Invisible Threat, Inequitable Impact

In February 2023, the US Environmental Protection Agency (EPA) is likely to propose a rule that would require certain facilities that emit ethylene oxide (EtO) to reduce those emissions (Reed 2022). This flammable, colorless gas is used largely to sterilize medical devices, as well as some plastics and packaged spices. It is also used to make chemicals for manufacturing antifreeze and other products. Despite its wide usage, the EPA, National Toxicology Program, and International Association of Research on Cancer all classify EtO as a carcinogen, a substance that can cause cancer (NCEA 2016, NTP 2021, IARC 2012).

In effect, the very product used to sterilize critical medical equipment also endangers people who live, work, or attend school near these facilities. Chronic exposure through inhaling EtO is associated with the development of cancers of white blood cells, such as non-Hodgkin's lymphoma, as well as breast cancer in women. Children are particularly vulnerable: their bodies experience frequent cell divisions as they grow, and they are susceptible to DNA mutations caused by EtO exposure. Acute inhalation of EtO can also contribute to respiratory issues, headaches, nausea, vomiting, and fatigue (EPA, n.d.a).

The EPA's forthcoming rule will affect only commercial sterilizers, which use EtO to sterilize medical devices, spices, and other products. In 2020, the EPA updated emissions standards for facilities regulated under a source category¹ known as "Miscellaneous Organic Chemical Manufacturing" or MON (EPA 2022a; EPA 2022b). MON facilities can emit EtO (and several other hazardous air pollutants) as a byproduct of producing chemicals used to make solvents, plastics, and pesticides. In December 2022, the EPA affirmed its decision to adopt stronger EtO emissions standards from MON facilities; however, the agency has yet to close emissions loopholes and require fenceline monitoring at these facilities (Earthjustice 2022).

However, commercial sterilizers and MON facilities are just two of several types of facilities that emit EtO—a piece of the puzzle of air toxics that contribute to pollution in communities. Hundreds of facilities in a variety of chemical sectors nationwide emit EtO. For example, two Union Carbide facilities less than five miles apart in Institute and South Charleston, West Virginia, do so (WVDEP, n.d.). But because EtO emissions are regulated under different source categories, stronger emissions requirements for one facility may not apply to the other. This piecemeal, convoluted approach to regulating hazardous air pollutants leaves communities in the dark and continuously exposed to this toxic gas.

The EPA's decision to consider requiring facilities to reduce EtO emissions comes after decades of communities and workers shouldering the burden of harmful exposure. For example, Clean Power Lake County has fought to limit EtO emissions from the Medline Industries commercial sterilizer in Waukegan, Illinois (Clean Power Lake County, n.d.). The community where the facility is located has a high concentration of Latino and low-income residents. Or take Allentown, Pennsylvania: more than 30 residents have sued the B. Braun commercial sterilizer facility, charging that its EtO emissions are responsible for elevated cancer rates (Pradelli and Mettendorf 2022). This community also has a higher proportion of people of color, people with low incomes, and people with limited English language proficiency compared with the county overall. And in Geismar, Louisiana, the BASF chemical plant (regulated as a MON source category) is one of 13 plants producing EtO in the state's so-called "Cancer Alley" (Parker and Russell 2021). According to a ProPublica analysis, BASF is one of the nation's top five corporations for cancer-causing air emissions (Younes et al. 2021).

Community members and workers should not have to carry the burden of ensuring that facilities control cancer-causing air emissions. Indeed, the Clean Air Act requires the EPA to update emissions standards for hazardous air pollutants every eight years. However, the EPA, which was due to update its commercial sterilizer rule in 2014, is now nine years overdue on doing so (OIG 2021). Moreover, it has been seven years since the EPA's Integrated Risk Information System (IRIS)² program updated its risk assessment for EtO inhalation, determining that the chemical is a carcinogen (NCEA 2016). And in 2022, after assessing risks to communities near active commercial sterilizers, the EPA found an elevated cancer risk in 23 communities (EPA 2022c). In these communities, the maximum cancer risk level from EtO emissions exceeds the EPA's threshold of 100 additional cancer cases per 1 million people.³

