Climate Change and Children's Health and Well-Being in the United States Appendix C: Supplemental Information for Analyses in the Air Quality Chapter This appendix describes methods, data sources, and assumptions for the air quality analyses presented in Chapter 4 of the main report. First is the information for the detailed analysis of children's health outcomes linked with exposure to fine particulate matter (PM2.5) and ground- level ozone (O3). Second is information required for the discussion of emerging literature linking wildfire smoke and fetal health outcomes. Detailed Analysis of Air Quality and Children's Health This section includes details of the air quality and children's health analysis: a summary of climate studies used in the analysis, a summary of air quality epidemiological studies used in the analysis, analysis steps, detailed results, and limitations of the approach. SUMMARY OF CLIMATE STUDIES USED IN THIS ANALYSIS This analysis considers pollutant sources linked to climate change that result in heightened levels of PM2.5 and O3. These include the following: • Climate penalty, which refers to changes in air quality resulting from climate-induced changes in temperature humidity, precipitation, and wind patterns, which all increase the secondary formation of O3 and PM2.5. • Southwest dust, which refers to changes in ambient dust levels associated with increasing aridity and is restricted to four southwestern U.S. states: Utah, Colorado, Arizona, and New Mexico. • Wildfires, which herein refers to nationwide changes in pollutant concentrations and associated health impacts attributable specifically to wildfire activity in the western U.S. The following studies are those used to quantify health effects in children, associated with these sources of pollutants: CLIMATE PENALTY: FANN ET AL. (2021)1 Fann et al. estimated mortality risk associated with changing air quality; specifically, O3 and PM2.5 concentrations. The authors show that changes in climate increase the population-weighted O3 and PM2.5 concentrations throughout the U.S. This analysis uses the Fann et al. air quality surfaces (i.e., changes in concentrations of pollutants in response to changes in meteorology and emissions) to quantify health effects attributable to exposures to PM2.5 and O3. The underlying study modeled future pollutant concentrations using two GCMs (CanESM2 and GFDL-CM3) and two alternative simulated air pollutant emissions scenarios, one which uses a 2011 inventory that estimates pollution burden from all sources as of that year, and an alternative 2040 dataset that accounts for the April 2023 1 ------- Climate Change and Children's Health and Well-Being in the United States implementation of a suite of regulatory policies on stationary and mobile emissions sources. The analysis completed in this EPA report considers the average of health impacts across both GCMs, under the 2011 emissions scenario. Health impacts associated with the alternative 2040 emissions inventory are estimated to be approximately 40% lower than health impacts associated with the 2011 inventory used in this analysis. SOUTHWEST DUST: ACHAKULWISUT ET AL. (2019)2 Achakulwisut et al. estimated the health burden resulting from changes in fine and coarse airborne dust exposure due to climate change in the Southwest. They found that, by the end of the century, climate change could lead to fine dust levels increasing by 57%, and coarse dust levels increasing by 38%. This analysis used projected PM2.5 concentrations for six GCMs (CanESM2, CCSM4, GFDL-CM3, GISS-E2-R, HADGEM2, and MIROC5) derived from a network of 34 monitors from the underlying study, spanning Arizona, Colorado, New Mexico, and Utah. WILDFIRES: NEUMANN ETAL. (2021)3 Neumann et al. estimated health impacts from wildfire emissions of black carbon and organic carbon by modeling changes in wildfire activity for the western region of the contiguous U.S. They found that climatic factors increase wildfire pollutant emissions by an average of 0.40% to 0.71% per year, and these emissions result in spatially weighted wildfire PM2.5 concentrations more than double for some climate model projections by the end of the 21st century. Future concentrations of PM2.5 from western wildfires were projected for five GCMs (CanESM2, CCSM4, GISS-E2-R, HADGEM2, and MIROC5) and extend nationwide, as emissions associated with wildfires typically travel eastward across the country. SUMMARY OF AIR QUALITY EPIDEMIOLOGY STUDIES USED IN THIS ANALYSIS Numerous epidemiological studies document the relationship between degraded air quality and human morbidity or mortality. This analysis draws on evidence from seven studies that identify the magnitude of these relationships for children specifically (summarized below). These studies have been parameterized for use with the U.S. EPA's Environmental Benefits Mapping and Analysis Program (BenMAP, https://www.epa.gov/benmap), a tool that estimates the human health impacts of air quality changes at a refined spatial scale. BenMAP is used to determine the change in ambient air pollution based on user-specified air quality data and relates the change in pollution concentrations with certain health effects using concentration-response functions derived from epidemiology studies. BenMAP applies that relationship to the population experiencing the change in pollution exposure to calculate health impacts. Table 1 maps the studies described above to their risk measures and includes age groups, BenMAP surfaces, and pollutants. The studies described below are listed in the same order