reduce disparities in health impacts associated with these pollutants.

Methods:

Through a health impact assessment, we quantified census tract-level variations in health outcomes attributable to NO2 and PM2.5 using health impact functions that combine demographic data from the US Census Bureau; two spatially resolved pollutant datasets, which fuse satellite data with physical and statistical models; and epidemiologically derived relative risk estimates and incidence rates from the Global Burden of Disease study.

Results:

Despite overall decreases in the public health damages associated with NO2 and PM2.5, racial and ethnic relative disparities in NO2-attributable pediatric asthma and PM2.5-attributable premature mortality have widened in the US during the last decade. Racial relative disparities in PM2.5-attributable premature mortality and NO2-attributable pediatric asthma have increased by 16% and 19%, respectively, between 2010 and 2019. Similarly, ethnic relative disparities in PM2.5-attributable premature mortality have increased by 40% and NO2-attributable pediatric asthma by 10%.

Discussion:

Enacting and attaining more stringent air quality standards for both pollutants could preferentially benefit the most marginalized and minoritized communities by greatly reducing racial and ethnic relative disparities in pollution-attributable health burdens in the US. Our methods provide a semi-observational approach to track changes in disparities in air pollution and associated health burdens across the US. https://doi.org/10.1289/EHP11900

Introduction

Ambient nitrogen dioxide (NO2), a marker for the complex mixture of traffic-related pollution, and fine particulate matter with aerodynamic diameter 2.5μm (PM2.5) pose pernicious threats to public health.1 Exposure to PM2.5 has a well-established association with premature death due to several specific causes,24 and recent studies have found moderate to high confidence linking NO2 with new-onset pediatric asthma.59 While levels of these pollutants have decreased in the US following the passage of the Clean Air Act, its 1990 Amendments, and other regional measures,10 PM2.5 and NO2 continue to impact public health and lead to loss of human life. The economic value of these health effects is very high, with the value of statistical life estimated at nearly $10 million per statistical death in 2019 USD.11,12 Systemic racism embedded within the fabric of urban planning and land use in the US has led to modern-day disparities in exposure to these pollutants and their associated health impacts.
While studies consistently show that racialized and minoritized communities face higher levels of NO2 and PM2.5, recent work has led to different conclusions regarding whether relative PM2.5 exposure disparities are narrowing, remaining constant, or widening due, in part, to the different methods employed to assess disparities.1315 Many previous studies have only focused on disparities in pollutant exposure,13,14,1618 leaving a gap in understanding disparities in pollution-attributable health impacts in which exposure disparities may be amplified by variation in underlying disease rates and demographic factors. Despite the association of NO2 with one of the most inequitably distributed diseases, pediatric asthma, no study has examined disparities in NO2-attributable pediatric asthma across the entire US and their changes over time. Thus, there is a need to understand the public health burdens associated with ambient PM2.5 and NO2 across the US and track associated disparities with time, especially as commitments to address environmental justice require concerted efforts to identify and map areas burdened by injustices and inequities.19
Here, we conduct a comprehensive assessment of disparities in public health burdens due to NO2 and PM2.5 across the 50 US states; Washington, DC; and Puerto Rico. Recently developed datasets, which fuse satellite data with physical models, enable us to resolve neighborhood-level differences in NO2 and PM2.5 and thereafter assess inequities in cardiovascular, respiratory, and metabolic health outcomes attributable to these pollutants using demographic data and the latest epidemiological evidence linking exposure with health outcomes. The main contributions of our work are threefold. First, we compare injustices in NO2- and PM2.5-related health burdens. Second, we track how racial and ethnic disparities in the health impacts attributable to these pollutants have changed over the last decade, a period of declining emissions from multiple polluting source sectors. Finally, we explore the degree to which more stringent NO2 and PM2.5 ambient air quality standards could reduce inequitable pollution-related health burdens for the most racialized and minoritized communities in the US.

Methods

Population and Demographic Data

The US Census Bureau’s American Community Survey (ACS) provides estimates of the population, age structure, and demographics within the 74,000 census tracts in the fifty US states, the District of Columbia, and Puerto Rico.20 We specifically used ACS 5-year estimates for our study, which fuse data from decennial censuses and postcensal estimates in years following the last decennial census. Our analysis begins with 2006–2010 ACS 5-year estimates and includes all subsequent 5-year estimates through 2015–2019. Therefore, all demographic data in this study are based on the 2010 Census and corresponding tract boundaries, obtained from the US Census Bureau’s TIGER/Line geodatabase.21 These ACS estimates represent data from their full 5-year time periods; but, for brevity, we refer to them by the final year of the period (e.g., 2006–2010 ACS estimates are referred to as “2010”). Five-year estimates have a larger sample size and smaller margin of error than other ACS estimates with shorter timeframes, and the US Census Bureau weights responses to the ACS by race, age, and differing response rates by building type and census tract to correct for nonresponse biases. While ACS also provides 1-year estimates and provided 3-year estimates through 2013, these other estimates are only available for administrative areas with populations >65,000 and 20,000, respectively.

Pollutant Concentrations

Surface-level NO2 and PM2.5 concentrations were derived from two existing global datasets that combine physical models with satellite retrievals to produce high-fidelity 0.01°×0.01° (1km×1km) estimates of these deleterious pollutants.22,23 We used annual average concentrations from 2010 to 2019, consistent with the years for which demographic data are available and treated these concentrations as surrogates for exposure to NO2 and PM2.5. Our consideration of only NO2 and PM2.5 stems from the fact that, of the common (criteria) air pollutants, PM2.5 exposure leads to the greatest loss of life,1 while NO2 is the most inequitably distributed among racial and ethnic minorities.11 We do not consider other pollutants such as ozone or carbon monoxide, noting that these pollutants also have health impacts24,25 but exhibit less spatial heterogeneity.
While satellite data were used in both datasets to estimate NO2 and PM2.5, the two datasets incorporate these data using different methods. The 0.01°×0.01° NO2 dataset uses a land-use regression model from Larkin et al.26 that predicts mean concentrations for the period 2010–2012 and scales these concentrations to prior and subsequent years using annual average NO2 column densities from NASA’s Ozone Monitoring Instrument satellite.23 The 0.01°×0.01° PM2.5 dataset (V5.GL.02) combines aerosol optical depth retrievals from several satellites with GEOS-Chem chemical transport model output to represent PM2.5 concentrations27 and calibrates these estimates to ground-based PM2.5 observations using geographically weighted regression. Resultant surface-level PM2.5 estimates are thereafter aggregated to annual mean values.28 Text S1 further describes advantages to using these spatially complete datasets, and Text S1 and Figure 1 detail their performance compared with in situ monitors.
Figure 1. Location of in situ (A) nitrogen dioxide (NO2) and (B) fine particulate matter (PM2.5) monitors. Monitor locations represent the Air Quality System (AQS) monitor network during 2019. Scatterplots are colored by density and show a comparison of (C–E) NO2 from Anenberg et al.29 and (F–H) PM2.5 from van Donkelaar et al.30 against observations for 2010, 2015, and 2019. Dataset values represent census tract averages in the tract coincident with the AQS monitor. The reduced major axis linear regression is denoted by the blue lines (C–H). Inset text in the scatterplots indicates the slope (m) and intercept (b) of the regression, the number of in situ monitors (N), the normalized mean bias (NMB), and correlation coefficient (r). Monitors in panels A and B are colored by the difference between the observed and dataset values (<0 corresponds to an overestimate by the datasets). Note: PM2.5, fine particulate matter with aerodynamic diameter 2.5μm; ppbv, parts per billion by volume.

Risk and Rates

In this study, we used cause-specific relative risk (RR) curves from the Global Burden of Disease (GBD) 2021 and mortality and asthma incidence rates from GBD 2019. The GBD is an ongoing multinational research collaboration that assesses morbidity and premature mortality from several risk factors, including ambient air pollution. GBD estimates are updated annually, and recent GBD releases have included several methodological updates that improved upon earlier estimates.31 Rates from GBD 2021 were not yet available at the completion of this study.
RR curves measuring the association of long-term PM2.5 exposure with premature death and NO2 exposure with new cases of pediatric asthma were estimated from systematic reviews and meta-regression based on a Bayesian, regularized, trimmed approach (Figure S1; Excel Table S1).31,32 This approach incorporates RR curves from studies that controlled for confounders including age, sex, education, and income.31 We base our approach, similar to that of the GBD, on relationships between specific pollutants and distinct health outcomes, incorporating an explicit assumption that biological relationships between an exposure and health impact do not vary by population or location unless there is compelling evidence of distinct differences. This allows for inclusion of a larger number of studies considered to estimate RR curves, which is a more holistic assimilation of the available evidence. With this approach, differences in attributable burdens are a function of exposure and underlying disease rates, the latter of which reflects differences in lifestyle, medical care, genetics, and nutrition.
We applied RR curves for NO2-attributable pediatric asthma uniformly to the population aged 0 to 18 across the US. Our consideration of this risk-outcome pair is based on a large body of scientific literature linking NO2 exposure with pediatric asthma development.6,33,34 We included PM2.5-attributable premature mortality for six different end points in our study: chronic obstructive pulmonary disease (COPD); ischemic heart disease; ischemic and intracerebral hemorrhagic stroke (“stroke”); lung, tracheal, and bronchial cancer (“lung cancer”); lower respiratory infection; and type 2 diabetes. RR curves for lower respiratory infection were applied uniformly to the entire population across the US, while the other premature mortality end points were applied to the population aged 25 years and older. These six premature mortality end points (along with low birthweight and short gestation, which we do not consider in this study) represent risk-outcome pairs associated with pollution exposure for which the GBD currently estimates RR curves. We note that other meta-analyses suggest that the impacts of PM2.5 on health may extend beyond these six outcomes.9 Similarly, a 2022 Health Effects Institute (HEI) report detailed several additional outcomes potentially associated with NO2 exposure beyond pediatric asthma development such as new-onset adult asthma and all-cause, circulatory, and ischemic heart disease mortality.9 By not considering these additional outcomes and only using the NO2-pediatric asthma risk-outcome pair from GBD 2020, we are likely underestimating the public health damages associated with NO2.
The uncertainty interval for the RR curves for pediatric asthma conferred by NO2 exposure spans 1 due to between-study heterogeneity unexplained by study design (Excel Table S1). Note that between-study heterogeneity is typically not considered in other health impact analyses, even though heterogeneity is high in most air pollution meta-analyses. Despite the uncertainty interval spanning 1, the mean risk association indicates increasing risk of new-onset pediatric asthma with NO2 and was significant and therefore met criteria for inclusion in the GBD. Additionally, the HEI classified the association of new-onset pediatric asthma with traffic-related air pollution as having medium to high confidence, and the HEI’s NO2-pediatric asthma RR curves had an uncertainty interval that only marginally spanned 19 even when between-study heterogeneity was not accounted for. We chose not to use statistical significance as the sole determining factor for inclusion in our study because this reliance for the convenience of statistical properties may neglect historically excluded groups.35
The theoretical minimum risk exposure levels (TMREL) for PM2.5 and NO2, the level below which we assume no increased risk of PM2.5-attributable premature mortality or NO2-attributable pediatric asthma, are modeled by the GBD as uniform distributions bounded by the minimum and fifth percentiles of exposure distributions from ambient air pollution cohort studies.31 We treated the midpoints of these distributions (i.e., 4.15μgm3 for PM2.5, 5.37 ppbv for NO2) as our TMRELs.
We obtained mortality rates per 100,000 population for COPD, ischemic heart disease, stroke, lung cancer, lower respiratory infection, and type 2 diabetes and incidence rates per 100,000 for the pediatric population for asthma from the GBD 2019 study for each year and state in our analysis (Figure S2). The GBD study estimates the burden of disease for our end points using data extracted from sources such as censuses, disease registries, civil registration and vital statistics, household surveys, and other sources. These data were mapped by the GBD to a cause list, and unspecified cases or deaths or those with inadequate specification were reassigned to probable causes using proportionate redistribution based on expert judgement or published studies or statistical algorithms. Nonreference case definitions or measurement methods in these data inputs were addressed by estimating correction factors using network meta-regression. Additional details are found in Vos et al.36 For each end point, the rates vary by 5-year age groups (e.g., <5, 5–9, 10–14, etc.). Incidence rates are generally higher in the Southeastern and Eastern US for most end points, while some end points such as COPD have less consistent spatial heterogeneity and substantially vary, even among bordering states (Figure S2).