That said, the concept of "acceptable risk" must be questioned when people's lives are at stake. The EPA's assessments of EtO from commercial sterilizers do not consider exposure to other chemicals or pollutants that may be in a community. Nor do they consider other cumulative impacts, such as poverty, that can increase susceptibility to poor health outcomes. And even communities adjacent to facilities not identified in the EPA's assessment or that fall below the agency's risk threshold may still face elevated cancer risks due to EtO emissions. Thus, the EPA's risk estimates likely underestimate the cancer risk for communities near these facilities. Moreover, while the EPA has identified facilities that emit EtO, who lives near these facilities has been less clear. Nor has the agency completed an equity analysis to identify possible environmental injustices.

To understand who EtO emissions from commercial sterilizers and MON facilities4 most endanger, the Union of Concerned Scientists (UCS) has analyzed emissions and enforcement data for 104 facilities, as well as sociodemographic characteristics of the communities surrounding these facilities. Our analysis, covering US states and Puerto Rico, includes 96 commercial sterilizers that the forthcoming rule may affect, as well as another eight facilities with MON-related EtO emissions (EPA 2022d; EPA 2019). The methodology for this analysis can be found at the end of this document.

What UCS Found

Many people may not know they live near a facility that emits EtO. Unlike chemical plants or refineries, commercial sterilizers may look like warehouses or large office buildings and be situated in or near residential areas. Often, they do not have large smokestacks or appear to be sites of major industrial activity.

UCS found that roughly 14.2 million people nationwide live within five miles of the 104 facilities in our analysis (Figure 1). More than 10,000 schools and childcare centers are within those same areas. California (15 facilities), Texas (nine), Puerto Rico (seven), Florida (six), Minnesota (six), and Georgia (five) are the states and territories with the most commercial sterilizers and EtO-emitting MON facilities. The commercial sterilizers in the top 10 most populous communities overall (more than 300,000 people living within five miles of at least one facility) are in California (six), New Jersey (two), Texas (one), and Utah (one).

Figure 1: More than 14 million people live within five miles of the 104 ethylene oxide-emitting facilities shown here. Many are people of color, people with low incomes, and/or people with limited English language proficiency.

The UCS analysis also identified EtO "sterilizer hotspots"—communities within five miles of two or more commercial sterilizers (in other words, two or more sterilizers that are less than 10 miles apart). We found that 28 percent of commercial sterilizers are in hotspots, potentially exposing people to multiple EtO sources (Table 1). Note that this finding does not include other EtO-emitting facilities that may be in those communities.

Living in a sterilizer hotspot does not mean that a person is absolutely exposed to hazardous levels of EtO, but someone could be exposed to EtO from multiple facilities. Nevertheless, the EPA does not account for these cumulative exposures. Three hotspots with multiple commercial sterilizers are home to facilities identified by the EPA as having an elevated cancer risk from EtO emissions: in Denver, Colorado; Hanover-Jessup, Maryland; and Richmond, Virginia.

Several facilities may be violating the Clean Air Act. According to the UCS analysis, eight of the 46 commercial sterilizers identified by the EPA as contributing to elevated cancer risks, in hotspots, or both have been in noncompliance with the Clean Air Act at least once during the last 12 quarters or been subject to informal or formal EPA enforcement actions in the last five years (Table 1).

Unfortunately, penalties for violating the Clean Air Act too often represent a small cost of doing business. For example, the Sterigenics facility in Charlotte, North Carolina, paid $11,500 in penalties in 2020 for a Clean Air Act violation, a drop in the bucket for a company that reported more than $931.5 million in total net revenues in 2021 (ECHO 2020; SEC 2021).

The UCS analysis found that communities adjacent to commercial sterilizers and EtO-emitting MON facilities are exposed to above-average cancer risks from toxic air pollutants. According to the EPA's AirToxScreen 2018,⁵ Americans face an average cancer risk of 20 additional cases per 1 million people if they breathe air toxics based on 2018 emissions levels for 70 years (EPA, n.d.c). This estimate incorporates most Clean Air Act--regulated air toxics, including EtO.