as they appear in Table 1. Note that these studies calculate outputs such as hazard ratios, rate ratios, relative risks, or odds ratios, which are alternative measures of association between an exposure (in this case, to air pollution) and the incidence of a specific April 2023 2 ------- Climate Change and Children's Health and Well-Being in the United States adverse health effect. Some studies instead statistically estimate a regression function, where the relevant coefficient on the exposure variable provides the estimate of the association between exposure to air pollution and incidence. INCIDENCE OF ASTHMA: TETREAULT ET AL. (2016)4 Tetreault et al. investigated the relationship between childhood asthma onset and long-term pollution exposure (PM2.5, O3, and NO2). The authors followed a cohort of 1,200,000 children born in Quebec, Canada, from 1996 to 2011, from birth to approximately age 6, and mapped asthma incidence with residential exposures to air pollutants. The study defined the onset of asthma as a hospital-discharged diagnosis of asthma or two reports of asthma from two separate physicians within a two-year period. The authors used Cox proportional hazard models to estimate the association between asthma onset and pollution exposure, controlling for demographics and socioeconomic status. The coefficient and standard error for PM2.5 were estimated from a hazard ratio of 1.33 (95% CI 1.31-1.34) for a 6.53 |ig/m3 increase in annual PM2.5 concentration. The coefficient and standard error for O3 were estimated from a warm-season hazard ratio of 1.07 (95% CI 1.06-1.08) for a 3.26 ppb increase in annual O3 concentrations. INCIDENCE OF HAY FEVER: PARKER ET AL. (2009)5 Parker et al. investigated the associations between long-term O3 exposure and respiratory allergies (defined as hay fever or respiratory allergy symptoms) among 73,000 children nationwide aged 3-17, between 1999 and 2005. The analysis was conducted using logistic regression models, adjusted for demographic and socioeconomic factors. The coefficient and standard error for PM2.5 are based on the odds ratio of 1.29 (95% CI 1.07-1.56) for a 10 |ig/m3 increase in PM2.5 concentration. The coefficient and standard error for O3 are based on the odds ratio of 1.18 (95% CI 1.09-1.27) for a 10 ppb increase in warm-season daily mean O3. SCHOOL DAYS LOST: GILLILAND ET AL. (2001)5 Gilliland et al. examined the association between air pollution and school absenteeism among fourth grade children (aged 9-10) in twelve southern California communities in 1996. The relationship is applied here to all school-age children (aged 5-17). The authors used school records to collect daily absence data and parental telephone interviews to identify causes. Using an average length of absence at baseline, they determined how this could relate to limiting new absences in the future. The authors used 15- and 30-day distributed lag models to quantify the association between O3 and school absences. O3 levels were positively associated with all school absence measures. The coefficient and standard error are based on a percent increase of 16.3% (95% CI -2.6%-38.9%) associated with a 20 ppb increase in 8-hour average O3 concentration. EMERGENCY DEPARTMENT VISITS FOR ASTHMA: ALHANTI ETAL. (2016)7 Alhanti et al. studied the relationship between daily PM2.5 concentrations and emergency department (ED) visits for asthma among residents of all ages (patient-level data) in Atlanta (1993- 2009), Dallas (2006-2009), and St. Louis (2001-2007). The authors ran city-specific daily time-series April 2023 3 ------- Climate Change and Children's Health and Well-Being in the United States Poisson regression models by age group (0-4, and 5-18 were included in this analysis) and performed additional analyses stratified by race and sex. The coefficient and standard error for PM2.5 are estimated from rate ratios of 1.01 (95% CI 1.00-1.02) and 1.02 (95% CI 1.01-1.04) associated with an 8 |ig/m3 increase in PM2.5 concentration for children aged 0-4 and 5-18, respectively. EMERGENCY DEPARTMENT VISITS FOR ASTHMA: MAR AND KOENIG (2009)8 Mar and Koenig studied the relationship between O3 exposure and asthma hospitalizations in the Seattle area from 1998 to 2002. The authors used hospital data on daily asthma cases with local monitored O3 concentrations to assess the association between asthma visits to the ED and air pollution. The coefficient and standard error are estimated from a relative risk of 1.11 (95% CI 1.02- 1.21) for a 10 ppb increase in daily 8-hour maximum summer O3 concentration. HOSPITAL ADMISSIONS FOR RESPIRATORY ISSUES: OSTRO ETAL. (2009)9 Ostro et al. estimated the association between ambient PM2.5 and respiratory diseases in children aged 5 to 19 in California. Hospital admission data was aggregated for all respiratory diseases to the county level to create a daily time series of admissions for each county. Authors analyzed data using a Poisson regression with time, day of the week, temperature, relative humidity, and pollutant as explanatory variables. They controlled for seasonality and time dependent effects by including a natural spline smoother for the daily time trend and meteorology. The coefficient and standard error are estimated from an excess risk of 4.1% (95% CI 1.8%-6.4%) for a 14.6 |ig/m3 increase in daily mean PM2.5 concentration. INFANT MORTALITY: WOODRUFF ETAL. (2008)10 Woodruff et al. examined the relationship between long-term exposure to PM2.5 air pollution and postneonatal (i.e., from 28 days through the first year of life) infant mortality in 3,600,000 live births