Health Impact and Environmental Justice Assessment

To facilitate comparison of pollutant concentrations with the populations they impact, we averaged the NO2 and PM2.5 datasets to underlying census tracts in the US (Figure S3). The median area of all (urban) census tracts is 5.2km2 (3.7km2) and supports this averaging approach. There are, however, 5.2% of tracts too small in area to contain coincident grid cells. Following Kerr et al.,37 we used inverse distance weighting to interpolate pollutant concentrations to the centroid of these small tracts. We generally observed good spatiotemporal agreement between tract-averaged NO2 or PM2.5 and in situ observations from the Environmental Protection Agency’s (EPA) Air Quality System (AQS) network of administrative and regulatory (not low-cost) monitors. We found that the NO2 dataset was biased high compared with observations, with the largest biases in rural areas, and the PM2.5 dataset was biased low (Text S1; Figure 1). The sparse coverage offered by the in situ monitors precludes us from understanding how the performance of the PM2.5 and NO2 datasets vary across census tracts or comprehensively characterizing uncertainty in pollution estimates. However, the complete spatial coverage offered by these datasets allows us to comprehensively assess health and environmental justice impacts.
We conducted a health impact assessment for each year in the period 2010–2019 by calculating the population attributable fraction (PAF), that is, the fraction of the burden of disease that might be attributable to PM2.5 or NO2 exposure, for our end points of interest using the census tract as our unit of analysis. For a given pollutant concentration X in tract t, the PAF was calculated following Ostro38 as
PAF(Xt)={RR(Xt)1RR(Xt)RR(TMREL)1RR(TMREL), for XtTMREL0, for Xt<TMREL.
(1)
The PAF was then used to calculate the total NO2-attributable pediatric asthma burden or PM2.5-attributable premature mortality burden in each tract as
Burdent=popt×PAFt×ks.
(2)
Here, Burden refers to the number of PM2.5-attributable premature deaths or new cases of NO2-attributable pediatric asthma in tract t; pop corresponds to the susceptible population in tract t; k corresponds to baseline incidence and death rates from the GBD; and s corresponds to state, the highest level of granularity available from the GBD. This use of state-level data likely masks fine-scale variations in premature mortality and pediatric asthma rates, and exploring how higher spatial resolution premature death rates impact our findings will be the basis of subsequent discussion. Both cause-specific PM2.5-attributable premature deaths from these six end points and their sum are presented in our analysis. While pollutant concentrations (Xt) and baseline disease rates (ks) in Equations 1 and 2 are based on annual estimates, the population estimates (popt) represent 5-year period estimates. We acknowledge that using these 5-year estimates could introduce temporal inconsistency into our health impact assessment but believe their increased statistical reliability39 and availability at the census tract level justifies their use.
Uncertainty in pollution-attributable health burdens was primarily characterized using the 95% uncertainty interval of RR curves. Other terms in the health impact function (Equation 2) also have uncertainties which are more difficult to quantify. For example, the satellite data and physical models used to estimate NO2 and PM2.5 have known biases (Text S1; Figure 1), but comprehensively characterizing associated uncertainty given the paucity of in situ monitors is not possible. Despite this limitation, the high resolution and complete spatial coverage enable us to conduct a state-of-the-science assessment of exposure and health impacts in individual census tracts. Furthermore, in a global study on NO2-attributable pediatric asthma, Achakulwisut et al.40 investigated the uncertainty in underlying disease incidence rates, finding this source of uncertainty to be the least influential in estimating health burdens, and Ostro et al.41 showed that uncertainty in risk relationships dominated that of exposure. Given the form of Equation 2, we expect any uncertainties in death and incidence rates would linearly scale our results and likely not substantially affect relative differences across demographic groups or overall trends.
We also report results aggregated to metropolitan statistical areas (MSAs) and to the national level. MSAs have at least one urbanized area of 50,000 or more residents and not only encompass cities’ urban cores but also their suburban and peri-urban areas.42 A majority of the US population (89%) lived in one of the 389 MSAs in 2019. We refer to MSAs by their colloquial names (e.g., Los Angeles-Long Beach-Anaheim, CA MSA = Los Angeles).
Race and ethnicity are distinct social constructs that capture unique information related to health-associated exposures and outcomes.43 While many different measures of race and ethnicity exist in epidemiological research,43 this study relies on the Office of Management and Budget definition, to which the US Census Bureau adheres. Regarding their racial identity, ACS respondents self-identify from five preset categories: “American Indian or Alaska Native”; “Asian”; “Black or African American”; “Native Hawaiian or Other Pacific Islander”; and “White.” In addition to respondents’ race(s), respondents self-identify their ethnicity as “Hispanic or Latino” or “Not Hispanic or Latino.” Following this distinction, we characterized environmental injustices stemming from PM2.5, NO2, and the associated health burdens for both racial and ethnic groups using two complementary methods:
1.
Top and bottom deciles of population subgroups. Census tracts were designated as the “most white” and “least white’’ or “most Hispanic” and “least Hispanic” using the top and bottom 10% (decile) of the white or Hispanic population distribution. This approach allows us to understand pollution-attributable health burdens in the most minoritized communities of the US and contrast with the burdens experienced by majority communities and has been previously used in the literature.15,37,44 These population subgroups based on deciles do not include only tracts in certain states or geographic regions. For example, the 7,330 census tracts that comprise the most white and least white classifications in the US include tracts from 52 and 49 states, territories, or districts. This subgroup definition does reflect urban-rural population differences to a certain extent when urban tracts are defined as those lying within the boundaries of MSAs. With this definition, 60% of the most white and 90% of the least white tracts are considered urban.
2.
Population-weighted. Population-weighted metrics were calculated with the following equation:
Xg=t=iNpopt,g×Xtt=iNpopt,g,
(3)
where X represents pollutant concentrations or disease rates, pop represents the population, g represents a population subgroup, and t represents a census tract.
The population age structure varies between top and bottom decile subgroups (Figure S4). Presenting NO2-attributable pediatric asthma crude rates or PM2.5-attributable premature mortality crude rates does not account for these different age distributions (Text S2). Whenever rates are presented for the top and bottom deciles of population subgroups, they represent age-standardized rates directly adjusted to the entire US population corresponding to the same year. Age standardization was conducted by multiplying each 5-year age-specific rate by the fraction of 5-year age group population to the entire US population. Since the standardization is to the national population, it does not account for subnational differences in the age distribution.
We conducted several scenarios where NO2 and PM2.5 reach target concentrations to assess how meeting these targets will reduce the associated health burdens and potentially advance environmental justice. Targets represent the National Ambient Air Quality Standards (NAAQS) established by the EPA and the World Health Organization (WHO) interim targets (ITs) and air quality guidelines (AQGs), updated in 2021.45 If NO2 or PM2.5 concentrations in a particular tract were larger than a target level, we assigned the concentration to the target value.
We tested whether differences in distributions of pollutants and associated disease burdens significantly vary across different ethnic and racial groups with the nonparametric Kolmogorov-Smirnov test. The significance of trends in pollutants, burdens, and disparities was assessed with least-squares regression. If the p-values associated with the Kolmogorov-Smirnov test statistic or the regression fell below 0.05, we classified the difference between distributions or trends as statistically significant.
Costs associated with PM2.5-attributable premature deaths were estimated with the EPA’s value of statistical life used for valuing mortality risk changes ($7.4 million in 2006 USD or $9.4 million in 2019 USD).46 This value represents the marginal rate of substitution between money and small changes in the risk of death. The body of literature on the economic burden of pediatric asthma is limited, but a 2018 study synthesized publications reporting on health care costs and health care utilization for pediatric asthma and found average annual costs per child ranged from $3,076 to $13,612 in 2015 USD.47 We used the midpoint of these values, adjusted for inflation to 2019 USD, as our estimate ($8,473).
We provide supplementary data files with information on tract-level NO2 and PM2.5 concentrations and pollution-attributable health burdens and rates for 2019 to make data used in this study accessible for stakeholders and enable scientific transparency and reproducibility. These files and their contents are described in Text S3.