According to the UCS analysis, the average cancer risk from air toxics in the census tracts where these 104 facilities are located is nearly 60 additional cancer cases per 1 million people—nearly three times the national average (AirToxScreen 2022). Furthermore, EtO emissions contribute to roughly one-third of the total cancer risk in these census tracts, representing a significant portion of cancer risk among 138 air toxics included in AirToxScreen (EPA, n.d.c). The overall air toxics cancer risk in the census tracts where these facilities are located ranges from 10 to 400 additional cancer cases per 1 million.

Fifteen facilities are in census tracts with a total air toxics cancer risk of at least 100 per 1 million, which is the EPA's threshold for acceptable risk. These facilities are in Arizona (two), Florida (one), Georgia (two), Louisiana (one), Missouri (one), Puerto Rico (two), Tennessee (one), and Texas (five).

Exposure to EtO is an urgent environmental justice issue. UCS has estimated the total number of people living within five miles of each facility, including the proportion of people of color, low-income residents, and people with limited English language proficiency. Of the roughly 14.2 million people living within five miles of at least one facility, nearly 8.5 million identify as people of color; 4.8 million identify as people with low incomes, and 1.2 million identify as people with limited English language proficiency (Table 2).

Compared with the counties in which the facilities are located, 57 commercial sterilizers (59 percent) are in communities with a higher proportion of people of color; 41 (43 percent) are located in communities with a higher proportion of people with low incomes; and 24 (25 percent) are in communities with a higher proportion of people with limited English language proficiency. Of the eight facilities with MON-related EtO emissions, five are in communities with a higher proportion of people of color compared with the county averages. For 16 commercial sterilizers and two MON facilities, the proportion of people of color, people with low incomes, and people with limited English language proficiency are all greater than the average for the counties in which the facilities are located.

These differences are even more pronounced when compared with the nation overall. Twenty-seven commercial sterilizers and two MON facilities (28 percent of facilities included in this analysis) are in communities with a higher proportion of people of color, people with low incomes, and people with limited English language proficiency compared with the US average. Indeed, on average, commercial sterilizers and EtO-emitting MON facilities appear to be concentrated in communities and counties with a higher proportion of people of color, people with low incomes, and people with limited English language proficiency compared with the United States overall (Table 2).

Our findings align with those of an EPA Inspector General's 2021 report. It stated that "minorities or [people] part of low-income households" comprise more than half of the people living in census block groups of 14 (of 22) EtO-emitting facilities contributing to elevated cancer risks (OIG 2021).

The UCS analysis substantiates earlier analyses that communities of color are treated as sacrifice zones for toxic pollution (Lerner 2010). Nearly 62 percent (64) of commercial sterilizers and facilities with MON-related EtO emissions are in communities where the proportion of people of color is greater than the US average.

These disproportionate impacts are particularly stark in sterilizer hotspots. Seventeen of the 27 sterilizers (63 percent) in hotspots have a greater proportion of people of color within five miles compared with county averages. On average, the proportion of people of color within five miles of these facilities in hotspots is nearly 10 percent greater than communities within five miles of sterilizers in non-hotspot areas. Multiple commercial sterilizers and other polluting facilities may expose the communities in these hotspots to harmful EtO emissions. Furthermore, more than half (13) of the 23 commercial sterilizers contributing to excess cancer risk from EtO emissions are in communities with a higher proportion of people of color compared with county averages.

Indeed, communities of color face higher cancer risks from air toxics overall. Among the facilities in communities with a higher proportion of people of color than the US average (40 percent), the average air toxics cancer risk is nearly 70 cases per million—which is 17 percent greater than the average for census tracts with commercial sterilizers and EtO-emitting MON facilities overall. And finally, among the facilities in our analysis, 15 (12 commercial sterilizers and 3 MON facilities) have been found to be in noncompliance with the Clean Air Act in at least one of the last 12 quarters. Ten are in communities with a higher proportion of people of color than the US average.