from 96 counties across the U.S. between 1999 and 2002. They used logistic regression models that incorporated generalized estimating equations to estimate odds ratios for all-cause and cause- specific postneonatal mortality as a result of exposure to air pollution. The coefficient and standard error are estimated from an odds ratio of 1.04 (95% CI 0.98-1.11) associated with a change of 7|ig/m3 in mean PM2.5 exposure level. ANALYSIS STEPS Chapter 4 of this report quantifies the effects of pollutant exposures on children's health. This analysis relies on pollutant source information from Fann et al. (2021), Achakulwisut et al. (2019), and Neumann et al. (2021) and effect estimates from numerous epidemiological studies, as summarized above. Table 2 details the analytic steps, data sources, and assumptions used to project the various measures of children's health impacts resulting from air quality degradation linked to climate change. As described in the table, this analysis summarizes impacts by degree of global warming. For more information on how the analysis applies thresholds of degrees of global warming, see methods described in Chapter 2 of the main report and Appendix A. This analysis considers all April 2023 4 ------- Climate Change and Children's Health and Well-Being in the United States geographies in the contiguous United States, except for the Southwest dust pollutant source, which is limited to four southwestern states (UT, CO, AZ, NM). April 2023 5 ------- Climate Change and Children's Health and Well-Being in the United States Table 1: Risk Measures, Studies, Age Groups, and BenMAP Surfaces Considered for Each Pollutant Source Age Range Pollutant Source Risk Measure Study Climate Penalty Southwest Dust Wildfire PM2.5 o3 (pm25) (pm25) Incidence of asthma Tetreault et al. (2016) 0-17 X X X X Incidence of hay fever (rhinitis) Parker et al. (2009) 3-17 X X X X School days lost, all cause Gilliland et al. (2001) 5-17 X ED visits associated with asthma Alhanti et al. (2016) Mar and Koenig (2009) 0-18 X X X X Hospital admissions for respiratory issues Ostro et al. (2009) 0-18 X X X Infant mortality Woodruff et al. (2008) 0-0* X X X * Infant mortality estimated for postneonatal infants (i.e., those aged 28-365 days) April 2023 6 ------- Climate Change and Children's Health and Well-Being in the United States Table 2: Analytic Steps in Climate Change Impacts on Air Quality and Children's Health Analysis Step Data Methods, Assumptions, Notes 1. Identify baseline incidence of health and County- or national-level This analysis used the finest scale data where available. ai well-being impacts under baseline climate incidence by health effect Specifically, county-level baseline incidence data for lost school and population obtained from BenMAP days, asthma ED visits, infant mortality, and respiratory hospital — ^ admissions were used. This analysis includes national-level IS 00 baseline incidence data for new cases of asthma and incidence of hay fever (allergic rhinitis). 2. Utilize projected PM2.5 and 03 Present and future pollutant Pollutant data is available at different spatial scales and for concentrations related to climate penalty, concentrations: different geographic regions. Different climate models are utilized southwest dust, and wildfires • Climate penalty: Fann et within a set of six CMIP5 scenarios for each analysis, and results 1— 0 (/) al. (2021) are binned based on 21st century arrival times for each GCM. 0) • Southwest dust: • Climate Penalty: Nationwide analysis, using climate data at ¦*-» to Achakulwisut et al. 36-km scale. Two GCMs: CanESM2, GFDL-CM3 0) (2019) • Southwest dust: Analysis limited to four southwestern states ro E r 1 • Wildfires: Neumannetal. (UT, CO, AZ, NM). Baseline pollutant concentrations from 34 (2021) monitor sites. Six GCMs: CanESM2, CCSM4, GFDL-CM3, GISS- W ai E2-R, HADGEM2, and MIROC5 3 • Wildfires: Nationwide analysis, using climate data at 0.25 x 3 u_ 0.25-(latitude/longitude) degree scale. Five GCMs: CanESM2, CCSM4, GISS-E2-R, HADGEM2, and MIROC5. PM25 air quality outputs generated at 0.5 x 0.625 (latitude/longitude) degree grid scale. 3. Estimate the increase in incidence of Health impact functions are The health impact functions used in this analysis are specific to health effects associated with each degree-C derived from epidemiological children of different age ranges, presented in Table 1. These c 0 increase in global mean temperatures studies described previously, represent the best available studies with effects specific to 4-1 r— parameterized in BenMAP children used in other EPA analyses. This analysis excludes health (j c aj impacts to children outside of these age ranges (e.g., it does not it "D LU See Chapter 2 of the main quantify incidence of hay fever/rhinitis among those younger 0) -E »- U report and Appendix A for than three years old). 