Sensitivity Analysis Considering High-Resolution Incidence Rates

While our main analysis relies on the aforementioned state-specific rates, which represent the highest level of granularity currently available from the GBD, incidence rates of morbidity- and mortality-related outcomes have been shown to vary on neighborhood scales with higher values in areas with lower socioeconomic status and a higher percentage of minorities. Recently, the EPA’s Benefits Mapping and Analysis Program—Community Edition (BenMAP-CE) software used to estimate air pollution-related health impacts has included estimates of census tract all-cause mortality rates using life tables from the US Small-Area Life Expectancy Estimates Project (USALEEP).48 These rates are based on death records over the period 2010–2015.
We use these tract-level incidence rates and investigate how they affect ethnoracial disparities. Since these rates represent all-cause mortality (rather than cause-specific mortality), we combine these higher-resolution incidence rates with all-cause RR estimates from Turner et al.49 of 1.06 per 10μgm3 annual average PM2.5. We did not apply a TMREL when calculating all-cause premature mortality with the Turner et al.49 RR estimates. Therefore, we did not expect the total number of PM2.5-attributable premature deaths and ethnoracial absolute disparities to match the results presented elsewhere in the main text; however, we hypothesize that examining the relative disparities using these different methods will allow us to test whether our results are robust to different incidence rates and RR estimates.
In this sensitivity analysis, we calculate national-level ethnoracial relative disparities for the following cases:
1.
Turner et al.49 RR estimates with state-level all-cause mortality rates. Burdens and rates are calculated for the population aged 30 and older for each 5-year age group (30–34, 35–39, etc.) and thereafter standardized to account for differences in the age structure across population subgroups.
2.
Same as 1, but no age standardization is applied.
3.
Turner et al.49 RR estimates with tract-level all-cause mortality rates. Tract-level rates from Raich et al.48 are available for 10-year age groups (25–34, 35–44, etc.), so we apply the RR estimates to the population aged 25 and older in 10-year age groups and standardize for different age structures. We acknowledge that disparities calculated with these methods (for population aged 25 and older) are not directly comparable with the disparities from 1 and 2 (for population aged 30 and older); however, we expect differences to be minimal.
4.
Same as 3 but no age standardization is applied.
Based on the period represented by USALEEP tract-level rates, all results for this sensitivity test represent 2015 values, and the age structure is standardized to the full US population for that year.

