Abstract
Background.
Environmental justice mandates that no person suffers disproportionately from environmental exposures. The Environmental Justice Index (EJI) provides an estimate of the environmental burden for each census tract but has not yet been utilized in asthma populations.
Objective.
We hypothesized that children from census tracts with high environmental injustice determined by the EJI would have a greater burden of asthma exacerbations, poorer asthma control, and poorer lung function over 12 months.
Methods.
Children 6–18 years with asthma (N=575) from metropolitan Atlanta, Georgia, completed a baseline research visit. Participant addresses were geocoded to obtain the EJI Social-Environmental ranking for each participant’s census tract, which was divided into tertiles. Medical records were reviewed for 12 months for asthma exacerbations. A subset of participants completed a second research visit involving spirometry and questionnaires.
Results.
Census tracts with the greatest environmental injustice had more racial/ethnic minorities, lower socioeconomic status, more hazardous exposures (particularly to airborne pollutants), and greater proximity to railroads and heavily trafficked roadways. Children with asthma residing in high injustice census tracts had a longer duration of asthma, greater historical asthma-related healthcare utilization, poorer asthma symptom control and quality of life, and more impaired lung function. By 12 months, children from high injustice census tracts also had more asthma exacerbations with a shorter time to exacerbation and persistently more symptoms, poorer asthma control, and reduced lung function.
Conclusion.
Disparities in environmental justice are present in metropolitan Atlanta that may contribute to asthma outcomes in children. These findings require additional study and action to improve health equity.
Keywords: Air pollution, Asthma control, Asthma exacerbation, Asthma in children, Disparity, Environment, Environmental Justice Index, Health equity, Social determinants of health
Introduction
Environmental justice is defined by the United States Environmental Protection Agency as the “fair treatment and meaningful involvement of all people with respect to environmental laws, regulations and policies.”(1) Fair treatment demands that no group of people suffer disproportionately from the negative effects of environmental exposures.(1) However, associations between race and air pollutants have been reported since the 1970s.(2) In his pivotal study in 1983, Dr. Robert Bullard found that Black residents in Houston, Texas were more likely to live near waste disposal sites than non-Black residents.(3) In a recent interview, Bullard stated that “Black people made up on 25 percent of Houston’s population [in the 1970s], yet 82 percent of the garbage in the city was dumped on them.”(2) Forty years later, associations between pollutant exposure and historically marginalized and socially disadvantaged populations persist.(4–6)
Once exposed, the effects of pollutants are also not equal across the population. Children are more susceptible, particularly to airborne pollutants.(7) Children have higher respiratory rates and minute ventilation, which increase pollutant inhalation, and more narrow airways, which amplify particulate trapping, absorption and epithelial inflammation and oxidative stress.(8, 9) Children also have developing lungs. Exposure to airborne pollutants has been associated with lower lung function measures during infancy(10) and impairment of lung growth from childhood to adulthood.(11) These negative effects are increased even further in socially vulnerable populations of children and are associated with significantly poorer health,(5) including asthma.(12) Indeed, Black and Hispanic children with asthma are more likely to live in lower opportunity neighborhoods due to systemic structural racism and historic red lining practices,(13) are exposed to significantly more pollutants,(14, 15) and have more severe asthma exacerbations attributable to air pollution.(7, 12, 16–21) These same children are also projected to have a 1.3-fold increased risk of living in a neighborhood with the highest projected increases in asthma due to increased particulate matter concentrations.(22)
To promote environmental health equity, the Department of Health and Human Services’ Office of Environmental Justice and the Centers for Disease Control and Prevention’s Agency for Toxic Substances and Disease Registry developed an Environmental Justice Index (EJI) in 2022. This tool is based on a battery of composite indicators and provides a single score to estimate the cumulative effects of environmental burden for each census tract, which is the smallest subdivision of land for which data are consistently available.(23, 24) To better understand the local factors associated with asthma outcomes and to improve awareness of environmental injustice in children from metropolitan Atlanta, Georgia, we compared these EJI rankings in a highly characterized sample of children with asthma who consented to participate in clinical research. We hypothesized that children who resided in census tracts with high environmental injustice would have differing clinical features of asthma at enrollment and a greater burden of asthma exacerbations, poorer asthma control, and poorer lung function over 12 months of follow-up.
Methods
Children aged 6 through 18 years with physician-diagnosed asthma at an academic specialty center for asthma at Children’s Healthcare of Atlanta in Atlanta, Georgia were eligible for the study if they were receiving current treatment for asthma and had historical evidence of ≥ 12% reversibility in their forced expiratory volume in one second (FEV1) relative to baseline after bronchodilator administration. Children’s Healthcare of Atlanta is the largest pediatric healthcare system in Georgia and consists of 3 pediatric hospitals, an outpatient Center for Advanced Pediatrics, and 18 neighborhood locations. Each year, Children’s Healthcare of Atlanta manages more than 414,000 patients from all 159 counties in Georgia. The population served is 39% White, 36% Black, and 17% Hispanic. However, many of the patients presenting for asthma care are from two surrounding counties which are 34% White, 61% Black, and 3% Hispanic. Legal guardians of children meeting inclusion criteria were identified through a review of electronic medical records and were contacted by email, phone, or in person during clinical encounters. Exclusion criteria included residence outside of Georgia, premature birth before 35 weeks of gestation and other chronic airway disorders that could mimic asthma such as pulmonary aspiration or vocal cord dysfunction. Permission to proceed with this study was granted by the Emory University and Children’s Healthcare of Atlanta Institutional Review Boards. Informed written consent was obtained from legal guardians and participants 18 years of age. Verbal assent was obtained from children 6–10 years and written assent was obtained from children 11 through 17 years.
Study design and outcome measures.