3 +-< details on population 3 u_ methods and data sources used throughout the analysis. April 2023 7 ------- Climate Change arid Children's Health and Well-Being in the United States PROJECTIONS OF PM2.5 AND O3 Figure 1 shows the change from 2000 baseline levels of PM2.5 associated with a 2°C (top panel) and 4°C (bottom panel) increase in global mean temperature, based on projections used in Fann et al. Figure 2 shows the change from 2000 baseline levels of O3 associated with a 2°C (top panel) and 4°C (bottom panel) increase in global mean temperature, based on projections used in Fann et al. Figure 1: Future PM2.5 Concentrations at 2°C and 4°C Increase in Global Mean Temperature 2°C of Global Warming Change in Fine Particulate Matter from Historical Baseline jjg/m3 <-0.23 -0.23 - 0 ¦10-0.09 ¦ 0.09-0.19 ¦ 0.19-0.31 ¦ 0.31 -0.50 ¦ 0.50-1.31 4°C of Global Warming Change in Fine Particulate Matter from Historical Baseline |jg/m3 <-0.23 -0.23 - 0 ¦ 0-0.09 ¦ 0.09-0.19 ¦ 0.19-0.31 ¦ 0.31 -0.50 ¦ 0.50-1.31 Source: USE PA (2021)11 April 2023 8 ------- Climate Change arid Children's Health and Well-Being in the United States Figure 2: Future 03 Concentrations at 2°C and 4°C increase in Global Mean Temperature 2°C of Global Warming Change in Ozone from Historical Baseline ppb -4.00 - -0.79 -0.79 - 0 0 - 0.53 0.53-1.35 1.35-2.42 2.42-3.77 3.77 -24.34 4°C of Global Warming Change in Ozone from Historical Baseline ppb -4.00 - -0.79 -0.79-0 0 - 0.53 0.53-1.35 1.35-2.42 2.42-3.77 3.77-24.34 Source: Fann et al. (2021) and USEPA (2021)12 April 2023 ------- Climate Change and Children's Health and Well-Being in the United States EFFECTS ON CHILDREN RESULTS Table 3 presents the results of the analysis assuming population growth (see Chapter 2 and Appendix A). The analysis estimates additional health impacts attributable to climate change relative to the baseline period and sums the impacts for each health effect across the three pollutant sources (climate penalty, southwest dust, and wildfire). Table 4 provides the same estimates but assumes population remains constant at 2010 levels, isolating the influence of climate change specifically. Table 3: Projected Annual Risks to Children's Health Associated with Future PM2.5 and 03 (with Population Growth) (1) (2) (3) (4) (5) (6) Degree of Global New Cases of Incidence of Hay School Days Lost ED Visits for Hospital Admissions for Infant Warming (°C) Asthma Fever/Rhinitis from All Causes Asthma Respiratory Illness Mortality (Aged 0-17) (Aged 3-17) (Aged 5-17) (Aged 0-18) (Aged 0-18) (Aged 0-0) 19,200 126,000 1,270,000 3,450 173 4 rc (14,400 to (92,000 to 159,000) (960,000 to 580,000) (2,560 to (117 to 224) (2 to 6) 24,800) 4,370) 34,500 228,000 2,240,000 6,240 332 7 2°C (27,900 to (179,000 to 276,000) (1,850,000 to (5,210 to (230 to 430) (4 to 10) 42,800) 2,630,000) 7,330) 57,900 367,000 3,590,000 10,300 537 11 3°C (51,400 to (318,000 to 418,000) (3,570,000 to (9,930 to (292 to 782) (5 to 16) 66,600) 3,610,000) 10,800) 89,600 554,000 5,480,000 15,800 785 15 4°C (74,100 to (447,000 to 662,000) (5,170,000 to (14,500 to (353 to 1,220) (6 to 25) 108,000) 5,790,000) 17,200) 5°C 134,000 771,000 7,630,000 22,400 1,160 24 Notes: All estimates presented in the table are incremental relative to baseline risks and convey impacts per year: (1) 841,000 new asthma cases, (2) 11.9 million incidences of hay fever/rhinitis, (3) 183 million school days lost from all causes, (4) 733,000 ED visits for asthma, (5) 429,000 hospital admissions for respiratory illness, and (6) 8,960 infant deaths. The table displays the average and range across climate models; a range for 5°C is not feasible because only one climate model reaches this temperature threshold before 2100. See Table 2 for analytic details. April 2023 10 ------- Climate Change and Children's Health and Well-Being in the United States Table 4: Projected Annual Risks to Children's Health Associated with Future PM2.5 and 03 (2010 Population) (1) (2) (3) (4) (5) (6) Degree of Global New Cases of Incidence of Hay School Days Lost from ED Visits for Hospital Admissions for Infant Warming (°C) Asthma Fever/Rhinitis All Causes Asthma Respiratory Illness Mortality (Aged 0-17) (Aged 3-17) (Aged 5-17) (Aged 0-18) (Aged 0-18) (Aged 0-0) 18,600 112,000 1,150,000 3,180 143 6 rc (13,700 to (81,900 to 141,000) (867,000 to 1,440,000) (2,340 to (95 to 187) (3 to 8) 23,400) 4,010) 32,200 194,000 1,950,000 5,510 258 11 2°C (25,500 to (152,000 to 235,000) (1,600,000 to (4,550 to (178 to 338) (7 to 15) 39,000) 2,300,000) 6,470) 49,300 294,000 2,940,000 8,460 393 17 3°C (43,600 to (255,000 to 333,000) (2,930,000 to (8,210 to (204 to 586) (8 to 25) 55,100) 2,960,000) 8,710) 72,900 431,000 4,360,000 12,600 561 24 4°C (60,200 to (351,000 to 510,000) (4,160,000 to (11,600 to (231 to 890) (9 to 38) 85,500) 4,560,000) 13,500) 5°C 100,000 585,000 5,880,000 17,100 840 36 Notes: See Table 3. April 2023 11 ------- Climate Change and Children's Health and Well-Being in the United States Figures 3 and 4 show the estimated change in childhood asthma diagnoses per 100,000 children aged 0-17 at 2°C and 4°C of global warming at the county level. For each figure, the top panel shows the combined impacts across pollution sources, which are split out by source below. The five states with largest impacts per 100,000 children are outlined in black for each pollutant source and listed below each map. Tables 5 and 6 then follow with the number of cases per 100,000 children for each state at 2°C and 4°C of global warming specifically to provide perspective on the range of impacts across states, although there can be considerable heterogeneity within states (see Figures 3 and 4). Figure 5 shows the change in total childhood asthma diagnoses for children aged 0-17 at 2°C and 4°C of global warming at the county level. Impacts are generally highest in areas with the large children's populations. The five states with largest total impacts are outlined in black and listed below each map. The relevant quantities or rates presented in each figure are provided in parentheses after the state name in the lists of top 5 states. Figure 6 then shows the county-level impacts across pollutant sources for another impact of air quality on children's well-being: the number of annual school days lost. This additional endpoint demonstrates that the spatial distribution is fairly consistent across impacts considered in this analysis. April 2023 12 ------- Climate Change arid Children's Health and Well-Being in the United States Figure 3: Estimated Change in New Annual Asthma Diagnoses Per 100,000 Children (Aged 0-17) at 2°C Global Warming (with Population Growth) All Pollution Sources Top five states (rate/100,000 in parentheses): D.C. (90), OH (83), WA (81), KY (80), MD (80) Climate Penalty, PM2.5 Climate Penalty, 03 Top five states: SC (33), NC (31), GA (30), AL (29), WV (23) Southwest Dust vf V Top five states: IL (88), OH (87), D.C. (77), IN (73), MD (68) Wildfire Top four states: NM (13.0), AZ(13.0), UT (12.9), CO (12.9) <0 1 -27 28 - 47 list A J > l J U4 Top five states: MT (35), OR (31), ID (21), WY (19), CA (16) 48 -68 !