Results

As cause-specific premature mortality rates and PM2.5 have declined (Text S1; Figure 2A; Figure S2), total PM2.5-attributable deaths across the 50 US states; Washington, DC; and Puerto Rico have decreased by 28.5% from 69,000 (48,500–87,000) in 2010 to 49,400 (34,500–62,600) in 2019 (Figure 2A). New cases of NO2-attributable pediatric asthma have declined by an even larger percentage, 39.8%, from 191,000 (282,900–407,900) in 2010 to 114,900 (158,600–259,400) in 2019 (Figure 2B) even with positive trends in pediatric asthma incidence in all states besides Puerto Rico. This wide uncertainty interval in estimated NO2-attributable pediatric asthma cases stems from between-study heterogeneity (“Methods”). Rates of PM2.5-attributable premature death and NO2-attributable pediatric asthma, which account for the changing US population during this 10-year period, have similarly declined by 34.8% and 37.4%, respectively (Figure S6A,B).
Figure 2. Annual (A) PM2.5 and PM2.5-attributable mortality and (B) NO2 and NO2-attributable pediatric asthma in all 50 US states; Washington, DC; and Puerto Rico. Pollution concentrations in the black dots connected by lines represent population-weighted mean values, while health outcomes represent sums. Stacked bars in panel A, from the bottom to top, represent ischemic heart disease, stroke, lung cancer, COPD, type 2 diabetes, and lower respiratory infection. Note: COPD, chronic obstructive pulmonary disease; PM2.5, fine particulate matter with aerodynamic diameter 2.5μm; ppbv, parts per billion by volume.
The monetary value attributed to mortality risk (value of a statistical life) for premature deaths due to PM2.5 and the estimated direct costs of NO2-attributable pediatric asthma during 2019 translate to $466 billion in 2019 USD, roughly 2.2% of the 2019 US gross domestic product. These total costs were dominated by PM2.5-attributable mortality, which accounted for $464 billion, or 99.7%, of the estimated costs. Although the value of a statistical life and direct costs are fundamentally different quantities (the former representing a willingness to pay for a small reduction in mortality risk and the latter representing the costs directly associated with patient care), they nonetheless speak to the substantial economic consequences of PM2.5 and NO2 pollution in the US.
PM2.5-attributable mortality rates were generally highest in MSAs in the Ohio River Valley and Gulf Coast (Figure 3A). MSAs with the 10 highest rates (Birmingham, AL; Mobile, AL; Gulfport, MS; Evansville, IN; Daphne, AL; Punta Gorda, FL; Mansfield, OH; Weirton, WV; Hot Springs, AR; and Kokomo, IN) are generally located in these two regions, which contain heavy manufacturing and petrochemical industries. The PM2.5-attributable premature mortality rate averaged over these 10 metropolitan areas was 42.1 deaths per 100,000, nearly double the rate averaged over all MSAs (21.5 deaths per 100,000). These increased rates stem from the high population-weighted PM2.5 concentrations in these MSAs (8.25μgm3 vs. the MSA average of 6.96μgm3) and elevated underlying mortality rates (the rates summed over our six premature mortality end points in states containing these MSAs is 1,070 deaths per 100,000 vs. the US-average of 905 deaths per 100,000; Figure S2).
Figure 3. Bubble plot points represent individual metropolitan statistical areas (MSAs), and their size is proportional to the (A) PM2.5-attributable deaths in 2019 per 100,000 population and (B) new cases of NO2-attributable pediatric asthma in 2019 per 100,000 pediatric population. Rates are discretized into five categories: <30th, 30–60th, 60–90th, 90–95th, and >95th percentiles of MSA rates. Alaska, Hawaii, and Puerto Rico are not to scale. Note: PM2.5, fine particulate matter with aerodynamic diameter 2.5μm.
NO2-attributable pediatric asthma rates in MSAs have more spatial heterogeneity than PM2.5-attributable rates, and even relatively isolated MSAs can experience higher-than-average rates (Figure 3B). The highest NO2-attributable pediatric asthma rates mostly occurred in the largest MSAs in the US and in MSAs near oil and gas production (Figure 3B). MSAs with the 10 highest rates include New York, NY; Chicago, IL; Houston, TX; Boston; MA; Detroit, MI; Lubbock, TX; Fargo, ND; Midland, TX; Odessa, TX; and Grand Forks, ND. Rates in these MSAs (316.1 cases per 100,000 children) were over three times higher than the rate averaged over all MSAs (93.4 cases per 100,000 children).
Among the most salient features in Figure 3B are the large NO2-attributable asthma rates in the Permian Basin. The NO2-attributable asthma rate averaged over the five largest MSAs in the Permian Basin (El Paso, TX; Lubbock, TX; Amarillo, TX; Midland, TX; and Odessa, TX) was 252.8 new cases of asthma per 100,000 children. This rate is nearly four times higher than the aforementioned rate averaged over all MSAs and even slightly greater than the rate in nearby Dallas-Fort Worth, TX (248.2 cases per 100,000 children). Oil and gas production in the Permian Basin has been linked to elevated levels of NO2, methane, and volatile organic compounds50,51 and increased pediatric asthma hospitalizations.52
Despite long-term decreases of PM2.5 and NO2 (Figure 2), the least white and most Hispanic communities consistently faced higher concentrations of PM2.5 and NO2 than the most white and least Hispanic communities in the 2010s (Figure 4; Figure S7). During this decade, trends in concentration disparities, which are characterized using the top and bottom deciles of racial and ethnic population subgroups, differ depending on the pollutant. In the case of PM2.5, we find that racial and ethnic relative disparities—that is, the ratio of concentrations for the minority subgroup to the majority subgroup—have increased, while disparities in NO2 concentrations have decreased over time (Figure S7). Disparities were substantially larger for NO2 than PM2.5 and for different racial subgroups than ethnic subgroups. For example in 2019, racial relative disparities in PM2.5 were 1.16, meaning that the least white communities in the US faced higher concentrations of PM2.5 by 16% compared with the most white communities. Ethnic PM2.5 disparities were 1.17. Conversely, racial and ethnic relative NO2 disparities were 2.11 and 1.69, respectively (Figure S7).
Figure 4. Racial and ethnic disparities in pollutant concentrations and associated pollution-attributable health burdens calculated for 2019 using the top and bottom deciles of population subgroups. Large scatter points correspond to concentrations or burdens for population subgroups calculated with all census tracts, while smaller jittered points correspond to these quantities in individual MSAs of the US. The row shaded in gray indicates that the difference between the MSA distributions is not significant (p=0.09), determined with the Kolmogorov-Smirnov test. Note: PM2.5, fine particulate matter with aerodynamic diameter 2.5μm; ppbv, parts per billion by volume.
The least white communities in the US experienced higher rates of cause-specific premature mortality attributable to PM2.5 from all end points compared to the most white communities in 2019, and relative disparities have a range of 1.25–1.41, depending on the specific end point considered (Figure 4A). Ethnic relative disparities exhibit a wider range (0.89–1.29; Figure 4B). We found that the least Hispanic communities in the US experienced slightly higher PM2.5-attributable premature mortality rates from COPD and lung cancer than the most Hispanic communities (Figure 4B). This finding could be related to the Hispanic paradox, in which Hispanic communities in the US have been shown to have higher prevalence of several risk factors related to premature mortality and yet lower mortality rates than non-Hispanic whites.53
Racial and ethnic disparities in NO2-attributable pediatric asthma are striking (Figure 4). NO2-attributable asthma rates in the least white and most Hispanic communities of the US were higher than rates in the most white and least Hispanic communities by a factor of 7.47 and 3.16 in 2019, respectively. In 28.8% of MSAs, all census tracts designated as most white had zero cases of NO2-attributable pediatric asthma (i.e., NO2 concentrations fell below the TMREL). This lack of NO2-attributable asthma cases in the least white tracts only occurred in 2.3% of MSAs.
While NO2-attributable pediatric asthma and PM2.5-attributable premature mortality rates have decreased across the US over the last decade, the magnitude of these decreases has not been uniform (Figure 5). Relative decreases in majority white and non-Hispanic communities outpaced relative decreases in majority nonwhite and Hispanic communities. As a result, relative racial disparities in NO2-attributable pediatric asthma have increased from a factor of 6.3 difference between most and least white communities in 2010 to a factor of 7.5 difference in 2019 (19% increase; Figure 5A). Similarly, relative racial disparities in PM2.5-attributable premature mortality grew by 16%, ethnic disparities in PM2.5-attributable premature mortality by 40%, and ethnic disparities in NO2-attributable pediatric asthma by 10%.
Figure 5. Trends in (A and B) NO2-attributable pediatric asthma and (C and D) PM2.5-attributable premature mortality rates for the most and least white and Hispanic tracts in the US. Black time series and corresponding text beneath each panel indicate the relative disparities, defined as the ratio of the rate for the bottom decile population subgroup (least white, most Hispanic) to the rate for top decile (most white, least Hispanic). A value of 1 for relative disparities implies that pollution-attributable burdens are equally shared across subgroups. For reference, rates for the first and last years of the analysis are indicated alongside the scatter points. Note: PM2.5, fine particulate matter with aerodynamic diameter 2.5μm.
At the beginning of the decade, the most Hispanic communities in the US faced lower PM2.5-attributable death rates (Figure 5D), similar to our findings for some cause-specific end points in Figure 4B. However, the ethnic subgroup most burdened with respect to PM2.5-attributable premature mortality reversed around 2015 (Figure 5D). By 2019, the most Hispanic communities had 8% higher PM2.5-attributable premature mortality rates than the least Hispanic communities. While racial and ethnic relative disparities in PM2.5-attributable mortality are generally around 1 (equality between subgroups) and therefore small compared to disparities in NO2-attributable pediatric asthma rates, their increasingly inequitable distribution is noteworthy. If trends in relative disparities of pollution-attributable pediatric asthma and premature mortality during the 2010s are an indication of the future, we expect that these relative disparities will continue to grow.
Examining trends and disparities in PM2.5 and NO2 concentrations can shed light on the drivers of the widening disparities in their attributable health impacts. Increasing relative disparities in PM2.5-attributable premature mortality mirror the increasing disparities that we found in PM2.5 concentrations; however, the increasing relative disparities in NO2-attributable pediatric asthma occurred as disparities in NO2 concentrations decreased (Figures 5; Figure S7). Differing trends for relative disparities in NO2 concentrations vs. NO2-attributable pediatric asthma suggest that changes in the population distribution and underlying pediatric asthma rates have counteracted the decreasing relative disparities in NO2 concentrations. Although these rates only vary by state (Figure S2S–U), a larger number of tracts belonging to a particular decile population subgroup located in a state in which pediatric asthma incidence has exhibited a greater increase relative to other states would lead to increasing relative disparities.
While our main focus has been on relative disparities among different population subgroups characterized by the top and bottom deciles (e.g., most and least white or Hispanic), alternative ways of defining disparities and population subgroups exist. When examining relative disparities with population-weighted pollutant concentrations and pollution-attributable disease rates (Figure S8), we find results consistent with Figure 4, although the magnitude of these population-weighted disparities was slightly smaller than disparities estimated using the top and bottom deciles. Trends in population-weighted relative disparities are more mixed than the trends calculated with top and bottom deciles (Figure S9). Racial relative disparities in population-weighted pollution-attributable asthma and premature mortality have nonsignificant positive trends, ethnic premature mortality disparities a significant positive trend, and ethnic asthma disparities a significant negative trend (Figure S9). When examining absolute disparities—that is, the difference and not the ratio—in concentrations or pollution-attributable disease rates between different population subgroups, we find mixed results (Figure S10). Absolute disparities in NO2 exposure and NO2-attributable pediatric asthma rates have decreased during the 2010s regardless of how population subgroups are defined (Figure S10A,B,E,F). However, racial absolute disparities in PM2.5 exposure and PM2.5-attributable premature mortality have generally remained constant (Figure S10C,G), while ethnic absolute disparities have grown (Figure S10D,H).
The NAAQS do not adequately protect the public from the adverse effects of PM2.5 and NO2 based on our own assessment of the health burdens that occur when the NAAQS were attained in the vast majority of tracts (Figure 6) as well as numerous toxicological, clinical, and epidemiological studies that highlight health effects of these pollutants at levels below the current NAAQS.5456 The current annual PM2.5 NAAQS of 12μgm3, last revised in 2012, was met in all but 486 (0.7%) of census tracts in 2019, and the highest 2019 NO2 concentration in all census tracts of the US (28.3 ppbv) was about half the annual NO2 NAAQS of 53 ppbv, which has not been revised since 1971. Yet, Figure 2A,B highlights the major public health damages and racial and ethnic disparities associated with these pollutants even when the NAAQS are attained. This result complements a recent study by Wang et al.18 that showed an NAAQS-like approach may not eliminate disparities since achieving reductions in one overburdened area would also benefit other, less-burdened areas.
Figure 6. Air quality, health, and environmental justice benefits achieved by attaining (A) PM2.5 and (B) NO2 standards in tracts where pollutant concentrations exceeded these standards in 2019. PM2.5 standards include the Environmental Protection Agency (EPA) National Ambient Air Quality Standard (NAAQS) of 12μgm3, the World Health Organization (WHO) Interim Target 4 (IT-4) of 10μgm3, the lower bound of the recommended range (810μgm3) recommended by the Clean Air Scientific Advisory Committee (CASAC) in their March 2022 letter to the EPA Administrator, and the WHO Air Quality Guidelines (AQG) of 5μgm3. NO2 standards include the WHO IT-1 of 21.3 ppb (assuming an ambient pressure of 1,013.25 hPa and temperature of 298.15 K), the WHO IT-2 of 16 ppb, the WHO IT-3 of 10.6 ppb, and the WHO AQG of 5.3 ppb. Stacked bars in panel A, from the bottom to top, represent ischemic heart disease, stroke, lung cancer, COPD, type 2 diabetes, and lower respiratory infection. Note: COPD, chronic obstructive pulmonary disease; PM2.5, fine particulate matter with aerodynamic diameter 2.5μm.
Enacting and attaining more stringent PM2.5 and NO2 standards could reduce pollution-attributable health burdens, with potentially outsized benefits for communities of color (Figure 6). As an example, we consider how a PM2.5 standard of 8μgm3 could advance environmental justice. This level is the lower end of the range recommended by EPA’s Clean Air Scientific Advisory Committee (CASAC) in March 2022 and slightly lower than the 910μgm3 range the EPA announced in January 2023 in response to CASAC’s recommendation.57 If a new PM2.5 standard of 8μgm3 was adopted and met in all tracts where this level is not currently met, the percentage decrease in PM2.5-attributable premature mortality rates in the least white communities of the US would be roughly four times larger than the percentage decrease in the most white communities (Figure 6A). Similarly, if the WHO interim target-3 (IT-3) was met, total pediatric NO2-attributable asthma burdens would drop by 20%, but the least white communities in the US would experience a fivefold greater reduction in pediatric asthma rates than in the most white communities (32.6% vs. 6.3%; Figure 6B).
Reducing 2019 NO2 and PM2.5 to the stringent WHO AQGs in all tracts where these guidelines are not met would lead to a 73.2% reduction in PM2.5-attributable mortality and eliminate NO2-attributable pediatric asthma (Figure 6). Attaining the AQGs also effectively eliminates the current patterns of injustice by which communities of color experience greater pollution-attributable health burdens. For NO2-attributable pediatric asthma, this elimination of inequity is due to the fact that NO2 concentrations in all census tracts would be below our assumed TMREL, and therefore no health impacts would be incurred. For PM2.5-attributable premature mortality, some health impacts would still be experienced under the AQG scenario since the AQG PM2.5 guideline exceeds our TMREL; however, inequities in PM2.5-attributable premature mortality are eliminated under this AQG scenario since the overburdened, least white tracts experience preferential decreases in PM2.5 that achieve parity with concentrations in the most white tracts (4.9μgm3).