Eligible participants were identified from routine clinical encounters at Children’s Healthcare of Atlanta and were invited to participate in a research study. After informed consent for the study was obtained, participants who were clinically stable completed a research visit involving characterization procedures. After this visit, the electronic medical records of each participant were reviewed for up to 12 months for occurrence of an asthma exacerbation treated with systemic corticosteroids, an asthma exacerbation prompting an Emergency Department visit, and hospitalization for asthma. Spirometry and asthma control and quality of life determination was also performed in a subset of participants who consented to a research-only, in-person follow-up visit at 12 months.
Residential geocoding and Environmental Justice Index quantification.
Residential geocoding was performed as described previously(25, 26) by mapping participant residential addresses to census tracts using the R package tidygeocoder (R version 4.0.2).(27) U.S. Census 2020 Geographic Identifiers (GEOIDs) were mapped to the 2010 GEOIDs for each census tract in Georgia using the R package tigris.(27, 28) The Center for Disease Control and Prevention’s EJI Social-Environmental Ranking,(23) which is calculated by combining rankings from only the Environmental Burden Module and the Social Vulnerability Module (excluding the Health Vulnerability Module), was determined for each participant census tract. The EJI Social-Environmental Ranking model ranks each census tract on 31 environmental and social factors and groups them into two overarching modules (Social Vulnerability Module and Environmental Burden module). The Social Vulnerability module contains four domains with a total of 14 individual indicators and the Environmental Burden module contains five domains with a total of 17 individual indicators. Percentile ranks for all individual indicators within each module are summed, producing a module score, which is then percentile ranked. The percentile ranked sum of the Social Vulnerability indicators (range 0–1) is then added to the percentile ranked sum of the Environmental Burden indicators (range 0–1) obtain an overall EJI Social-Vulnerability score, which is then percentile ranked. The EJI relies on historical data with varying time scales and is intended as a high-level mapping and screening tool that characterizes cumulative impacts and patterns of environmental injustice across communities. The Social-Environmental Burden ranking of the EJI is considered more suitable than the full EJI for research with health-related outcomes of interest.(24)
Clinical characterization procedures.
At the initial research visit, participants completed medical history and demographic questionnaires and the 6-item Asthma Control Questionnaire(29, 30) (ACQ) which questions about the occurrence of specific symptoms of asthma over the past week. The ACQ-6 was completed with technical assistance in children 5–10 years(31) and directly by participants aged 11–18 years. Participants also completed the Pediatric Asthma Quality of Life Questionnaire (PAQLQ).(32) Spirometry (KoKo® PDS, Ferraris, Louisville, Colorado) was performed according to technical standards.(33) The best of three forced vital capacity (FVC) maneuvers was interpreted for FVC, FEV1, and FEV11/FVC as percentages of predicted values according to Global Lung Function Initiative prediction equations.(34) Airflow obstruction was defined by FEV1 or FEV1/FVC values below the lower limit of normal.(34) Exhaled nitric oxide concentrations were measured according to technical standards(35) with a commercial device (NIOX®, Circrassia Inc., Morrisville, NC). Aeroallergen sensitization was assessed by specific immunoglobulin E (IgE) testing by the ImmunoCap method (Children’s Healthcare of Atlanta, Atlanta, GA) or by skin prick testing with eight extracts: tree mix, grass mix, weed mix, mold mix, dog dander, cat dander, Blatella germanica, and dust mite mix (Dermatophagoides farinae and Dermatophagoides pteronyssinus) (Greer® Laboratories, Lenoir, North Carolina). Venipuncture was also performed for quantification of blood eosinophils and serum IgE, which were measured in a hospital laboratory (Children’s Healthcare of Atlanta, Atlanta, Georgia).
Statistical analyses.
Statistical analyses were performed with IBM SPSS® Statistics (Version 29, IBM, Armonk, NY). Given the lack of scientific evidence supporting a specific weighting scheme for the modules and creation of the raw score for the EJI Social-Environmental Ranking score, this score was percentile ranked for each census tract consistent with technical recommendations,(24) with higher ranks corresponding to more potential distributive and procedural environmental injustice. EJI Social-Environmental percentile rankings were then split into “low,” “moderate” and “high” tertiles for comparison. Differences between the tertiles were compared with Chi-square tests or analysis of variance with Tukey’s Least Significant Difference post-hoc tests. Asthma outcomes of interest included asthma control, asthma-related quality of life, lung function, exacerbation occurrence by 12 months, and time to future exacerbation over 12 months. Exacerbations were defined according to a Working Group Report(36) as an acute worsening of respiratory symptoms necessitating treatment with systemic corticosteroids. Time to future exacerbation was analyzed with Cox regression with age in years, sex, and tobacco smoke exposure included as covariates in the model. A p-value <0.05 was used as the threshold for statistical significance for all analyses.
Results
Five hundred seventy-five participants with GEOIDs were included in the analysis and were distributed across the metropolitan Atlanta area (Figure 1A, B). Raw scores for the EJI Social-Environmental Ranking and its Social Vulnerability and Environmental Burden modules were normally distributed in the study population (Figure 2). Percentile rankings for the EJI Social-Environmental Ranking were then divided into tertiles for further analysis using cut-points of 0.5288 (corresponding to the 33.3 percentile) and 0.8007 (corresponding to the 66.7 percentile). Although the EJI Social-Environmental Ranking is not intended as a representation of risk or exposure for a given area,(24) tertiles were labeled as “low”, “moderate”, and “high” injustice for comparative and visualization purposes.
Figure 1.
Flowchart of participant enrollment and follow-up (panel A) and distribution of participants (shown as circles), by census tract (panel B). The colors of the census tracts represent the EJI percentile ranked score. Darker EJI census tract scores reflect less environmental injustice. EMR = electronic medical record; GEOID = Geographic Identifier.
Figure 2.
Histograms of the raw (top) and percentile ranked (bottom) Environmental Justice Index (EJI) Social-Environmental Ranking, (panel A) the Social Vulnerability Module (panel B), and the Environmental Burden Module (panel C). The black lines reflect the distribution curve.