In sum, race is a key indicator of exposure to ethylene oxide from commercial sterilizers and MON facilities. People of color are disproportionately exposed to EtO emissions, face elevated cancer risks, and are in communities that are more likely to have multiple commercial sterilizers and facilities that fail to comply with the Clean Air Act.

Going Deeper: 10 Case Studies

UCS looked deeper into communities in nine states and Puerto Rico that may be disproportionately burdened by EtO emissions from commercial sterilizers and MON facilities. Our analysis considered the results within buffer zones around each facility of one mile (red in the maps), three miles (orange), and five miles (yellow). We retrieved data on facility noncompliance and EtO emissions from the EPA's ECHO database.⁶

California has 15 commercial sterilizers, and more than 4.8 million Californians (roughly 12 percent of the state population) live within five miles of at least one of these facilities (US Census Bureau 2020). Nine of these facilities are located in metropolitan Los Angeles. See more here.

Colorado has four commercial sterilizers. More than 592,000 Coloradans live within five miles of at least one of these facilities (roughly 10 percent of the state population) (US Census Bureau 2020). Two of the metropolitan Denver sterilizers—Terumo BCT in Lakewood and Livanova in Arvada—are a hotspot roughly five miles apart. See more here.

Georgia has five commercial sterilizers. More than 412,000 Georgians (roughly 4 percent of the state population) live within five miles of at least one of these facilities (US Census Bureau 2020). Two facilities in Atlanta—Sterigenics and Sterilization Services of Georgia—are less than 10 miles apart. See more here.

Maryland has four commercial sterilizers. More than 343,000 Marylanders live within five miles of at least one of these facilities (nearly 6 percent of the state population) (US Census Bureau 2020). The two sterilizers in metropolitan Baltimore, in Hanover and Jessup, are roughly three miles apart. See more on here.

All six of Minnesota's commercial sterilizers are in metropolitan Minneapolis-St. Paul (Figure 6). Nearly 842,000 Minnesotans (nearly 15 percent of the state population) live within five miles at least one of these facilities (US Census Bureau 2020). See more here.

New Jersey has three commercial sterilizers. More than 716,000 New Jersey residents (roughly 8 percent of the state population) live within five miles of them (US Census Bureau 2020). In metropolitan New York City, just over the border in New Jersey, two sterilizers are roughly 12 miles apart—EtO Sterilization in Linden and Steris Isomedix in South Plainfield. See more here.

The commonwealth of Puerto Rico bears a disproportionate burden of EtO pollution from commercial sterilizers. Only California and Texas, the two most populous states, have more sterilizers, even though Puerto Rico has roughly one-tenth the population of Texas and 1 percent of its land area. Puerto Rico has seven commercial sterilizers. The EPA has identified four as contributing to elevated cancer risks. In all, the seven sterilizers are within five miles of more than 413,000 Puerto Ricans (roughly 13 percent of the commonwealth's population) and nearly 300 schools and childcare centers (US Census Bureau 2020). Puerto Rico has a very high proportion of Latino residents, with people of color making up 99 percent of the population within five miles of these sterilizers.⁸
See more here.

Tennessee has two commercial sterilizer facilities, Sterilization Services of Tennessee in Memphis and DeRoyal Industries/Royal Sterilization in New Tazewell. Both sterilize medical equipment and nearly 141,000 Tennesseans (roughly 2 percent of the state population) live within five miles of them (US Census Bureau 2020). Organizers from Memphis Community Against Pollution have been pushing federal and local officials to hold the facility accountable and protect public health (MCAP, n.d.). See more here.

Texas has eight commercial sterilizers and one facility with MON-related EtO emissions. More than 1.1 million Texans (roughly 4 percent of the state population) live within five miles of at least one of these facilities (US Census Bureau 2020). Two sterilizers—Sterigenics and Cosmed of Texas—are less than a mile apart in Grand Prairie, between Dallas and Fort Worth. See more here.

Virginia has four commercial sterilizers, two in metropolitan Richmond and two in Virginia Beach. Both areas are sterilizer hotspots. Nearly 726,000 Virginians (more than 8 percent of the state population) live within five miles of at least one of these facilities (US Census Bureau 2020). In Richmond, Sterilization Services of Virginia in Henrico and Central Virginia Health Network/Bon Secours Mercy Health are located less than 10 miles apart. See more here.