Discussion

Our study documents the substantial impact of air pollution on human health from 2010 through 2019, exploring how communities of color shoulder a disproportionate share of this burden. Results paint a mixed picture of progress: despite overall decreases in NO2 and PM2.5 and associated health impacts during the 2010s, significant racial and ethnic disparities in the health impacts attributable to these pollutants remain. We found that relative disparities in NO2-attributable pediatric asthma are several times larger than relative disparities in PM2.5-attributable premature mortality, and relative disparities in PM2.5 concentrations and pollution-attributable health impacts from PM2.5 and NO2 are widening.
Our finding that disparities in PM2.5 and associated health burdens are growing in spite of overall decreases in attributable burdens is an important conclusion of this study and complements recent work by Jbaily et al.,15 who highlighted increasing PM2.5 disparities among racial and ethnic groups but did not examine associated health impacts. Other studies that have examined trends in disparities in PM2.5 exposure have found different results than our findings and those of Jbaily et al.,15 namely that exposure disparities have declined.13,14 The disagreement among these studies likely comes from a combination of differences in how disparities were defined (i.e., relative vs. absolute metrics), how population subgroups were characterized (i.e., top and bottom deciles vs. population-weighted metrics), and the precise time period examined. One potential explanation for the widening PM2.5 disparities could be the declining importance of the power generation sector.58 The largest benefits of power plant closures have accrued to the white population.44 As the role of the power generation sector on PM2.5 decreased, light-duty and heavy-duty vehicles have become an increasingly important source of primary PM2.5. Our previous work has shown the collocation of marginalized and minoritized neighborhoods with the roadways used by these vehicles.37 The increasing relative disparities in NO2-attributable pediatric asthma that occurred while disparities in NO2 concentrations decreased is another noteworthy finding of this study. Future place-based policies that yield even larger decreases in NO2 concentrations specifically in marginalized and minoritized communities of the US could provide a means to reduce or even reverse the growing disparities in NO2-attributable pediatric asthma.
Measures of socioeconomic status such as educational attainment and income have often been used in environmental justice studies. Here, we have chosen to focus on race and ethnicity as racial and ethnic disparities are not a proxy for socioeconomic disparities. Mikati et al.59 showed that the magnitude of racial disparities exceeds that of disparities based on poverty status. Tessum et al.16 demonstrated that people of color at every income level face disproportionate PM2.5 exposure. Policies to reduce pollution burdens based strictly on socioeconomic status may not do so equitably, thus buttressing our focus on racial and ethnic patterns of injustice.
Systems and practices that introduce and perpetuate systemic racism and discrimination are responsible for these disparities.60 Marginalized and minoritized communities are disproportionately exposed to virtually all major emissions sectors; traffic (particularly heavy-duty diesel traffic), industry, and construction have been pointed out as the most important in explaining PM2.5 and NO2 disparities.16,37,61 Disparities in pollution and associated health impacts have been linked to “redlining,” a practice beginning in the 1930s by which financial services were denied to residents in certain urban areas based on their race or ethnicity.6264 While this discriminatory practice officially ended in 1968, its numerous effects on present-day zoning practices and the placement of highways and industries in racialized and minoritized neighborhoods have been documented.6466
Our assessment of the economic costs caused by PM2.5 and NO2 agrees well with a global economic assessment conducted by Yin et al.,67 who found that PM2.5-attributable economic costs amount to 2.7% of the gross domestic product of the US. Furthermore, premature mortality and pediatric asthma burdens documented in this study generally align with other recent studies.12,23,58,68,69 We note, however, that our estimates are lower. One key reason for this discrepancy is that our TMRELs are higher than many other studies, which assume, for example, that there is no level below which PM2.5 would not increase the risk of death.12,70 If we recalculate the number of PM2.5-attributable premature deaths and NO2-attributable cases of pediatric asthma assuming a TMREL of zero (Figure S6C,D), our estimated burden of disease attributable to these pollutants is more closely aligned with other published studies. For example, Achakulwisut et al.40 found 290,000 new cases of pediatric asthma attributable to NO2 in the US using no TMREL, annual average 2010–2012 NO2 concentrations, and 2015 population counts. Using our methodology but with a TMREL of 0 ppbv, we estimate 349,000 new cases of attributable pediatric asthma in 2015 (Figure S6D). Our consideration of TMRELs also accounts for the existence of nonanthropogenic sources of PM2.5 and NO2 (e.g., biogenic, lightning, sea salt, dust, etc.) that impact health but cannot be readily regulated and controlled in the same sense as anthropogenic PM2.5 or NO2. There is a lack of consensus on how to apply TMRELs and uncertainty surrounding risks of pollution-attributable health impacts at low exposure. For example, the WHO Health Risks of Air Pollution in Europe (HRAPIE) project considers TMRELs for certain pollutants such as PM2.5 but not for others like NO2.71 Approaches employed by the GBD include TMRELs given background levels of air pollutants and lack of knowledge about the shape of the concentration-response function at the lowest exposure levels.72 We acknowledge that specification of the TMRELs will lead to different pollution-attributable disease burdens, and here we have provided burden estimates with and without the application of nonzero TMRELs. A growing number of studies specifically analyzing health effects of pollutants at low concentrations73 will continue to increase the community’s understanding of low-level health effects.
As with many environmental epidemiological studies, our results are susceptible to the ecological fallacy. This study was done at the population level by combining areal data inputs at the census tract and state levels. Therefore, our results are not intended for drawing conclusions about individual-level associations between pollution exposure and health outcomes. There are several potential confounders related to the NO2- and PM2.5-health effects associations. Our health impact and environmental justice assessments do not specifically control for confounding; however, the RR estimates from the GBD study are based on meta-analysis of epidemiological studies that each adjusted for population variations and major confounders, including age, sex, education, and income.31 Still, we acknowledge that the RR estimates used here could also be subject to other confounders, including exposure to indoor air pollution or meteorology (e.g., temperature), that could accentuate or attenuate the precise magnitude of our disease burdens or disparities.
The state-level cause-specific mortality rates and pediatric asthma incidence rates have spatial resolution much coarser than our exposure and demographic data, and their use in our study could impact our results. Underlying incidence rates are often higher in minoritized and marginalized communities70,74 due to, for example, racial differences in access to and quality of health care services that affect severity of disease and risk of mortality for the same pollution concentration. We hypothesize that incorporating rates at a higher spatial resolution would likely increase the magnitude of the disparities uncovered in this study. To test this hypothesis, we considered a recently developed dataset estimating tract-level all-cause mortality rates in 2015 (Text S2). Since these rates represent all-cause mortality rather than the six specific causes we examined in this study, we combined these tract-level rates with all-cause mortality RR estimates from Turner et al.,49 which were also recently used in an EPA review of the NAAQS for PM2.5. Consistent with our hypothesis, we find that using tract-level mortality rates leads to even higher burdens placed on the least white and most Hispanic communities of the US (Text S2; Figure S5). The analysis in Figure S6 also suggests that similar conclusions regarding disparities and the most exposed population subgroup are found using cause-specific RR curves from the GBD or the all-cause RR curves from Turner et al.49 We do not have pediatric asthma incidence rates at the census tract level for a similar sensitivity analysis of NO2-attributable pediatric asthma. Beyond the tract-level all-cause mortality dataset we have considered, other datasets with higher-than-state spatial resolution often have high rates of data suppression for confidentiality reasons, hindering their use in our study.
Our choice of RR estimates could also impact the magnitude of disparities and could explain some of the differences among studies that, as previously mentioned, have at times reached different conclusions regarding the burden of disease attributable to pollution and trends in disparities. We relied on RR estimates uniformly applied to the entire population in this study, but risk may differ among different demographic groups due to social determinants of health or biological differences. Akinbami et al.75 found that children belonging to racial and ethnic minority groups had as high or higher relative risk for asthma diagnoses than non-Hispanic white children, and Spiller et al.76 showed RR estimates applied uniformly to the population, rather than race-ethnicity specific estimates, underestimated pollution-related health impacts for minority communities. However, the literature on this topic is not consistent: Alexeeff et al.77 did not find a difference in the association between exposure to PM2.5 and COPD by race or ethnicity, and Parker et al.78 similarly found that the association between PM2.5 and heart disease mortality was not statistically different for non-Hispanic white adults vs. black or Hispanic adults. Hicken et al.79 reviewed studies examining the association between race and air pollution exposure on adult mortality and determined no consistent role of race in the association between exposure to PM2.5 and mortality. Since the literature remains inconsistent on this topic, it is unknown how race- and ethnicity-specific RR estimates would impact the findings uncovered herein, but it is possible that our use of uniform RR estimates might mask some differences in the pollution-attributable outcomes we investigated. Future work might leverage these race- and ethnicity-specific RR estimates, such as those developed by Di et al.55 from Medicaid-eligible persons in the US.
Ambient PM2.5 and NO2 continue to challenge public health in the US, leading to an estimated 49,400 premature deaths and 114,900 new cases of pediatric asthma, respectively, in 2019. Minoritized, racialized, and marginalized communities in the US persistently experience disproportionate air pollution-attributable disease burdens. Racial and ethnic health disparities due to NO2-attributable pediatric asthma are substantially larger than those from PM2.5-attributable premature mortality, but relative disparities for both of these health outcomes in the most vs. least minoritized communities of the US have widened in the past decade. Alternative ways of defining disparities (e.g., absolute, population-weighted, most vs. least burdened) indicate that the exact sign and significance of trends can be somewhat metric specific. The overall decreases in pollution concentrations; pollution-attributable health burdens; and, in the case of NO2, absolute disparities in exposure and attributable pediatric asthma can be reasonably expected given the focus of the Clean Air Act and its amendments on attaining minimum national standards for air quality rather than on distributional effects. However, recent trends in relative disparities in the US have clearly not matched the obvious macro-level reductions in ambient NO2 and PM2.5 pollution due to the Clean Air Act and related measures.
Increasing the stringency of the NAAQS for PM2.5 and NO2 to be in alignment with the 2021 WHO AQGs could have outsized benefits for marginalized and minoritized communities. Codification and attainment of these AQGs would substantially reduce current patterns of injustice and broadly reduce pollution-attributable health burdens across the nation. Accomplishing sufficient pollution remediation will require reductions from almost every emission sector given their disproportionate impacts on marginalized and minoritized communities. Recent efforts to reduce emissions from transportation (e.g., plug-in electric vehicle tax credits, EPA’s proposed heavy-duty engine and vehicle standards) and rethink land use (e.g., Department of Transportation’s Reconnecting Communities Pilot Program) are steps in the right direction and urgently needed. The investments needed to develop new control technologies and implement other mitigation measures are not trivial. If the benefit-to-cost ratio of the Clean Air Act, which is estimated at thirty to one,80 is any indication, the potential economic benefits of future pollution-reducing policies could be enormous.

Acknowledgments

The authors thank Neal Fann, Maria Harris, and Ananya Roy for their helpful feedback. We gratefully acknowledge the computing resources provided on the High-Performance Computing Cluster operated by Research Technology Services at the George Washington University.81
This study was funded by NASA (grants 836683, 875721, and 80NSSC21K0508).

Article Notes

G.H.K. reports that he has served as a consultant for the Environmental Defense Fund, Department of Justice, and California Air Resources Board. S.C.A. reports that she has served as a consultant on related topics for the Environmental Defense Fund, Department of Justice, and Environmental Integrity Project. The remaining authors report no conflicts of interest relevant to this article.