Social vulnerability and environmental burden of census tracts.
The EJI Social-Environmental Ranking consists of two modules. The Social Vulnerability Module’s domains and indicators for the tertile groups are shown in Table 1. Census tracts with high injustice, compared to tracts with moderate or low injustice, had more racial/ethnic minorities, greater poverty, lesser education, more unemployment, more persons who rent their homes, lesser household income, and less access to health insurance and Internet services. Households in these census tracts were also slightly younger with more reported disability and less English fluency (Table 1). The Environmental Burden Module’s domains and indicators for the tertile groups are shown in Table 2. Air pollution exposure was high in each tertile, but there was significantly more diesel exposure in high versus moderate or low injustice census tracts. Census tracts with high injustice, compared to tracts with moderate or low injustice, also had more homes within a 1-mile buffer of an Environmental Protection Agency Toxic Release Inventory site, less green space and less walkability, more homes built before 1980 and more homes within a 1-mile buffer of a railroad, major roadway or highway (Table 2).
Table 1.
Social Vulnerability Module domains and domain indicators for census tracts separated into tertiles based on the Environmental Justice Index Social-Environmental Ranking. Data represent the mean ± standard deviation or the median (25th, 75th percentile).
| Domains and indicators | Low Injustice N=190 | Moderate Injustice N=190 | High Injustice N=195 |
|---|---|---|---|
| Racial/ethnic minority status (domain percentile ranking) | 0.62 ± 0.23 | 0.83 ± 0.17 * | 0.88 ± 0.11 * # |
| Minorities (%) | 48.5 ± 27.2 | 76.6 ± 23.6* | 84.3 ± 17.1*# |
| Socioeconomic status (domain percentile ranking) | 0.29 ± 0.18 | 0.64 ± 0.20 * | 0.86 ± 0.12 * # |
| Persons <200% poverty (%) | 20.1 ± 10.0 | 35.7 ± 12.6* | 51.6 ± 13.3*# |
| Persons age 25+ years with no high school diploma (%) | 6.0 ± 3.4 | 11.1 ± 5.5* | 18.1 ± 8.0*# |
| Persons unemployed (%) | 4.7 ± 2.4 | 7.2 ± 3.6* | 9.3 ± 4.4*# |
| Persons who rent (%) | 8.4 ± 7.2 | 16.2 ± 9.4* | 20.4 ± 8.0*# |
| Households <$75,000/year (%) | 21.3 ± 7.3 | 34.2 ± 9.4* | 41.9 ± 9.5*# |
| Persons uninsured (%) | 8.7 ± 3.9 | 13.6 ± 5.9* | 18.7 ± 6.0*# |
| Persons without internet (%) | 8.8 ± 5.7 | 17.1 ± 7.5* | 26.7 ± 11.8*# |
| Household characteristics (domain percentile ranking) | 0.30 ± 0.20 | 0.36 ± 0.22 * | 0.49 ± 0.22 * # |
| Persons aged 65+ years (%) | 13.5 ± 4.8 | 12.4 ± 5.7 | 11.3 ± 4.9*# |
| Persons 0–17 years (%) | 24.2 ± 5.5 | 22.6 ± 6.4* | 25.5 ± 6.2*# |
| Population with disability (%) | 9.5 ± 3.3 | 12.5 ± 5.4* | 13.2 ± 4.6* |
| Persons 5+ years who speak English “less than well” | 0.9 (0.2, 1.8) | 1.0 (0.2, 2.3)* | 1.6 (0.6, 5.1*# |
| Housing type (domain percentile ranking) | 0.29 ± 0.25 | 0.39 ± 0.30 * | 0.47 ± 0.27 * # |
| Mobile homes (%) | 0 (0, 1.1) | 0 (0, 1.8)* | 0.6 (0, 2.8)* |
| Persons in group quarters (%) | 0 (0, 0.2) | 0.2 (0, 0.5)* | 0.3 (0, 1.0)* |
p<0.05 vs. Low
p<0.05 vs. Moderate
Table 2.
Environmental Burden Module domains and domain indicators for census tracts separated into tertiles based on the Environmental Justice Index Social-Environmental Ranking. Data represent the mean ± standard deviation or the median (25th, 75th percentile).