¦ 69- 107 ¦108-692 Note: These maps describe the projected change in new annual asthma diagnoses per 100,000 children at 2°C of global warming relative to the baseline (1986-2005). Darker shading conveys larger increases while lighter shading conveys small increases. The five states with the largest increases on average are outlined in black. April 2023 13 ------- Climate Change and Children's Health and Well-Being in the United States Table 5: Estimated New Annual Asthma Diagnoses Per 100,000 Children by State with 2°C Global Warming (with Population Growth) State Incidence Per 100,000 Children State Incidence Per 100,000 Children Washington, DC 90 Rhode Island 40 Ohio 83 Alabama 40 Washington 81 Oregon 39 Kentucky 80 Wyoming 37 Maryland 80 Iowa 36 Virginia 77 Arkansas 36 West Virginia 76 Montana 34 Delaware 72 Connecticut 31 Colorado 70 New Mexico 30 Illinois 70 Nevada 29 New Jersey 65 South Dakota 29 Tennessee 61 Michigan 26 Indiana 61 Mississippi 25 North Carolina 60 Idaho 24 Pennsylvania 57 Wisconsin 24 Arizona 52 Minnesota 24 Kansas 52 Georgia 20 New York 51 Louisiana 13 South Carolina 50 North Dakota 10 Utah 48 Texas 8 Missouri 46 Florida 0 Nebraska 46 New Hampshire -4 Massachusetts 46 Maine -14 Oklahoma 46 Vermont -16 California 45 -- Notes: This table describes the projected new annual diagnoses per 100,000 children at 2°C of global warming using the methods described in Table 2 averaged to the state level. States are listed from largest to smallest impacts. April 2023 14 ------- I I Figure 4: Estimated Change in New Annual Asthma Diagnoses Per 100,000 Children (Aged 0-17) at 4°C Global Warming (with Population Growth) All Pollution Sources Climate Change arid Children's Health and Well-Being in the United States Top five states (rate/100,000 in parentheses): D.C. (214), OH (205), WA (203), MD (178), IL (169) Climate Penalty, PM2.5 Climate Penalty, 03 Top five states: AL (59), GA (54), SC (54), NC (50), WV (46) Southwest Dust Top five states: OH (200), IL (189), DC (186), WA (158), IN (155) Wildfire <0 1 -27 28-47 48 -68 ¦69- 107 ¦1108-692 Note: These maps describe the projected change in new annua! asthma diagnoses per 100,000 children at 4°C of global warming relative to the baseline (1986-2005). Darker shading conveys larger increases while lighter shading conveys small increases. The five states with the largest increases on average are outlined in black. April 2023 15 ------- Climate Change and Children's Health and Well-Being in the United States Table 6: Estimated New Annual Asthma Diagnoses Per 100,000 Children by State with 40C Global Warming (with Population Growth) State Incidence Per 100,000 Children State Incidence Per 100,000 Children Washington, DC 214 Michigan 91 Ohio 205 South Carolina 89 Washington 203 Georgia 88 Maryland 178 California 83 Illinois 169 Alabama 83 West Virginia 168 Iowa 74 Delaware 162 Arkansas 72 Virginia 162 Oregon 71 New Jersey 160 Wyoming 70 Kentucky 156 Wisconsin 68 Pennsylvania 143 Minnesota 65 Indiana 142 New Mexico 61 New York 141 South Dakota 58 Massachusetts 136 Mississippi 52 Colorado 133 Montana 47 Rhode Island 122 Idaho 43 Kansas 110 Nevada 40 Tennessee 109 Louisiana 39 North Carolina 106 North Dakota 32 Arizona 105 Texas 26 Missouri 100 New Hampshire 22 Utah 99 Florida 2 Connecticut 98 Vermont -9 Nebraska 97 Maine -11 Oklahoma 95 -- Notes: This table describes the projected new annual diagnoses per 100,000 children at 4°C of global warming using the methods described in Table 2 averaged to the state level. States are listed from largest to smallest impacts. April 2023 16 ------- Climate Change arid Children's Health and Well-Being in the United States Figure 5: Estimated Change in Total New Annual Asthma Diagnoses Among Children Aged o (with Population Growth) 2°C of Global Warming Top five states (excess diagnoses in parentheses): CA (3,640), NY (2,740), IL (2,570), OH (2,280), NJ (1,710) 4°C of Global Warming Top five states: CA (9,490), NY (8,760), IL (7,050), OH (5,430), NJ (5,200) <0 1 -3 4-6 7-9 ¦10-30 H 31 -3950 Note: These maps describe projected total change in new annual asthma diagnoses at 2°C and 4°C of global warming relative to baseline (1986-2005). The five states with the highest impacts are outlined in black. See Table 2 for analytic details. April 2023 ------- Climate Change arid Children's Health and Well-Being in the United States Figure 6: Estimated Change in Annual Lost School Days Due to Climate Change Per 100,000 Children (Aged 5-17) (with Population Growth) 2°C of Global Warming Top five states, excess incidence in parentheses: IL (9,430), OH (9,350), D.C. (8,410), IN (7,950), MD (7,360) 4°C of Global Warming Top five states: OH (20,800), IL (19,620), D.C. (19,550), MD (16,410), IN (16,390) <0 1 -2,000 2,001 -4,200 