Supplementary Material

File (ehp11900.s001.original.acco.pdf)
File (ehp11900.smcontents.508.pdf)
File (ehp11900.s001.acco.pdf)
File (ehp11900.s002.codeanddata.acco.zip)

References

1.
Landrigan PJ, Fuller R, Acosta NJR, Adeyi O, Arnold R, Basu NN, et al. 2017. The Lancet commission on pollution and health. Lancet 391(10119):462–512, https://doi.org/10.1016/S0140-6736(17)32345-0.
2.
Dockery DW, Pope CA, Xu X, Spengler JD, Ware JH, Fay ME, et al. 1993. An association between air pollution and mortality in six U.S. Cities. N Engl J Med 329(24):1753–1759. https://pubmed.ncbi.nlm.nih.gov/8179653/, https://doi.org/10.1056/NEJM199312093292401.
3.
Kloog I, Ridgway B, Koutrakis P, Coull BA, Schwartz JD. 2013. Long- and short-term exposure to PM2.5 and mortality: using novel exposure models. Epidemiology 24(4):555–561. https://pubmed.ncbi.nlm.nih.gov/23676266/, https://doi.org/10.1097/EDE.0b013e318294beaa.
4.
Liu C, Chen R, Sera F, Vicedo-Cabrera AM, Guo Y, Tong S, et al. 2019. Ambient particulate air pollution and daily mortality in 652 cities. N Engl J Med 381(8):705–715. https://pubmed.ncbi.nlm.nih.gov/31433918/, https://doi.org/10.1056/NEJMoa1817364.
5.
United States Environmental Protection Agency. 2016. Integrated Science Assessment for Oxides of Nitrogen–Health Criteria. Washington, DC: Office of Research and Development, National Center for Environmental Assessment. https://cfpub.epa.gov/ncea/isa/recordisplay.cfm?deid=310879 [accessed 5 July 2022].
6.
Khreis H, Kelly C, Tate J, Parslow R, Lucas K, Nieuwenhuijsen M. 2017. Exposure to traffic-related air pollution and risk of development of childhood asthma: a systematic review and meta-analysis. Environ Int 100:1–31. https://pubmed.ncbi.nlm.nih.gov/27881237/, https://doi.org/10.1016/j.envint.2016.11.012.
7.
Garcia E, Urman R, Berhane K, McConnell R, Gilliland F. 2019. Effects of policy-driven hypothetical air pollutant interventions on childhood asthma incidence in Southern California. Proc Natl Acad Sci USA 116(32):15883–15888. https://pubmed.ncbi.nlm.nih.gov/31332016/, https://doi.org/10.1073/pnas.1815678116.
8.
Thurston GD, Balmes JR, Garcia E, Gilliland FD, Rice MB, Schikowski T, et al. 2020. Outdoor air pollution and new-onset airway disease. An official American Thoracic Society workshop report. Ann Am Thorac Soc 17(4):387–398. https://pubmed.ncbi.nlm.nih.gov/32233861/, https://doi.org/10.1513/AnnalsATS.202001-046ST.
9.
HEI Panel on the Health Effects of Long-Term Exposure to Traffic-Related Air Pollution. 2022. Systematic Review and Meta-Analysis of Selected Health Effects of Long-Term Exposure to Traffic-Related Air Pollution. Boston, MA: Health Effects Institute. https://www.healtheffects.org/system/files/hei-special-report-23_1.pdf [accessed 6 July 2022].
10.
Sullivan TJ, Driscoll CT, Beier CM, Burtraw D, Fernandez IJ, Galloway JN, et al. 2018. Air pollution success stories in the United States: the value of long-term observations. Environ Sci Policy 84:69–73, https://doi.org/10.1016/j.envsci.2018.02.016.
11.
United States Environmental Protection Agency. 2012. Regulatory Impact Analysis for the Final Revisions to the National Ambient Air Quality Standards for Particulate Matter. Research Triangle Park, NC: Office of Air Quality Planning and Standards, Health and Environmental Impacts Division. https://www.epa.gov/sites/default/files/2020-07/documents/naaqs-pm_ria_final_2012-12.pdf [accessed 5 July 2022].
12.
Fann N, Coffman E, Timin B, Kelly JT. 2018. The estimated change in the level and distribution of PM2.5-attributable health impacts in the United States: 2005–2014. Environ Res 167:506–514. https://pubmed.ncbi.nlm.nih.gov/30142626/, https://doi.org/10.1016/j.envres.2018.08.018.
13.
Colmer J, Hardman I, Shimshack J, Voorheis J. 2020. Disparities in PM2.5 air pollution in the United States. Science 369(6503):575–578. https://pubmed.ncbi.nlm.nih.gov/32732425/, https://doi.org/10.1126/science.aaz9353.
14.
Liu J, Clark LP, Bechle MJ, Hajat A, Kim S-Y, Robinson AL, et al. 2021. Disparities in air pollution exposure in the United States by race/ethnicity and income, 1990–2010. Environ Health Perspect 129(12):127005. https://pubmed.ncbi.nlm.nih.gov/34908495/, https://doi.org/10.1289/EHP8584.
15.
Jbaily A, Zhou X, Liu J, Lee T-H, Kamareddine L, Verguet S, et al. 2022. Air pollution exposure disparities across US population and income groups. Nature 601(7892):228–233. https://pubmed.ncbi.nlm.nih.gov/35022594/, https://doi.org/10.1038/s41586-021-04190-y.
16.
Tessum CW, Paolella DA, Chambliss SE, Apte JS, Hill JD, Marshall JD. 2021. PM2.5 polluters disproportionately and systemically affect people of color in the United States. Sci Adv 7(18):eabf4491, https://doi.org/10.1126/sciadv.abf4491.
17.
Liu J, Marshall JD. 2023. Spatial decomposition of air pollution concentrations highlights historical causes for current exposure disparities in the United States. Environ Sci Technol Lett 10(3):280–286. https://pubmed.ncbi.nlm.nih.gov/36938149/, https://doi.org/10.1021/acs.estlett.2c00826.
18.
Wang Y, Liu P, Schwartz J, Castro E, Wang W, Chang H, et al. 2023. Disparities in ambient nitrogen dioxide pollution in the United States. Proc Natl Acad Sci USA 120(16):e2208450120. https://pubmed.ncbi.nlm.nih.gov/37036985/, https://doi.org/10.1073/pnas.2208450120.
19.
Cluck W. 23 April 2021. Environmental justice is a Central focus of the Biden administration. SSRN J, https://doi.org/10.2139/ssrn.3819845.
20.
Manson S, Schroeder J, Van Riper D, Kugler T, Ruggles S. 2020. IPUMS National Historical Geographic Information System: version 14.0. Minneapolis, MN: IPUMS. https://doi.org/10.18128/D050.V15.0 [accessed 7 June 2022].
21.
U.S. Census Bureau. 2019. 2019 TIGER/Line Shapefiles (Machine-Readable Data Files). https://www.census.gov/cgi-bin/geo/shapefiles/index.php [accessed 3 June 2021].
22.
van Donkelaar A, Hammer MS, Bindle L, Brauer M, Brook JR, Garay MJ, et al. 2021. Monthly global estimates of fine particulate matter and their uncertainty. Environ Sci Technol 55(22):15287–15300. https://pubmed.ncbi.nlm.nih.gov/34724610/, https://doi.org/10.1021/acs.est.1c05309.
23.
Anenberg SC, Mohegh A, Goldberg DL, Kerr GH, Brauer M, Burkart K, et al. 2022. Long-term trends in urban NO2 concentrations and associated paediatric asthma incidence: estimates from global datasets. Lancet Planet Health 6(1):e49–e58. https://pubmed.ncbi.nlm.nih.gov/34998460/, https://doi.org/10.1016/S2542-5196(21)00255-2.
24.
Jerrett M, Burnett RT, Pope CA, Ito K, Thurston G, Krewski D, et al. 2009. Long-term ozone exposure and mortality. N Engl J Med 360(11):1085–1095. https://pubmed.ncbi.nlm.nih.gov/19279340/, https://doi.org/10.1056/NEJMoa0803894.
25.
Chen K, Breitner S, Wolf K, Stafoggia M, Sera F, Vicedo-Cabrera AM, et al. 2021. Ambient carbon monoxide and daily mortality: a global time-series study in 337 cities. Lancet Planet Health 5(4):e191–e199. https://pubmed.ncbi.nlm.nih.gov/33838734/, https://doi.org/10.1016/S2542-5196(21)00026-7.
26.
Larkin A, Geddes JA, Martin RV, Xiao Q, Liu Y, Marshall JD, et al. 2017. Global land use regression model for nitrogen dioxide air pollution. Environ Sci Technol 51(12):6957–6964. https://pubmed.ncbi.nlm.nih.gov/28520422/, https://doi.org/10.1021/acs.est.7b01148.
27.
Bey I, Jacob DJ, Yantosca RM, Logan JA, Field BD, Fiore AM, et al. 2001. Global modeling of tropospheric chemistry with assimilated meteorology: model description and evaluation. J Geophys Res Atmospheres 106(D19):23073–23095, https://doi.org/10.1029/2001JD000807.
28.
van Donkelaar A, Martin RV, Li C, Burnett RT. 2019. Regional estimates of chemical composition of fine particulate matter using a combined geoscience-statistical method with information from satellites, models, and monitors. Environ Sci Technol 53(5):2595–2611. https://pubmed.ncbi.nlm.nih.gov/30698001/, https://doi.org/10.1021/acs.est.8b06392.
29.
Anenberg SC, Mohegh A, Goldberg DL, Kerr GH, Brauer M, Burkart K, et al. 2022. Long-term trends in urban NO2 concentrations and associated paediatric asthma incidence: estimates from global datasets. The Lancet Planetary Health 6(1):e49–e58, https://doi.org/10.1016/S2542-5196(21)00255-2.
30.
Van Donkelaar A, Hammer MS, Bindle L, Brauer M, Brook JR, Garay MJ, et al. 2021. Monthly global estimates of fine particulate matter and their uncertainty. Environ Sci Technol 55(22):15287–15300, https://doi.org/10.1021/acs.est.1c05309.
31.
Murray CJL, Aravkin AY, Zheng P, Abbafati C, Abbas KM, Abbasi-Kangevari M, et al. 2020. Global burden of 87 risk factors in 204 countries and territories, 1990–2019: a systematic analysis for the global burden of disease study 2019. The Lancet 396(10258):1223–1249, https://doi.org/10.1016/S0140-6736(20)30752-2.
32.
Global Burden of Disease Collaborative Network. 2021. Particulate Matter Risk Curves. Seattle, WA: Institute for Health Metrics and Evaluation (IHME). https://ghdx.healthdata.org/record/ihme-data/global-burden-disease-study-2019-gbd-2019-particulate-matter-risk-curves [accessed 23 June 2023].
33.
Anderson HR, Favarato G, Atkinson RW. 2013. Long-term exposure to air pollution and the incidence of asthma: meta-analysis of cohort studies. Air Qual Atmos Health 6(1):47–56, https://doi.org/10.1007/s11869-011-0144-5.
34.
Khreis H, Cirach M, Mueller N, de Hoogh K, Hoek G, Nieuwenhuijsen MJ, et al. 2019. Outdoor air pollution and the burden of childhood asthma across Europe. Eur Respir J 54(4):1802194. https://pubmed.ncbi.nlm.nih.gov/31391220/, https://doi.org/10.1183/13993003.02194-2018.
35.
Lett E, Adekunle D, McMurray P, Asabor EN, Irie W, Simon MA, et al. 2022. Health equity tourism: ravaging the justice landscape. J Med Syst 46(3):17. https://pubmed.ncbi.nlm.nih.gov/35150324/, https://doi.org/10.1007/s10916-022-01803-5.
36.
Vos T, Lim SS, Abbafati C, Abbas KM, Abbasi M, Abbasifard M, et al. 2020. Global burden of 369 diseases and injuries in 204 countries and territories, 1990–2019: a systematic analysis for the global burden of disease study 2019. The Lancet 396(10258):1204–1222, https://doi.org/10.1016/S0140-6736(20)30925-9.
37.
Kerr GH, Goldberg DL, Anenberg SC. 2021. COVID-19 pandemic reveals persistent disparities in nitrogen dioxide pollution. Proc Natl Acad Sci USA 118(30):e2022409118. https://pubmed.ncbi.nlm.nih.gov/34285070/, https://doi.org/10.1073/pnas.2022409118.
38.
Ostro BD. 2004. Outdoor Air Pollution: Assessing the Environmental Burden of Disease at National and Local Levels. Geneva, Switzerland: World Health Organization.
39.
Spielman SE, Folch DC. 2015. Reducing uncertainty in the American community survey through data-driven regionalization. PLoS One 10(2):e0115626. https://pubmed.ncbi.nlm.nih.gov/25723176/, https://doi.org/10.1371/journal.pone.0115626.
40.
Achakulwisut P, Brauer M, Hystad P, Anenberg SC. 2019. Global, national, and urban burdens of paediatric asthma incidence attributable to ambient NO2 pollution: estimates from global datasets. Lancet Planet Health 3(4):e166–e178. https://pubmed.ncbi.nlm.nih.gov/30981709/, https://doi.org/10.1016/S2542-5196(19)30046-4.
41.