| Domains and indicators | Low Injustice N=190 | Moderate Injustice N=190 | High Injustice N=195 |
|---|---|---|---|
| Air pollution (domain percentile ranking) | 0.86 ± 0.12 | 0.91 ± 0.11 * | 0.92 ± 0.11 * |
| Ozone (days/year above regulatory standard) | 3.3 ± 1.4 | 3.7 ± 1.4* | 3.6 ± 1.3* |
| PM2.5 (days/year above regulatory standard) | 10.5 ± 0.3 | 10.7 ± 0.3* | 10.7 ± 0.3* |
| Diesel (ambient concentration in PM/m3) | 0.45 ± 0.16 | 0.56 ± 0.18*# | 0.61 ± 0.17*# |
| Cancer probability over lifetime (assuming continuous exposure) | 44.8 ± 5.4 | 47.9 ± 9.5* | 49.9 ± 15.1* |
| Hazardous/toxic sites (domain percentile ranking) | 0.18 ± 0.24 | 0.31 ± 0.26 * | 0.61 ± 0.27 * # |
| Area within 1-mile buffer of EPA National Priority List site (%) | 0 | 0 | 0 |
| Area within 1-mile buffer of EPA Toxic Release Inventory site (%) | 0 (0, 6.2) | 10.5 (0, 30.6)* | 38.2 (6.3, 70.3)*# |
| Area within 1-mile buffer of EPA Treatment, Storage and Disposal site (%) | 0 | 0 | 0 |
| Area within 1-mile buffer of EPA risk management plan site (%) | 0 | 0 | 0 |
| Area within 1-mile buffer of coal mines (%) | 0 | 0 | 0 |
| Area within 1-mile buffer of lead mines (%) | 0 | 0 | 0 |
| Built environment (domain percentile ranking) | 0.44 ± 0.23 | 0.33 ± 0.23 * | 0.41 ± 0.24 # |
| Lack of green space (area outside a 1-mile buffer of a park or recreational area (%)) | 43.5 ± 34.8 | 64.5 ± 36.3* | 72.8 ± 33.4*# |
| Houses built before 1980 (%) | 24.0 ± 18.9 | 40.7 ± 20.9* | 55.7 ± 23.9*# |
| Lack of walkability (index ranking) | 7.3 ± 2.9 | 9.6 ± 3.1* | 10.0 ± 2.8* |
| Transportation infrastructure (domain percentile ranking) | 0.23 ± 0.27 | 0.46 ± 0.23 * | 0.63 ± 0.24 * # |
| Area within 1-mile buffer of railroad (%) | 0 (0, 21.3) | 26.2 (0, 64.8)* | 52.6 (15.0, 92.1)*# |
| Area within 1-mile buffer of highway (%) | 10.8 (0, 50.6) | 68.9 (38.6, 95.1)* | 84.5 (44.7, 100)*# |
| Area within 1-mile buffer of airport (%) | 0 | 0 | 0 |
| Water pollution (domain percentile ranking) | 0.43 ± 0.16 | 0.51 ± 0.17 * | 0.51 ± 0.17 * |
| Intersects an impaired/impacted watershed at the HUC12 level (%) | 39.8 ± 23.4 | 51.4 ± 25.3* | 51.5 ± 25.9* |
EPA = Environmental Protection Agency; HUC = hydrologic unit code; PM = particulate matter
p<0.05 vs. Low
p<0.05 vs. Moderate
Demographic and clinical features in participants.
Asthma features of the individual participants from low, moderate and high injustice census tracts are shown in Table 3. Although there were no differences in age, self-reported sex or ethnicity, participants from moderate and high injustice census tracts, compared to those from low injustice census tracts, were more likely to report a race other than White, had a longer duration of asthma, and had significantly more lifetime and current healthcare utilization for asthma despite no differences in prescribed asthma medications. Children from high versus low injustice census tracts also had more tobacco smoke exposure. Nearly all participants had some evidence of Type 2 inflammation, with 84% of participants overall sensitized to at least one aeroallergen (low, 77.7%; moderate, 88.4%; high, 86.6%, p=0.054) and 74% of participants overall with multiple aeroallergen sensitization (low, 66.1%; moderate, 76.9%; high, 77.2%, p=0.093). However, children from moderate and high injustice census tracts, compared to children from low injustice census tracts, were more likely to have sensitization to dust mite (low, 56.3%; moderate, 71.9%; high, 71.7%, p=0.015) and cockroach (low, 21.4%; moderate, 40.5%; high, 31.7%, p=0.007), with no differences in sensitization to molds, pollens, or pet dander. Children from moderate and high injustice census tracts also had higher exhaled nitric oxide and serum IgE concentrations (Table 3).
Table 3.
Features of the participants at enrollment. Data represent the mean ± standard deviation, median (25th, 75th percentile) or the number of participants (%).
| Low Injustice N=190 | Moderate Injustice N=190 | High Injustice N=195 | |
|---|---|---|---|
| Age (years) | 11.3 ± 6.7 | 11.2 ± 3.6 | 11.5 ± 3.7 |
| Asthma duration (years) | 7.1 ± 4.0 | 8.5 ± 4.0* | 9.2 ± 4.8* |
| Males | 98 (51.6) | 117 (61.6) | 116 (59.5) |
| Hispanic ethnicity | 5 (2.6) | 5 (2.6) | 5 (2.6) |
| Self-reported race | |||
| White | 64 (33.7) | 24 (12.6)* | 19 (9.7)* |
| Black | 103 (54.2) | 164 (86.3)* | 167 (85.6)* |
| Multiple | 15 (7.9) | 1 (0.5)* | 9 (4.6)# |
| Other | 9 (4.2) | 1 (0.5)* | 0* |
| Obesity (body mass index >95th percentile) | 44 (23.2) | 46 (24.2) | 48 (24.6) |
| Indoor exposures | |||
| Cat | 24 (12.6) | 15 (7.9)* | 14 (7.1)* |
| Dog | 68 (35.8) | 40 (21.1)* | 42 (21.5)* |
| Tobacco smoke | 28 (14.7) | 36 (18.9) | 49 (25.1)* |
| Asthma medications (prescribed) | |||
| Inhaled corticosteroid | 131 (68.9) | 135 (71.1) | 149 (76.4) |
| Long-acting beta agonist | 73 (38.4) | 83 (43.7) | 87 (44.6) |
| Leukotriene receptor antagonist | 97 (51.1) | 97 (51.1) | 96 (49.2) |
| Biologic | 6 (3.2) | 1 (0.5) | 7 (3.6) |
| Asthma healthcare utilization | |||
| Hospitalization (ever) | 82 (43.2) | 113 (59.5)* | 121 (62.1)* |
| Intensive care unit admission (ever) | 65 (34.2) | 79 (42.6) | 87 (44.6)* |
| Emergency Department (past year) | 82 (43.2) | 105 (55.3)* | 122 (62.6)* |
| Hospitalization (past year) | 60 (31.6) | 81 (42.6)* | 81 (41.5)* |
| Intubation for asthma (ever) | 20 (10.5) | 26 (13.7) | 28 (14.4) |
| Other allergic features | |||
| Eczema (ever) | 93 (48.9) | 114 (60.0) | 100 (51.2) |
| Exhaled nitric oxide (ppb) | 19 (10, 43) | 23 (11,44)* | 27 (11, 52)* |
| Blood eosinophils (cells/microliter) | 278 (141, 423) | 280 (126, 480) | 241 (139, 418) |
| Positive aeroallergens (of 8) | 3.3 ± 2.9 | 4.2 ± 3.2* | 3.9 ± 3.1 |
| Serum IgE (kU/L) | 277 (66, 656) | 485 (171, 1169)* | 307 (131, 797)* |
p<0.05 vs. Low
p<0.05 vs. Moderate
Asthma symptom control, quality of life and lung function.