4,201 -6,300 ¦6,301 - 10,000 H 10,001 - 60,000 Note: These maps describe projected change in annual school days lost due to climate change-induced changes i air quality at 2°C and 4°C of global warming relative to baseline (1986-2005). The five states with the highest impacts are outlined in black. See Table 2 for analytic details. April 2023 ------- Climate Change and Children's Health and Well-Being in the United States Figures 7 and 8 present the results of the social vulnerability analysis (see Chapter 2 and Appendix A for methods, data sources, and assumptions). These results are presented separately for PM2.5 (Figure 7) and O3 (Figure 8). The estimated risks for each socially vulnerable group are presented relative to each group's "reference" population, defined as all individuals other than those in the group analyzed. Positive numbers indicate the group is disproportionately affected by the referenced impact. Negative numbers indicate the group is less likely to live in the areas with the highest projected impacts. April 2023 19 ------- Climate Change arid Children's Health and Well-Being in the United States Figure 7: Social Vulnerability Analysis Results for PM2.5 and New Asthma Diagnoses Among Children Limited English Speaking Low Income BIPOC No Health Insurance American Indian or Alaska Native Asian Black or African American Pacific Islander Hispanic or Latino White, non-Hispanic 2°C 23% 4°C Figure 8: Social Vulnerability Analysis Results for 03 and New Asthma Diagnoses Among Children Limited English Speaking Low Income BIPOC No Health Insurance 2°C -23% -12% -23% -25% 4°C -11% -13% -17% -27% American Indian or Alaska Native Asian Black or African American Pacific Islander Hispanic or Latino White, non-Hispanic -46% -41% -44% 19% -57% 12% -45% -37% 31% 23% 21% April 2023 ------- Climate Change and Children's Health and Well-Being in the United States LIMITATIONS Below are several limitations of the analysis. See Fann et al. (2021), Achakulwisut et al. (2019), and Neumann et al. (2021) for additional limitations of the underlying sectoral impact models. 1. There is limited epidemiological literature specifically on children's health effects of air pollution. This analysis relies on standard health functions used by the U.S. EPA for regulatory impacts analyses that are relevant to children. This set of functions focuses on respiratory morbidity effects, and mortality effects are restricted to the postneonatal population. Children are likely to experience additional morbidity and mortality effects that are not quantified by this analysis. 2. Impacts of coarse particulate matter on children's health are omitted from this analysis. The quantitative air quality analyses in this report focus on the impacts of PM2.5 and O3 on children's health. As noted in the main text of the report, additional impacts may be associated with other air pollutants. In particular, there is epidemiological evidence that exposure to coarse particulate matter (PM10-PM2.5) is associated with emergency department visits for asthma.13 Coarse particle exposure among children is expected to increase as a result of both wildfire smoke exposure and increased levels of airborne fugitive dust. As described in Achakulwisut et al., while it is common to assume that impacts attributable to fine and coarse PM fractions are additive because there is technically no overlap in the diameter range of the two PM fractions, in practice, this issue is still up for debate owing to uncertainties in separating the health impacts attributable to fine and coarse PM in epidemiological studies. To avoid the potential for double-counting, we therefore omit quantitative consideration of coarse particulate matter on children's health - in the process we may underestimate the full impact of particulate matter of all size fractions on the health endpoints we assess. 3. The connection between climate change and air quality, particularly particulate matter, remains uncertain and is currently characterized by relatively few lines of evidence. As noted in the above-cited literature used in this report and in Dawson et al. (2014)14, connections between climate change and air quality are complex, particularly with respect to particulate air quality. The modeling work utilized here (Fann et al. 2021) represents an important step forward in modeling finer scale meteorology which affects air quality, making use of state-of- the-art meteorological down-scaling and air quality models. The complexity of the relationship is illustrated by the finding in Fann et al. that some areas of the U.S. could experience improvements in air quality as a result of climate change, while most of the U.S. is expected to experience a decline in air quality. The Fann et al., work has not yet been supported by additional lines of evidence, and as a result may be subject to additional uncertainty. 4. Results of this analysis are available at the county level as the finest spatial scale. The BenMAP analysis was run using county-level baseline incidence and population data, which limits the geographic level to which health impacts associated with pollutant changes can be April 2023 21 ------- Climate Change and Children's Health and Well-Being in the United States specified. As a result, results may underrepresent the spatial precision of the gridded air quality data from underlying climate studies summarized at the beginning of this Appendix. 