Policy Recommendations

More than 14 million people across the country—and in particular, Black, Latino, and other communities of color and non-English-speaking communities—have been breathing unsafe levels of ethylene oxide for far too long. The EPA needs to act urgently on the science, updating its emissions standards for facilities emitting EtO during the sterilization process and in the production of chemicals.

• The EPA must continue to rely on its own assessment of the cancer risk from EtO. The EPA's IRIS assessments use the best available science to determine whether chemicals harm human health. As industry seeks to undermine the IRIS EtO cancer risk value, the EPA needs to continue using its own robust scientific research, relying on it to support EtO regulations (Desikan 2022). Members of the scientific community have spoken out on this issue (UCS 2022). In December 2022, the EPA affirmed that it will use its IRIS values in future EtO rulemakings (Earthjustice 2022).
• In addition to its rule on commercial sterilizers, the EPA must update air toxics rules for all categories of facilities that emit ethylene oxide, including hospital sterilizers, in the manufacture of polymers and resins (neoprene), synthetic organic chemical manufacturing, polyether polyols production, and "chemical manufacturing area sources" (known as CMAS). The EPA must also grant communities' reconsideration petition to reduce cancer risk from MON facilities' EtO emissions, close emissions loopholes, and require fenceline monitoring at these facilities (Earthjustice 2022).
• The EPA must require controls for fugitive EtO emissions. These leaks or other uncontrolled emissions include emissions from warehouses, which are currently unpermitted sources with significant EtO emissions.
• The EPA must ensure that new risk reviews include a margin of safety to protect public health (OIG 2021). The EPA uses so-called residual risk reviews of chemicals to determine the need for stronger health-protective standards. In its EtO review, the EPA must prioritize having an ample margin of safety to ensure protection for children and other people especially at risk of harm from EtO exposure. In accordance with EPA guidance, all risk reviews must also include a publicly available environmental justice analysis and assess cumulative risks and hazards (EPA 2022p; EPA 2023). The EPA must also implement additional controls to alleviate disproportionate harms identified in the analysis.

• The EPA must remove loopholes that allow periodic, uncontrolled emissions from industry processes, which continue to be a major problem.
• The EPA must increase penalties for EtO-emitting facilities that continuously fail to comply with Clean Air Act standards. Corporations should no longer be able to treat penalties as a normal cost of doing business.
• The EPA must implement and enforce EtO monitoring around the fencelines of all commercial sterilizers, MON facilities, and all other source categories of EtO emissions. Communities need and deserve strong protections. This starts with thorough data collection regarding EtO levels in fenceline communities and making this information publicly accessible and available in real time.

• The EPA must reach out to communities affected by EtO emissions, particularly facility workers, people near the highest-risk facilities, and people living with cumulative air pollution exposures in fenceline and hotspot communities. Communities must have opportunities to directly discuss their concerns with the EPA and to inform agency decisionmaking about facilities that directly affect their health. Furthermore, public databases like ECHO must be more accessible and easier to navigate for people who want to learn about the facilities and noncompliance in their communities.
• The EPA must ensure language access by providing appropriate materials and resources on the commercial sterilizer and forthcoming rulemaking processes in both English and other community-appropriate languages. Given the high concentration of facilities in Spanish-speaking communities, materials and interpretation in Spanish is critical.
• The EPA must work toward requiring a complete phaseout of EtO. Less toxic alternatives to EtO are available (Shahbandar 2018). The EPA should continue working to phase out EtO at all commercial sterilizers. Moreover, the Food and Drug Administration, which regulates medical devices and is working to identify alternatives to EtO for sterilization, must collaborate with the EPA (FDA, n.d.).
• The EPA must consider cumulative impacts and protect overburdened communities. Communities can be exposed to EtO from multiple sources, other sources of toxic pollution, and social stressors that can increase the risk of developing cancer and other adverse health outcomes. The EPA and other agencies should use a cumulative impacts framework to better understand the full effects of air toxics on people's health and safety. The agencies should implement safeguards that mitigate these effects and alleviate harm to disproportionately impacted communities.
• The EPA must alleviate chemical hazards to fenceline communities, including by strengthening the Risk Management Program (RMP), which covers some facilities that use and emit EtO, and EPA and Congress should align with the policy goals outlined in the Louisville Charter for Safer Chemicals (Coming Clean 2021).
• Congress should enact legislation, such as the Environmental Justice for All Act, that explicitly advances measures to protect fenceline communities, accounts for cumulative impacts, and requires fenceline monitoring (NRC, n.d.).