Asthma symptom control and quality of life at enrollment are shown in Figure 3. Asthma symptom control reflected by total scores on the ACQ-6 instrument was significantly poorer in children from moderate and high versus low injustice census tracts (Figure 3). However, children from high injustice census tracts, compared to children from moderate and low injustice census tracts, had significantly higher (i.e., worse) ACQ-6 scores for the individual questions pertaining to nocturnal symptoms (High versus moderate and low combined, p=0.002), wheezing (p=0.035), and short-acting bronchodilator use (p=0.006), with no differences in morning symptoms (p=0.096), activity limitation (p=0.482) and dyspnea (p=0.216). Asthma quality of life was poorer in children from high versus low injustice census tracts (Figure 3B) and these scores correlated with ACQ-6 scores the combined sample (r = −0.709, p<0.001). Differences in quality of life were driven by the symptom and emotional domains of the AQLQ instrument (symptom domain: 6.06 ± 1.12 vs. 5.59 ± 1.28 vs. 5.52 ± 1.34 for low vs. moderate vs. high, p=0.009; emotional domain: 6.29 ± 1.11 vs. 5.73 ± 1.41 vs. 5.77 ± 1.34 for low vs. moderate vs. high, p=0.007) with no differences in the activity domain (5.93 ± 1.21 vs. 5.81 ± 1.10 vs. 5.58 ± 1.27 for low vs. moderate vs. high, p=0.131).
Figure 3.
Baseline Asthma Control Questionnaire (ACQ)-6 scores (panel A), Asthma Quality of Life Questionnaire (AQLQ) scores (panel B), FEV1 values (panel C) and FEV1/FVC values (panel D) in children from low (magenta), moderate (yellow), and high (blue) environmental injustice census tracts. Boxplot whiskers represent the 5th-95th percentile and individual participants are shown as dots. *p<0.05
Lung function values at enrollment are also shown in Figure 3. Children from census tracts with moderate and high environmental injustice had lower FEV1 than children from tracts with low injustice (Figure 3C). Differences in FEV1/FVC (Figure 3D) were also present in children from high versus low injustice census tracts. More children from high injustice census tracts also had FEV1 values below the lower limit of normal (FEV1: 14.1% vs. 21.4% vs. 28.4%, for low vs. moderate vs. high, p=0.010). The proportion of children with FEV1/FVC values below the lower limit of normal in each census tract grouping did not reach the threshold of statistical significance (32.7% vs. 40.7% vs. 43.9%, p=0.117).
Asthma outcomes at 12 months.
By 12 months, in the baseline sample of 575 participants, 155 children (27.0%) had an asthma exacerbation treated with systemic corticosteroids that was identified from a review of electronic medical records. Exacerbation occurrence was significantly greater in children from high injustice census tracts compared to low injustice census tracts (Figure 4A). Similar differences between children from high and low injustice census tracts were observed for Emergency Department visits (Figure 4A). Hospitalizations were also numerically greater in children from high injustice census tracts but this comparison did not meet the threshold of statistical significance (p=0.061) (Figure 4A). The time to the subsequent exacerbation was also significantly shorter in children from high versus injustice census tracts as shown in Figure 4B. The time to exacerbation did not meet the threshold of significance for children from high versus moderate injustice tracts (hazard ratio 1.39, 95% confidence interval, 0.94–2.04, p=0.096).
Figure 4.
Twelve-month exacerbation, emergency department (ED), and hospitalization occurrence (panel A), and time to subsequent exacerbation (panel B) in children from low (magenta), moderate (yellow), and high (blue) environmental injustice census tracts. HR= hazard ratio, CI = confidence interval.
At 12 months, 203 of the 575 included participants (35.3%) consented to an in-person, follow-up research visit. The percentage of participants participating in the follow up was not different between the census tract groups (low injustice, n=68 (35.8%); moderate injustice, n=65 (34.2%), high injustice, n=70 (35.9%), p=0.928). Children who participated in the follow-up visit, compared to children who did not participate, were similar in age (p=0.647) and gender (p=0.207) with no differences in inhaled corticosteroid prescription (p=0.676) but were less likely to be Hispanic (p=0.019) and more likely to identify a race other than White (p=0.028). At 12 months, children from high environmental injustice census tracts continued to have poorer asthma symptom control (Figure 5A), poorer quality of life (Figure 5B), lower FEV1 percent predicted values (Figure 5C), and lower FEV1/FVC percent predicted values (Figure 5D) than children residing in census tracts with lower environmental injustice.
Figure 5.
Asthma Control Questionnaire (ACQ)-6 scores (panel A), Asthma Quality of Life Questionnaire (AQLQ) scores (panel B), FEV1 values (panel C), and FEV1/FVC values (panel D) in children from low (magenta), moderate (yellow), and high (blue) environmental injustice census tracts. Boxplot whiskers represent the 5th-95th percentile and individual participants are shown as dots. *p<0.05
Since the EJI Social-Environmental ranking combines the scores from the Social Vulnerability Module and the Environmental Burden Module, we then examined how these two separate module scores were associated with asthma outcomes at 12 months. Exacerbations treated with systemic corticosteroids and exacerbations prompting Emergency Department visits were associated with social vulnerability but not with environmental burden (Table E1). Social vulnerability was also associated with asthma control, quality of life, FEV1 and FEV1/FVC at 12 months (Table E2). However, asthma control and FEV1 were also associated with environmental burden independent of social vulnerability (Table E2).