5. This analysis does not capture fine-scale health effects ofpopula tions tha t may beat grea ter risk of exposure or disproportionate impacts, including BIPOC children, low income individuals, children with housing uncertainty, and children with various comorbidities. This analysis estimates health effects at the county-level using 36-km-squared air quality concentrations and may not capture localized health effects experienced by fenceline and near-road children, who are likely disproportionately vulnerable. 6. The airborne dust component of this analysis is limited to the southwestern region of the U.S. While dust exposures are known to be large in the southwestern U.S., this analysis does not consider health effects from dust in other regions of the U.S., which are likely smallerthan those in the Southwest but nonzero. 7. Respiratory health may degrade for other climate-related reasons. The health effects presented in this chapter are associated with changes in air quality linked with O3 and PM2.5. Respiratory health is likely to worsen among children for other climate-induced reasons, including changes and shift in plant pollen production (see Chapter 5 and Appendix D). DATA SOURCES Table 7: Summary of Data Sources Used in the Air Quality and Children's Health Analysis Data Type Description Data Documentation and Availability Climate modeling See Appendix A for data sources. Air quality modeling Climate Penalty: The Community Multiscale Air Quality (CMAQ) model estimated air quality over the conterminous US for five 11-year periods centered on 2000, 2030, 2050, 2075, and 2095. Southwest Dust: Seasonal mean concentrations of PM2.5 measured at 35 monitoring sites and projected for 20-year periods centered on 2030, 2050, 2070, and 2090. Wildfires: Estimated PM2.5 concentrations over the coterminous US for all years 2006- 2100 using GEOS-Chem chemical transport model. U.S. Environmental Protection Agency. (2020). CMAQ (Version 5.3.2). Available from https://doi. org/10.5281/zenodo.4081737 Climate penalty PM2.5 and 03 air quality results by degree of warming estimated from U.S. Environmental Protection Agency. 2021. "Climate Change and Social Vulnerability in the United States: A Focus on Six Impacts." EPA 430-R-21-003. Achakulwisut, P., Anenberg, S.C., Neumann, J.E., Penn, S.L, Weiss, N., Crimmins, A., Fann, N., Martinich, J., Roman, H. and Mickley, L.J., 2019. Effects of increasing aridity on ambient dust and public health in the US Southwest under climate change. GeoHealth, 3(5), pp. 127-144. httDs://doi.org/10.1029/2019GH000187 April 2023 22 ------- Climate Change and Children's Health and Well-Being in the United States Data Type Description Data Documentation and Availability GEOS-Chem chemical transport model (Version 12.6). Available from http://acmg. seas, harvard.edu/geos/ Emissions inventory estimates Climate Penalty: CMAQ was run using two emission inventory estimates: • The 2011 National Emissions Inventory which estimates the level and distribution of pollutants emitted from all sources • A 2040 emissions inventory which accounts for the implementation of a suite of Federal, state, and local air quality regulations on stationary mobile sources. US Environmental Protection Agency. 2011 National Emissions Inventory, Version 2: Technical Support Document. US Environmental Protection Agency; 2015. Available from https://www.epa.gov/air- emissionsinventories/2011-national- emissions-inventorvnei-data Baseline health effect incidence rates Mortality incidence rates projected from 2000 through 2060 were obtained from BenMAP-CE for one age group (postneonatal infants). Incidence rates for new cases of asthma were obtained from BenMAP-CE for three age groups (0-4, 5-11, and 12-17). Asthma prevalence rates were obtained from BenMAP-CE for two age groups (0-4 and 5-17). Asthma-related ED morbidity incidence rates were obtained from BenMAP-CE for one age group (0-17). Incidence rates for respiratory-related hospital admissions were obtained from BenMAP-CE for two age groups (0-1, 2-17, and 18-24). Prevalence rates of hay fever/rhinitis were obtained from BenMAP-CE for one age group (3-17). Baseline school days lost were obtained from BenMAP-CE for one age group (5-18). U.S. EPA. (2023). Environmental Benefits Mapping and Analysis Program: Community Edition (BenMAP-CE) User Manual and Appendices. Washington, DC. Future population of children See Appendix A for data sources Demographics for social vulnerability analysis See Appendix A for data sources April 2023 23 ------- Climate Change and Children's Health and Well-Being in the United States Wildfire Smoke and Fetal Health Chapter 4 features research on wildfire smoke exposure and risk of preterm births, a maternal health effect that may be exacerbated by climate change. This analysis estimates an additional 7,700 and 13,600 premature births per year at 2°C and 4°C of global warming, respectively, attributable to wildfire annually based on findings from Heft-Neal et al. (2022)15, information on singleton births in 2010 from CDC16, and population-weighted PM2.5 concentrations associated with western wildfire smoke from Neumann et al. (2021).17 Heft-Neal et al. estimated that 3.7% of preterm births in California were attributable to wildfire smoke exposure during the study period (2007-2012). This percentage is applied to the total number of singleton births in the continental U.S. from CDC in 2010 to estimate the number of births attributable to wildfire nationally in the baseline period. Total premature births associated with wildfire in the baseline period were multiplied by a ratio of change in wildfire-attributable PM2.5 concentrations at 2°C and 4°C of global warming to estimate additional premature births associated with wildfire smoke with global warming. Finally, baseline wildfire-attributable premature births were subtracted from projected premature births to estimate the incremental number of premature births presented above and in Chapter 4. DATA SOURCES Table 8: Summary of Data Sources Used in the Wildfire Smoke and Fetal Health Analysis Data Type Description Data Documentation and Availability Number of premature births National count of singleton births in 2010 and preterm singleton birth rate for 2010. Centers for Disease Control and Prevention. 