Ostro B, Spadaro JV, Gumy S, Mudu P, Awe Y, Forastiere F, et al. 2018. Assessing the recent estimates of the global burden of disease for ambient air pollution: methodological changes and implications for low- and middle-income countries. Environ Res 166:713–725. https://pubmed.ncbi.nlm.nih.gov/29880237/, https://doi.org/10.1016/j.envres.2018.03.001.
42.
Office of Management and Budget. 2010. 2010 Standards for Delineating Metropolitan and Micropolitan Statistical Areas. https://www.govinfo.gov/content/pkg/FR-2010-06-28/pdf/2010-15605.pdf [accessed 6 July 2022].
43.
Martinez RAM, Andrabi N, Goodwin AN, Wilbur RE, Smith NR, Zivich PN. 2023. Conceptualization, operationalization, and utilization of race and ethnicity in major epidemiology journals, 1995–2018: a systematic review. Am J Epidemiol 192(3):483–496. https://pubmed.ncbi.nlm.nih.gov/35938872/, https://doi.org/10.1093/aje/kwac146.
44.
Richmond-Bryant J, Mikati I, Benson AF, Luben TJ, Sacks JD. 2020. Disparities in distribution of particulate matter emissions from US coal-fired power plants by race and poverty status after accounting for reductions in operations between 2015 and 2017. Am J Public Health 110(5):655–661. https://pubmed.ncbi.nlm.nih.gov/32191524/, https://doi.org/10.2105/AJPH.2019.305558.
45.
Hoffmann B, Boogaard H, de Nazelle A, Andersen ZJ, Abramson M, Brauer M, et al. 2021. WHO air quality guidelines 2021–aiming for healthier air for all: a joint statement by medical, public health, scientific societies and patient representative organisations. Int J Public Health 66:1604465. https://pubmed.ncbi.nlm.nih.gov/34630006/, https://doi.org/10.3389/ijph.2021.1604465.
46.
US Environmental Protection Agency. 2010. Guidelines for Preparing Economic Analyses. Washington, DC: US EPA National Center for Environmental Economics Office of Policy. https://www.epa.gov/sites/production/files/2017-08/documents/ee-0568-50.pdf [accessed 20 May 2022].
47.
Perry R, Braileanu G, Palmer T, Stevens P. 2019. The economic burden of pediatric asthma in the United States: literature review of current evidence. Pharmacoeconomics 37(2):155–167. https://pubmed.ncbi.nlm.nih.gov/30315512/, https://doi.org/10.1007/s40273-018-0726-2.
48.
Raich W, Fant C, Jackson M, Roman H. 2020. Memorandum Supporting Near-Source Health Benefits Analyses Using Fine-Scale Incidence Rates. Cambridge, MA: Industrial Economics, Incorporated (IEc).
49.
Turner MC, Jerrett M, Pope CA, Krewski D, Gapstur SM, Diver WR, et al. 2016. Long-term ozone exposure and mortality in a large prospective study. Am J Respir Crit Care Med 193(10):1134–1142. https://pubmed.ncbi.nlm.nih.gov/26680605/, https://doi.org/10.1164/rccm.201508-1633OC.
50.
Zhang Y, Gautam R, Pandey S, Omara M, Maasakkers JD, Sadavarte P, et al. 2020. Quantifying methane emissions from the largest oil-producing basin in the United States from space. Sci Adv 6(17):eaaz5120. https://pubmed.ncbi.nlm.nih.gov/32494644/, https://doi.org/10.1126/sciadv.aaz5120.
51.
Dix B, Francoeur C, Li M, Serrano-Calvo R, Levelt PF, Veefkind JP, et al. 2022. Quantifying NOx emissions from U.S. Oil and gas production regions using TROPOMI NO2. ACS Earth Space Chem 6(2):403–414, https://doi.org/10.1021/acsearthspacechem.1c00387.
52.
Willis M, Hystad P, Denham A, Hill E. 2020. Natural gas development, flaring practices and paediatric asthma hospitalizations in Texas. Int J Epidemiol 49(6):1883–1896, https://doi.org/10.1093/ije/dyaa115.
53.
Cortes-Bergoderi M, Goel K, Murad MH, Allison T, Somers VK, Erwin PJ, et al. 2013. Cardiovascular mortality in Hispanics compared to non-Hispanic whites: a systematic review and meta-analysis of the Hispanic paradox. Eur J Intern Med 24(8):791–799. https://pubmed.ncbi.nlm.nih.gov/24095273/, https://doi.org/10.1016/j.ejim.2013.09.003.
54.
Frampton MW, Greaves IA, ATS Environmental Health Policy Committee. 2009. NOx - NOx: who’s there? Am J Respir Crit Care Med 179(12):1077–1078. https://pubmed.ncbi.nlm.nih.gov/19498060/, https://doi.org/10.1164/rccm.200903-0485ED.
55.
Di Q, Wang Y, Zanobetti A, Wang Y, Koutrakis P, Choirat C, et al. 2017. Air pollution and mortality in the Medicare population. N Engl J Med 376(26):2513–2522. https://pubmed.ncbi.nlm.nih.gov/28657878/, https://doi.org/10.1056/NEJMoa1702747.
56.
Independent Particulate Matter Review Panel. 2020. The need for a tighter particulate-matter air-quality standard. N Engl J Med 383(7):680–683, https://doi.org/10.1056/NEJMsb2011009.
57.
Environmental Protection Agency. 2023. National Ambient Air Quality Standards (NAAQS) for PM. https://www.epa.gov/pm-pollution/national-ambient-air-quality-standards-naaqs-pm [accessed 15 June 2023].
58.
Dedoussi IC, Eastham SD, Monier E, Barrett SRH. 2020. Premature mortality related to United States cross-state air pollution. Nature 578(7794):261–265. https://pubmed.ncbi.nlm.nih.gov/32051602/, https://doi.org/10.1038/s41586-020-1983-8.
59.
Mikati I, Benson AF, Luben TJ, Sacks JD, Richmond-Bryant J. 2018. Disparities in distribution of particulate matter emission sources by race and poverty status. Am J Public Health 108(4):480–485. https://pubmed.ncbi.nlm.nih.gov/29470121/, https://doi.org/10.2105/AJPH.2017.304297.
60.
Van Horne YO, Alcala CS, Peltier RE, Quintana PJE, Seto E, Gonzales M, et al. 2022. An applied environmental justice framework for exposure science. J Expo Sci Environ Epidemiol 33(1):1–11, https://doi.org/10.1038/s41370-022-00422-z.
61.
Demetillo MAG, Harkins C, McDonald BC, Chodrow PS, Sun K, Pusede SE. 2021. Space‐based observational constraints on NO2 air pollution inequality from diesel traffic in major US cities. Geophys Res Lett 48(17):e2021GL094333, https://doi.org/10.1029/2021GL094333.
62.
Nardone A, Chiang J, Corburn J. 2020. Historic redlining and urban health today in U.S. Cities. Environ Justice 13(4):109–119, https://doi.org/10.1089/env.2020.0011.
63.
Lane HM, Morello-Frosch R, Marshall JD, Apte JS. 2022. Historical redlining is associated with present-day air pollution disparities in U.S. Cities. Environ Sci Technol Lett 9(4):345–350, https://doi.org/10.1021/acs.estlett.1c01012.
64.
Bullard RD. 1993. The threat of environmental racism. Nat Resour Environ 7(3):23–26.
65.
Nardone A, Casey JA, Morello-Frosch R, Mujahid M, Balmes JR, Thakur N. 2020. Associations between historical residential redlining and current age-adjusted rates of emergency department visits due to asthma across eight cities in California: an ecological study. Lancet Planet Health 4(1):e24–e31. https://pubmed.ncbi.nlm.nih.gov/31999951/, https://doi.org/10.1016/S2542-5196(19)30241-4.
66.
Mohl RA. 2012. The expressway teardown movement in American cities: rethinking postwar highway policy in the post-interstate era. J Plan Hist 11(1):89–103, https://doi.org/10.1177/1538513211426028.
67.
Yin H, Brauer M, Zhang JJ, Cai W, Navrud S, Burnett R, et al. 2021. Population ageing and deaths attributable to ambient PM2.5 pollution: a global analysis of economic cost. Lancet Planet Health 5(6):e356–e367. https://pubmed.ncbi.nlm.nih.gov/34119010/, https://doi.org/10.1016/S2542-5196(21)00131-5.
68.
Mohegh A, Goldberg D, Achakulwisut P, Anenberg SC. 2021. Sensitivity of estimated NO2-attributable pediatric asthma incidence to grid resolution and urbanicity. Environ Res Lett 16(1):014019, https://doi.org/10.1088/1748-9326/abce25.
69.
Chowdhury S, Haines A, Klingmüller K, Kumar V, Pozzer A, Venkataraman C, et al. 2021. Global and national assessment of the incidence of asthma in children and adolescents from major sources of ambient NO2. Environ Res Lett 16(3):035020, https://doi.org/10.1088/1748-9326/abe909.
70.
Castillo MD, Kinney PL, Southerland V, Arno CA, Crawford K, van Donkelaar A, et al. 2021. Estimating intra‐urban inequities in PM2.5‐attributable health impacts: a case study for Washington, DC. GeoHealth 5(11):e2021GH000431. https://pubmed.ncbi.nlm.nih.gov/34765851/, https://doi.org/10.1029/2021GH000431.
71.
Héroux M-E, Anderson HR, Atkinson R, Brunekreef B, Cohen A, Forastiere F, et al. 2015. Quantifying the health impacts of ambient air pollutants: recommendations of a WHO/Europe project. Int J Public Health 60(5):619–627. https://pubmed.ncbi.nlm.nih.gov/26024815/, https://doi.org/10.1007/s00038-015-0690-y.
72.
Plass D, Hilderink H, Lehtomäki H, Øverland S, Eikemo TA, Lai T, et al. 2022. Estimating risk factor attributable burden – challenges and potential solutions when using the comparative risk assessment methodology. Arch Public Health 80(1):148. https://pubmed.ncbi.nlm.nih.gov/35624479/, https://doi.org/10.1186/s13690-022-00900-8.
73.
Papadogeorgou G, Kioumourtzoglou MA, Braun D, Zanobetti A. 2019. Low levels of air pollution and health: effect estimates, methodological challenges, and future directions. Curr Environ Health Rep 6(3):105–115. https://pubmed.ncbi.nlm.nih.gov/31090042/, https://doi.org/10.1007/s40572-019-00235-7.
74.
Kheirbek I, Wheeler K, Walters S, Kass D, Matte T. 2013. PM2.5 and ozone health impacts and disparities in New York City: sensitivity to spatial and temporal resolution. Air Qual Atmos Health 6(2):473–486. https://pubmed.ncbi.nlm.nih.gov/23710262/, https://doi.org/10.1007/s11869-012-0185-4.
75.
Akinbami LJ, Rhodes JC, Lara M. 2005. Racial and ethnic differences in asthma diagnosis among children who wheeze. Pediatrics 115(5):1254–1260. https://pubmed.ncbi.nlm.nih.gov/15867032/, https://doi.org/10.1542/peds.2004-0897.
76.
Spiller E, Proville J, Roy A, Muller NZ. 2021. Mortality risk from PM2.5: a comparison of modeling approaches to identify disparities across racial/ethnic groups in policy outcomes. Environ Health Perspect 129(12):127004. https://pubmed.ncbi.nlm.nih.gov/34878311/, https://doi.org/10.1289/EHP9001.
77.
Alexeeff SE, Deosaransingh K, Liao NS, Van Den Eeden SK, Schwartz J, Sidney S. 2021. Particulate matter and cardiovascular risk in adults with chronic obstructive pulmonary disease. Am J Respir Crit Care Med 204(2):159–167. https://pubmed.ncbi.nlm.nih.gov/33662228/, https://doi.org/10.1164/rccm.202007-2901OC.
78.
Parker JD, Kravets N, Vaidyanathan A. 2018. Particulate matter air pollution exposure and heart disease mortality risks by race and ethnicity in the United States: 1997 to 2009 national health interview survey with mortality follow-up through 2011. Circulation 137(16):1688–1697. https://pubmed.ncbi.nlm.nih.gov/29237717/, https://doi.org/10.1161/CIRCULATIONAHA.117.029376.
79.
Hicken MT, Payne-Sturges D, McCoy E. 2023. Evaluating race in air pollution and health research: race, PM2.5 air pollution exposure, and mortality as a case study. Curr Envir Health Rpt 10(1):1–11, https://doi.org/10.1007/s40572-023-00390-y.
80.
U.S. Environmental Protection Agency Office of Air and Radiation. 2011. The Benefits and Costs of the Clean Air Act from 1990 to 2020. https://www.epa.gov/sites/default/files/2015-07/documents/fullreport_rev_a.pdf [accessed 28 November 2022].
81.
The George Washington University High Performance Computing, Cluster. 2020. Building a Shared Resource HPC Center Across University Schools and Institutes: a Case Study. https://arxiv.org/abs/2003.13629/ [accessed 15 June 2022].