Discussion
In this analysis, we utilized the EJI Social-Environmental raking to better understand the census-tract level variables that contribute to asthma outcomes in a highly characterized sample of children with asthma residing in metropolitan Atlanta, Georgia. This analysis demonstrated that census tracts with highest environmental injustice had more racial/ethnic minority residents, lower socioeconomic status, more exposure to hazardous chemicals and pollutants (particularly airborne pollutants), and greater proximity to railroads and heavily trafficked roadways. Children with asthma residing in these high injustice census tracts also had features of more troublesome asthma at enrollment, with a longer duration of asthma, more historical asthma-related healthcare utilization, poorer asthma symptom control and asthma-related quality of life, and more impaired lung function measures with greater airflow limitation. By 12 months, children from high injustice census tracts also had a greater occurrence of an asthma exacerbation treated with systemic corticosteroids and continued to have poorer asthma symptom control and more impaired lung function measures than children from census tracts with less environmental injustice. These data highlight persistent disparities in environmental justice in metropolitan Atlanta, Georgia, that require additional study and action to improve health equity.
This is the first study to utilize the EJI in an asthma population. Although the EJI is not intended for definitive labeling of injustice or representation of risk for individual patients, the results are potentially useful for identifying areas that require additional action and for measuring progress in areas where new policies are implemented.(24) Our results are also consistent with other studies. In a recent analysis of adults with asthma in Pennsylvania, patients from census tracts with more people of color and impoverished residents had more exposure to traffic-related air pollutants (particularly nitrogen dioxide and black carbon) and greater odds of severe asthma.(37) In unadjusted analyses, patients living in these census tracts were also less likely to be White, more likely to have public insurance or be unemployed, and had lesser college degree attainment.(37) Similarly, a separate analysis of census tracts in Louisiana at or above the 75th percentile for the National Air Toxics Assessment’s Respiratory Hazard Index, particulate matter, and ozone found that those same census tracts had a disproportionately high burden of asthma, with higher asthma prevalence and hospitalizations.(38) Furthermore, a recent analysis of children in California using the CalEnviroScreen tool, a composite measure similar to the EJI, found that every unit increase in the CalEnviroScreen score was associated with a 1.6% increased rate of pediatric hospitalizations for asthma.(39) In that same study, every unit increase in racial/ethnic segregation and diesel particulate matter was also associated with an increase of 1.1% and 0.2% in the rate of childhood asthma.(39) However, the CalEnviroScreen model differs from the EJI in that it multiplies pollution burden scores by a single measure of population characteristics, and therefore gives the environmental factors representing exposures more weight in the model.(24)
In the present study, we enrolled children from metropolitan Atlanta, Georgia. Therefore, many of the airborne pollutant exposures such as ozone and particulate matter captured by the EJI Environmental Burden Module were quite high in most participants, with a percentile ranking of 0.86 versus 0.92 for low versus high injustice census tracts. Therefore, the air pollution differences that we observed in the present study were primarily related to diesel and residential proximity to railroads and heavily trafficked roads. Although our observation of greater exacerbation occurrence (with a non-significant trend toward greater hospitalization) in children residing in census tracts higher air pollution supports other literature,(7, 40) we recognize that there are limitations to the EJI with regard to the quantification of air pollution. First, the ozone and particulate matter indicators reflect a three-year average from 2014–2016 and cannot be interpreted as causal predictors. Diesel exhaust is also a complex mixture of a variety of gases, aerosols and particulate substances that varies according to the vehicle, motor and fuel utilized. However, our intent was not to model these factors as predictors with lag effects on asthma outcomes, but to instead to characterize cumulative impacts and patterns of environmental injustice across communities of children with asthma. This work was stimulated by a recent Working Group report of the American Academy of Allergy, Asthma and Immunology(41) and recognizes that although community engagement is a key component of successful environmental intervention programs for asthma,(42) identification of the communities at greatest need for intervention is an essential first step.
To our knowledge, this is one of the first studies to explore environmental justice and its potential effects in children with asthma through examination of both census-tract level variables and individual patient-level variables. A strength of the EJI is its direct incorporation of social vulnerability into the models, instead of adjusting for these crucial variables after the fact. However, although our sample size was relatively large and the children were well characterized, there are limitations with this work. First, this study did not leverage a population sample but instead utilized a convenience sample of children visiting an academic medical center who consented to participate in asthma research. The relatively high percentage of children who were hospitalized for asthma in the year prior to baseline evaluation also indicates a study population of high morbidity that is not necessarily representative of the general pediatric asthma population. Because of the convenience nature of our study, not all census tracts were represented, so it is possible that there are children living in census tracts with better or worse environmental justice who were not represented in the study. Our study patients were also predominantly English-speaking and may have had fewer challenges accessing care and improved health literacy. Although the prescribed asthma medications did not differ between our EJI tertiles, which potentially reflects a similar quality of care, we cannot rule out systemic or structural racism or other unmeasured variables related to asthma care. There are also a multitude of other variables that synergize and interact to influence health outcomes that were not captured in our study or by the EJI, such as adverse childhood experiences or indoor exposures such as pollutants and rodent sensitization. This study also involved relatively few patient-reported outcomes and may have underestimated (or overestimated) asthma burden attributable to the environment. The follow-up interval was also relatively short. Although some exacerbations may have occurred outside our healthcare system and may have been missed, our electronic medical record also integrates with the electronic medical records from the major adult healthcare systems in Georgia, which increases our ability to detect unscheduled healthcare utilization at other facilities. Finally, it is also recognized that the EJI is not a definitive tool for labeling environmental injustice and cannot be used to represent risk for a given community or individual.(23, 24)
In summary, this study demonstrates associations between environmental injustice and asthma outcomes in school-age children in metropolitan Atlanta, Georgia. Although these associations are multi-faceted and complex, this study provides critical evidence that residence matters. While additional investigation is needed, this study also provides preliminary data for prioritization of community-level interventions in areas with the greatest asthma-related, adverse environmental effects.
Supplementary Material
Highlights Box.
What is already known about this topic?