No one's ability to breathe clean air should depend on their race, income, or language. Everyone deserves clean air and environmental standards that protect their health and safety. The EPA must listen to the communities exposed to EtO—and to its own science showing the dangers. The problem is clear, the danger is real, and the agency must act now. The EPA urgently needs to adopt rules that strengthen controls of EtO emissions at commercial sterilizers and all other EtO-emitting facilities.

Acknowledgments

This analysis was made possible through the generous support of the Bernard F. and Alva B. Gimbel Foundation, David and Lucile Packard Foundation, Wilburforce Foundation, and UCS members.

At UCS, the author would like to thank Kristie Ellickson for her immense contributions to the data analysis; Genna Reed, Jessica Thomas, Andrés Bachelet, and Jacob Carter for their thoughtful review and input; Matt Heid, Seth Michaels, Seth Shulman, Heather Tuttle, Bryan Wadsworth, Marc Miller, Michelle Rama-Poccia, Cynthia DeRocco, Anthony Eyring, Felipe Castro, Omari Spears, and Nick Iannaco for editing, translation, and production. Finally, the author is grateful to Marvin C. Brown, Robyn Winz, Maya Nye, Beto Lugo-Martinez, Atenas Mena, and Celeste Flores for their invaluable feedback.

The opinions expressed herein do not necessarily reflect those of the organizations that funded the work or the individuals who reviewed it. The Union of Concerned Scientists bear sole responsibility for the report's contents.

Endnotes

1 National Emission Standards for Hazardous Air Pollutants regulate facilities classified under nearly 100 "source categories" (EPA 2022a). The EPA updates emissions standards by source category, rather than by pollutants emitted by each facility.

2 The EPA's IRIS program conducts risk assessments to identify and characterize the health hazards of chemicals. The toxicity values derived by these assessments often serve as the basis for environmental regulations. IRIS assessments represent the gold standard of chemical toxicity reviews at the federal, state, and local levels (IRIS, n.d.).

3 If EtO emissions contribute to cancer risk levels that exceed EPA's threshold for acceptable risk, the agency takes action to require facilities to reduce emissions.

4 At the time of this analysis, the EPA had not yet finalized its decision to require stronger emissions standards for MON facilities that emit EtO. However, we feel this information is important to include because it is not always clear which facilities are regulated under which source categories and communities potentially exposed to EtO have the right to this information.

5 The EPA's AirToxScreen 2018 (the latest available version at the time of this analysis) integrates emissions, ambient concentrations, and exposure estimates for 181 of the 189 Clean Air Act air toxics, plus diesel particulate matter. Cancer risks are included for nearly 140 of these. According to the EPA, it developed AirToxScreen to "help assess which air toxics and emission source types may pose health risks" (EPA, n.d.b).

6 The EPA's Enforcement and Compliance History Online (ECHO) database integrates data for roughly 800,000 facilities nationwide, including facility location, compliance and enforcement records, and pollution emissions (ECHO, n.d.). We provide links to facility webpages when available.

7 While a few thousand pounds of emissions may seem small compared with other pollutants, EtO is a potent, toxic gas. Among the 23 facilities that EPA identified as contributing to excess cancer risks, 2021 reported EtO emissions ranged from 40 to 6,000 pounds (ECHO, n.d.). Some facilities are only permitted 100 to 200 pounds of EtO emissions per year (EPA, n.d.d).