Information & Authors

Information

Published In

Environmental Health Perspectives
Volume 132Issue 3March 2024
PubMed: 38445892

History

Received: 22 July 2022
Revision received: 1 December 2023
Accepted: 16 January 2024
Published online: 6 March 2024
Corrected: 5 April 2024

Notes

Conclusions and opinions are those of the individual authors and do not necessarily reflect the policies or views of EHP Publishing or the National Institute of Environmental Health Sciences.

Authors

Affiliations

Department of Environmental and Occupational Health, The George Washington University, Washington, District of Columbia, USA
Aaron van Donkelaar
Department of Energy, Environmental & Chemical Engineering, Washington University in St. Louis, St. Louis, Missouri, USA
Randall V. Martin
Department of Energy, Environmental & Chemical Engineering, Washington University in St. Louis, St. Louis, Missouri, USA
Michael Brauer
Department of Health Metrics Sciences, Institute of Health Metrics and Evaluation, University of Washington, Seattle, Washington, USA
School of Population and Public Health, University of British Columbia, Vancouver, British Columbia, Canada
Katrin Bukart
Department of Health Metrics Sciences, Institute of Health Metrics and Evaluation, University of Washington, Seattle, Washington, USA
Sarah Wozniak
Department of Health Metrics Sciences, Institute of Health Metrics and Evaluation, University of Washington, Seattle, Washington, USA
Daniel L. Goldberg
Department of Environmental and Occupational Health, The George Washington University, Washington, District of Columbia, USA
Susan C. Anenberg
Department of Environmental and Occupational Health, The George Washington University, Washington, District of Columbia, USA

Notes

Address correspondence to Gaige Hunter Kerr, Department of Environmental and Occupational Health, George Washington University, 950 New Hampshire Ave., NW, Washington, DC 20052 USA. Email: [email protected]

Metrics & Citations

Metrics

About Article Metrics


Citations

Download citation

If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click DOWNLOAD.

Cited by

  • Neonatal intensive care admissions and exposure to satellite-derived air pollutants in the United States, 2018, Scientific Reports, 10.1038/s41598-024-84755-9, 15, 1, (2025).
  • TROPOMI Satellite Data Reshape NO 2 Air Pollution Land-Use Regression Modeling Capabilities in the United States , ACS ES&T Air, 10.1021/acsestair.4c00153, (2025).
  • Tracking Airborne Lead Exposures That Disrupt Children’s Fronto-executive Functions and Inhibitory Systems, American Journal of Public Health, 10.2105/AJPH.2024.307866, 114, 12, (1313-1316), (2024).
  • Invited Perspective: Increasing Disparities in Air Pollution Health Burdens in the United States—A Concerning Trend for and a Path Forward, Environmental Health Perspectives, 10.1289/EHP14233, 132, 3, (2024).
  • The last decade of air pollution epidemiology and the challenges of quantitative risk assessment, Environmental Health, 10.1186/s12940-024-01136-5, 23, 1, (2024).
  • Toward cleaner air and better health: Current state, challenges, and priorities, Science, 10.1126/science.adp7832, 385, 6707, (386-390), (2024).
  • PM 2.5 exposure disparities persist despite strict vehicle emissions controls in California , Science Advances, 10.1126/sciadv.adn8544, 10, 37, (2024).
  • Disproportionate Clean Air Act violations occur in communities of color throughout the United States, Environmental Research Letters, 10.1088/1748-9326/ad416d, 19, 5, (054052), (2024).
  • The nexus of transportation, the built environment, air pollution and health, Cities & Health, 10.1080/23748834.2024.2376389, (1-20), (2024).
  • Methods for Quantifying Source‐Specific Air Pollution Exposure to Serve Epidemiology, Risk Assessment, and Environmental Justice, GeoHealth, 10.1029/2024GH001188, 8, 11, (2024).
  • See more

View Options

View options

PDF

View PDF

Restore your content access

Enter your email address to restore your content access:

Note: This functionality works only for purchases done as a guest. If you already have an account, log in to access the content to which you are entitled.

Figures

Tables

Share

Share

Copy the content Link

Share on social media