Environmental justice demands that no group of people suffer disproportionately from the negative effects of environmental exposures, yet historically marginalized and socially disadvantaged populations continue to reside in areas with more exposures and asthma burden.
What does this article add to our knowledge?
Children residing in census tracts with the greatest environmental injustice have more exposure to pollutants, poorer asthma symptom control and quality of life, more impaired lung function and greater healthcare utilization for asthma.
How does this study impact current management guidelines?
Environmental injustice contributes to asthma outcomes in children and is associated with more exacerbations and persistently more asthma symptoms, poorer asthma control, and reduced lung function. Community-level interventions are needed for asthma equity.
Acknowledgments
This work was supported by the National Institutes of Health under grant award numbers R01 NR018666, R01 NR018666-05W1, K24 NR018866, and UL1 TR002378.
Abbreviations
- ACQ
Asthma Control Questionnaire
- EJI
Environmental Justice Index
- FEV1
Forced expiratory volume in one second
- FVC
Forced vital capacity
- GEOID
Geographic identifier
- IgE
Immunoglobulin E
- PAQLQ
Pediatric Asthma Quality of Life Questionnaire
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Author Disclosures: Jocelyn R. Grunwell, Abby D. Mutic, Idil D. Ezhuthachan, Carrie Mason, Mallory Tidwell, Cherish Caldwell, Jalicae Norwood, Sydney Zach, Natalie Jordan, and Anne M. Fitzpatrick have no disclosures pertaining to this submitted work.
References
- 1.Environmental Justice. United States Environmental Protection Agency. Available online at http://epa.gov/environmentaljustice/learn-about-environmental-justice. Last accessed September 26, 2023.
- 2.Funes Y The Father of Environmental Justice Exposes the Geography of Inequity. Nature. 2023;621(7979):S25–S9. [DOI] [PubMed] [Google Scholar]
- 3.Bullard RD. Solid waste sites and the black Houston community. Sociol Inq. 1983;53(2–3):273–88. [DOI] [PubMed] [Google Scholar]
- 4.Chen F, Chen JT, Hart JE, Coull BA, Scammell MK, Chu MT, et al. Disparities in joint exposure to environmental and social stressors in urban households in Greater Boston. Environ Res. 2023;238(Pt 1):117104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Patel VR, Jella T, Gupta A, Nassel A, Ibrahim A, Hussaini SMQ. Association of Neighborhood-Level Environmental Injustice With Health Status in the US. JAMA Intern Med. 2023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Querdibitty CD, Campbell J, Wetherill MS, Salvatore AL. Geographic and social economic disparities in the risk of exposure to ambient air respiratory toxicants at Oklahoma licensed early care and education facilities. Environ Res. 2023;218:114975. [DOI] [PubMed] [Google Scholar]
- 7.Bi J, D’Souza RR, Moss S, Senthilkumar N, Russell AG, Scovronick NC, et al. Acute Effects of Ambient Air Pollution on Asthma Emergency Department Visits in Ten U.S. States. Environ Health Perspect. 2023;131(4):47003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kline O, Prunicki M. Climate change impacts on children’s respiratory health. Curr Opin Pediatr. 2023;35(3):350–5. [DOI] [PubMed] [Google Scholar]
- 9.Dondi A, Carbone C, Manieri E, Zama D, Del Bono C, Betti L, et al. Outdoor Air Pollution and Childhood Respiratory Disease: The Role of Oxidative Stress. Int J Mol Sci. 2023;24(5). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Muttoo S, Jeena PM, Roosli M, de Hoogh K, Meliefste K, Tularam H, et al. Effect of short-term exposure to ambient nitrogen dioxide and particulate matter on repeated lung function measures in infancy: A South African birth cohort. Environ Res. 2022;213:113645. [DOI] [PubMed] [Google Scholar]
- 11.Yu Z, Merid SK, Bellander T, Bergstrom A, Eneroth K, Georgelis A, et al. Associations of improved air quality with lung function growth from childhood to adulthood: the BAMSE study. Eur Respir J. 2023;61(5). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Kelchtermans J, Mentch F, Hakonarson H. Ambient air pollution sensitivity and severity of pediatric asthma. J Expo Sci Environ Epidemiol. 2023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Lee EK, Donley G, Ciesielski TH, Gill I, Yamoah O, Roche A, et al. Health outcomes in redlined versus non-redlined neighborhoods: A systematic review and meta-analysis. Soc Sci Med. 2022;294:114696. [DOI] [PubMed] [Google Scholar]
- 14.United States Environmental Protection Agency (U.S. E.P.A.). America’s Children and the Environment, October 2019. Washington, D.C.: 2019. [Google Scholar]
- 15.Martenies SE, Milando CW, Williams GO, Batterman SA. Disease and Health Inequalities Attributable to Air Pollutant Exposure in Detroit, Michigan. Int J Environ Res Public Health. 2017;14(10). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Rosser F, Han YY, Rothenberger SD, Forno E, Mair C, Celedon JC. Air Quality Index and Emergency Department Visits and Hospitalizations for Childhood Asthma. Ann Am Thorac Soc. 2022;19(7):1139–48. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Cheng J, Tong S, Su H, Xu Z. Association between sub-daily exposure to ambient air pollution and risk of asthma exacerbations in Australian children. Environ Res. 2022;212(Pt D):113556. [DOI] [PubMed] [Google Scholar]
- 18.Ho YN, Cheng FJ, Tsai MT, Tsai CM, Chuang PC, Cheng CY. Fine particulate matter constituents associated with emergency room visits for pediatric asthma: a time-stratified case-crossover study in an urban area. BMC Public Health. 2021;21(1):1593. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Li H, Li X, Zheng H, Liu L, Wu Y, Zhou Y, et al. Ultrafine particulate air pollution and pediatric emergency-department visits for main respiratory diseases in Shanghai, China. Sci Total Environ. 2021;775:145777. [DOI] [PubMed] [Google Scholar]
- 20.Zheng XY, Orellano P, Lin HL, Jiang M, Guan WJ. Short-term exposure to ozone, nitrogen dioxide, and sulphur dioxide and emergency department visits and hospital admissions due to asthma: A systematic review and meta-analysis. Environ Int. 2021;150:106435. [DOI] [PubMed] [Google Scholar]
- 21.Byrwa-Hill BM, Venkat A, Presto AA, Rager JR, Gentile D, Talbott E. Lagged Association of Ambient Outdoor Air Pollutants with Asthma-Related Emergency Department Visits within the Pittsburgh Region. Int J Environ Res Public Health. 2020;17(22). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.United States Environmental Protection Agency. 2021. Climate change and social vulnerability in the United States: A Focus on Six Impacts. U.S. Environmental Protection Agency, EPA; 430-R-21–003. [Google Scholar]
- 23.Centers for Disease Control and Prevention and Agency for Toxic Substances Disease Registry. 2022. Environmental Justic Index. Available at http://www.atsdr.cdc.gov/placeandhealth/eji/index.html. Last accessed August 21, 2023.