8 The UCS analysis defined "people of color" as those who did not identify as "White Only, non-Hispanic" on the US Census; thus, we included those who identify as "White Hispanic" as "people of color."

References

AirToxScreen (US Environmental Protection Agency, AirToxScreen). 2022. "AirToxScreen Mapping Tool (based on 2018 emissions)." Accessed December 19, 2022. http://epa.maps.arcgis.com/apps/MapSeries/index.html?appid=e5a7e59018c7424eaddcb64b31ba4a41

Buchanan, Christopher. 2020. "Plant Near South Fulton Facing Fines for Cancer-Causing Gas." 11Alive, January 8, 2020. http://www.11alive.com/article/news/health/sterilization-services-of-georgia-ethylene-oxide-fine-epd/85-88f1eb9e-3242-4de3-8360-9e9cc7ce8b62

Clean Air Laredo Coalition. n.d. "Clean Air Laredo Coalition." Accessed December 16, 2022. http://cleanairlaredo.org/

Clean Power Lake County. n.d. " Clean Power Lake County." Accessed December 10, 2022. http://cleanpowerlakecounty.org/

Collier, Kiah, and Maya Miller. 2021. "A Plant that Sterilizes Medical Equipment Spews Cancer-Causing Pollution on Tens of Thousands of Schoolchildren." ProPublica, December 27, 2021. http://www.propublica.org/article/a-plant-that-sterilizes-medical-equipment-spews-cancer-causing-pollution-on-tens-of-thousands-of-schoolchildren

Collier, Kiah, and Maya Miller. 2022. "Texas-Mexico Border Town Approves Air Pollution Monitoring Following ProPublica and Texas Tribune Investigation." ProPublica, August 29, 2022. http://www.propublica.org/article/texas-mexico-border-town-approves-air-pollution-monitoring-following-propublica-and-texas-tribune-investigation

Coming Clean. 2021. "The Louisville Charter for Safer Chemicals." Accessed January 30, 2023. http://comingcleaninc.org/louisville-charter/endorse

Desikan, Anita. 2022. "EPA Rejects Industry Attempt to Downplay Ethylene Oxide Harms." The Equation (blog), March 11, 2022. http://blog.lkgear.com/anita-desikan/epa-rejects-industry-attempt-to-downplay-ethylene-oxide-harms

Earthjustice. 2022. "EPA Affirms Science-Based Cancer Risk Value for Ethylene Oxide." December 16, 2022. http://earthjustice.org/news/press/2022/epa-affirms-science-based-cancer-risk-value-for-ethylene-oxide

ECHO (US Environmental Protection Agency, Enforcement and Compliance History Online). n.d. "Enforcement and Compliance History Online." Accessed November 25, 2022. http://echo.epa.gov

ECHO (US Environmental Protection Agency, Enforcement and Compliance History Online). 2020. "Civil Enforcement Case Report, Case Number: NCMECA200186869." Accessed December 10, 2022. http://echo.epa.gov/enforcement-case-report?activity_id=3602208522

EPA (US Environmental Protection Agency). n.d.a. "Ethylene Oxide." Accessed December 16, 2022. http://www.epa.gov/sites/default/files/2016-09/documents/ethylene-oxide.pdf

---------. n.d.b. "2018 AirToxScreen." Accessed December 20, 2022. http://www.epa.gov/AirToxScreen/2018-airtoxscreen

---------. n.d.c. "AirToxScreen Frequent Questions." Accessed December 20, 2022. http://www.epa.gov/AirToxScreen/airtoxscreen-frequent-questions

---------. n.d.d. "Ethylene Oxide Emissions: Frequent Questions." Accessed January 4, 2023. http://www.epa.gov/il/ethylene-oxide-emissions-frequent-questions

EPA (US Environmental Protection Agency). 2019. "Residual Risk Assessment for the Miscellaneous Organic Chemical Manufacturing Source Category in Support of the 2019 Risk and Technology Review Proposed Rule." Accessed December 13, 2022. http://www.regulations.gov/document/EPA-HQ-OAR-2018-0746-0011