2012. "Births: Final Data for 2010." National Vital Statistics Reports (NVSS), 61(1). Available at: https://www.cdc.gov/nchs/data/nvsr/nvsr61/nvsr61_01. pdf Future wildfire- attributable PM2.5 Change in population-weighted wildfire-attributable PM2.5 concentrations by degree used to scale the number of preterm births attributable to wildfire in the baseline period. Neumann, J.E., Amend, M., Anenberg, S., Kinney, P.L, Sarofim, M., Martinich, J., Lukens, J., Xu, J.W. and Roman, H., 2021. Estimating PM2. 5-related premature mortality and morbidity associated with future wildfire emissions in the western US. Environmental Research Letters, 16(3), p.035019. Wildfire- attributable preterm births Baseline count of premature births estimated from percentage of premature births attributable to wildfire (2007-2012). Heft-Neal, S., Driscoll, A., Yang, W., Shaw, G. and Burke, M., 2022. Associations between wildfire smoke exposure during pregnancy and risk of preterm birth in California. Environmental Research, 203, p.111872. April 2023 24 ------- Climate Change and Children's Health and Well-Being in the United States References 1 Fann, N.L, Nolte, C.G., Sarofim, M.C., Martinich, J. and Nassikas, N.J., 2021. Associations between simulated future changes in climate, air quality, and human health. JAMA Network Open, 4(1), pp.e2032064-e2032064. 2 Achakulwisut, P., Anenberg, S.C., Neumann, J.E., Penn, S.L., Weiss, N., Crimmins, A., Fann, N., Martinich, J., Roman, H. and Mickley, L.J., 2019. Effects of increasing aridity on ambient dust and public health in the US Southwest under climate change. GeoHealth, 3(5), pp. 127-144. 3 Neumann, J.E., Amend, M., Anenberg, S., Kinney, P.L., Sarofim, M., Martinich, J., Lukens, J., Xu, J.W. and Roman, H., 2021. Estimating PM2. 5-related premature mortality and morbidity associated with future wildfire emissions in the western US. Environmental Research Letters, 16(3), p.035019. 4Tetreault, L.F., Doucet, M., Gamache, P., Fournier, M., Brand, A., Kosatsky, T. and Smargiassi, A., 2016. Childhood exposure to ambient air pollutants and the onset of asthma: an administrative cohort study in Quebec. Environmental Health Perspectives, 124(8), pp.1276-1282. 5 Parker, J.D., Akinbami, L.J. and Woodruff, T.J., 2009. Air pollution and childhood respiratory allergies in the United States. Environmental Health Perspectives, 117(1), pp.140-147. 6 Gilliland, F.D., Berhane, K., Rappaport, E.B., Thomas, D.C., Avol, E., Gauderman, W.J., London, S.J., Margolis, H.G., McConnell, R., Islam, K.T. and Peters, J.M., 2001. The effects of ambient air pollution on school absenteeism due to respiratory illnesses. Epidemiology, pp.43-54. 7 Alhanti, B.A., Chang, H.H., Winquist, A., Mulholland, J.A., Darrow, L.A. and Sarnat, S.E., 2016. Ambient air pollution and emergency department visits for asthma: a multi-city assessment of effect modification by age. Journal of Exposure Science & Environmental Epidemiology, 26(2), pp. 180-188. 8 Mar, T.F. and Koenig, J.Q., 2009. Relationship between visits to emergency departments for asthma and ozone exposure in greater Seattle, Washington. Annals of Allergy, Asthma & Immunology, 103(6), pp.474-479. 9 Ostro, B., Roth, L, Malig, B. and Marty, M., 2009. The effects of fine particle components on respiratory hospital admissions in children. Environmental Health Perspectives, 117(3), pp.475-480. 10 Woodruff, T.J., Darrow, L.A. and Parker, J.D., 2008. Air pollution and postneonatal infant mortality in the United States, 1999-2002. Environmental Health Perspectives, 116(1), pp.110-115. 11 U.S. Environmental Protection Agency. 2021. "Climate Change and Social Vulnerability in the United States: A Focus on Six Impacts." EPA430-R-21-003. 12 U.S. Environmental Protection Agency. 2021. "Climate Change and Social Vulnerability in the United States: A Focus on Six Impacts." EPA430-R-21-003. 13 Malig, B.J., Green, S., Basu, R. and Broadwin, R., 2013. Coarse particles and respiratory emergency department visits in California. American journal of epidemiology, 178(1), pp.58-69. 14 Dawson, J. P., B. J. Bloomer, D. A. Winner, and C. P. Weaver, 2014: Understanding the meteorological drivers of U.S. particulate matter concentrations in a changing climate. Bulletin of the American Meteorological Society, 95, 521-532. 15 Heft-Neal, S., Driscoll, A., Yang, W., Shaw, G. and Burke, M., 2022. Associations between wildfire smoke exposure during pregnancy and risk of preterm birth in California. Environmental Research, 203, p. 111872. 16 Centers for Disease Control and Prevention. 2012. "Births: Final Data for 2010." National Vital Statistics Reports (NVSS), 61(1). Available at: https://www.cdc.gov/nchs/data/nvsr/nvsr61/nvsr61_01.pdf 17 Neumann, J.E., Amend, M., Anenberg, S., Kinney, P.L, Sarofim, M., Martinich, J., Lukens, J., Xu, J.W. and Roman, H., 2021. Estimating PM2. 5-related premature mortality and morbidity associated with future wildfire emissions in the western US. Environmental Research Letters, 16(3), p.035019. April 2023 25 ------- |