- 24.Technical documentation for the Environmental Justice Index 2022. Available online at https://www.atsdr.cdc.gov/placeandhealth/eji/technical_documentation.html. Last accessed August 31, 2023.
- 25.Grunwell JR, Opolka C, Mason C, Fitzpatrick AM. Geospatial Analysis of Social Determinants of Health Identifies Neighborhood Hot Spots Associated With Pediatric Intensive Care Use for Life-Threatening Asthma. J Allergy Clin Immunol Pract. 2022;10(4):981–91 e1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Najjar N, Opolka C, Fitzpatrick AM, Grunwell JR. Geospatial Analysis of Social Determinants of Health Identifies Neighborhood Hot Spots Associated With Pediatric Intensive Care Use for Acute Respiratory Failure Requiring Mechanical Ventilation. Pediatr Crit Care Med. 2022;23(8):606–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Cambon J tidygeocoder: Geocoding Made Easy. R package version 1.0.1 ed., 2020. Available at https://jessecambon.github.io/tidygeocoder. [Google Scholar]
- 28.Walker K tigris: Load Census TIGER/Line Shapefiles. 2020.
- 29.Juniper EF, O’Byrne PM, Roberts JN. Measuring asthma control in group studies: do we need airway calibre and rescue beta2-agonist use? Respir Med. 2001;95(5):319–23. [DOI] [PubMed] [Google Scholar]
- 30.Juniper EF, Svensson K, Mork AC, Stahl E. Measurement properties and interpretation of three shortened versions of the asthma control questionnaire. Respir Med. 2005;99(5):553–8. [DOI] [PubMed] [Google Scholar]
- 31.Juniper EF, Gruffydd-Jones K, Ward S, Svensson K. Asthma Control Questionnaire in children: validation, measurement properties, interpretation. Eur Respir J. 2010;36(6):1410–6. [DOI] [PubMed] [Google Scholar]
- 32.Juniper EF, Guyatt GH, Feeny DH, Ferrie PJ, Griffith LE, Townsend M. Measuring quality of life in children with asthma. Qual Life Res. 1996;5(1):35–46. [DOI] [PubMed] [Google Scholar]
- 33.Stanojevic S, Kaminsky DA, Miller MR, Thompson B, Aliverti A, Barjaktarevic I, et al. ERS/ATS technical standard on interpretive strategies for routine lung function tests. Eur Respir J. 2022;60(1). [DOI] [PubMed] [Google Scholar]
- 34.Quanjer PH, Stanojevic S, Cole TJ, Baur X, Hall GL, Culver BH, et al. Multi-ethnic reference values for spirometry for the 3–95-yr age range: the global lung function 2012 equations. Eur Respir J. 2012;40(6):1324–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Horvath I, Barnes PJ, Loukides S, Sterk PJ, Hogman M, Olin AC, et al. A European Respiratory Society technical standard: exhaled biomarkers in lung disease. Eur Respir J. 2017;49(4). [DOI] [PubMed] [Google Scholar]
- 36.Fuhlbrigge A, Peden D, Apter AJ, Boushey HA, Camargo CA Jr., Gern J, et al. Asthma outcomes: exacerbations. J Allergy Clin Immunol. 2012;129(3 Suppl):S34–48. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Byrwa-Hill BM, Morphew TL, Presto AA, Fabisiak JP, Wenzel SE. Living in environmental justice areas worsens asthma severity and control: Differential interactions with disease duration, age at onset, and pollution. J Allergy Clin Immunol. 2023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Bakshi A, Van Doren A, Maser C, Aubin K, Stewart C, Soileau S, et al. Identifying Louisiana communities at the crossroads of environmental and social vulnerability, COVID-19, and asthma. PLoS One. 2022;17(2):e0264336. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Alcala E, Brown P, Capitman JA, Gonzalez M, Cisneros R. Cumulative Impact of Environmental Pollution and Population Vulnerability on Pediatric Asthma Hospitalizations: A Multilevel Analysis of CalEnviroScreen. Int J Environ Res Public Health. 2019;16(15). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Hardell J, Silver EJ, Kavouras I, Lee DS, Gross E. Childhood asthma in the Bronx, NY; the impact of pollutants on length of hospital stay. J Asthma. 2023:1–8. [DOI] [PubMed] [Google Scholar]
- 41.Burbank AJ, Hernandez ML, Jefferson A, Perry TT, Phipatanakul W, Poole J, et al. Environmental justice and allergic disease: A Work Group Report of the AAAAI Environmental Exposure and Respiratory Health Committee and the Diversity, Equity and Inclusion Committee. J Allergy Clin Immunol. 2023;151(3):656–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Bryant-Stephens TC, Strane D, Robinson EK, Bhambhani S, Kenyon CC. Housing and asthma disparities. J Allergy Clin Immunol. 2021;148(5):1121–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.