EPA (US Environmental Protection Agency). 2022a. "National Emissions Standards for Hazardous Air Pollutants (NESHAP)." Accessed December 11, 2022. http://www.epa.gov/stationary-sources-air-pollution/national-emission-standards-hazardous-air-pollutants-neshap-8

---------. 2022b. "Miscellaneous Organic Chemical Manufacturing: National Emissions Standards for Hazardous Air Pollutions (NESHAP)." Accessed December 11, 2022. http://www.epa.gov/stationary-sources-air-pollution/miscellaneous-organic-chemical-manufacturing-national-emission

---------. 2022c. "Ethylene Oxide Risk from Commercial Sterilizers." Accessed December 11, 2022. http://www.epa.gov/hazardous-air-pollutants-ethylene-oxide/forms/ethylene-oxide-risk-commercial-sterilizers

---------. 2022d. "Ethylene Oxide Commercial Sterilization Facilities." Accessed December 11, 2022. http://www.epa.gov/hazardous-air-pollutants-ethylene-oxide/ethylene-oxide-commercial-sterilization-facilities

---------. 2022e. "Lakewood, Colorado (Terumo BCT Sterilization Service, Inc.)." Accessed December 11, 2022. http://www.epa.gov/hazardous-air-pollutants-ethylene-oxide/forms/lakewood-colorado-terumo-bct-sterilization-service

---------. 2022f. "Hanover, Maryland (Elite Spice, Inc.)." Accessed December 11, 2022. http://www.epa.gov/hazardous-air-pollutants-ethylene-oxide/forms/hanover-maryland-elite-spice-inc

---------. 2022g. "Jessup, Maryland (Elite Spice, Inc.)." Accessed December 11, 2022. http://www.epa.gov/hazardous-air-pollutants-ethylene-oxide/forms/jessup-maryland-elite-spice-inc

---------. 2022h. "Linden, New Jersey (ETO Sterilization-Plant #2)." Accessed December 11, 2022. http://www.epa.gov/hazardous-air-pollutants-ethylene-oxide/forms/linden-new-jersey-eto-sterilization-plant-2

---------. 2022i. "Añasco, Puerto Rico (Edwards Lifesciences)." Accessed December 11, 2022. http://www.epa.gov/hazardous-air-pollutants-ethylene-oxide/forms/anasco-puerto-rico-edwards-lifesciences

---------. 2022j. "Fajardo, Puerto Rico (Customed, Inc.)." Accessed December 11, 2022. http://www.epa.gov/hazardous-air-pollutants-ethylene-oxide/forms/fajardo-puerto-rico-customed-inc

---------. 2022k. "Salinas, Puerto Rico (Steri-Tech, Inc.)." Accessed December 11, 2022. http://www.epa.gov/hazardous-air-pollutants-ethylene-oxide/forms/salinas-puerto-rico-steri-tech-inc

---------. 2022l. "Villalba, Puerto Rico (Medtronic PR Operation CO)." Accessed December 11, 2022. http://www.epa.gov/hazardous-air-pollutants-ethylene-oxide/forms/villalba-puerto-rico-medtronic-pr-operation-co

---------. 2022m. "Memphis, Tennessee (Sterilization Services of Tennessee)." Accessed December 11, 2022. http://www.epa.gov/hazardous-air-pollutants-ethylene-oxide/forms/memphis-tennessee-sterilization-services-tennessee

---------. 2022n. "New Tazewell, Tennessee (DeRoyal Industries)." Accessed December 11, 2022. http://www.epa.gov/hazardous-air-pollutants-ethylene-oxide/forms/new-tazewell-tennessee-deroyal-industries

---------. 2022o. "Henrico, Virginia (Sterilization Services of Virginia)." Accessed December 11, 2022. http://www.epa.gov/hazardous-air-pollutants-ethylene-oxide/forms/henrico-virginia-sterilization-services-virginia

---------. 2022p. "EPA Legal Tools to Advance Environmental Justice." May 2022. http://www.epa.gov/ogc/epa-legal-tools-advance-environmental-justice

EPA (US Environmental Protection Agency). 2023. "EPA Legal Tools to Advance Environmental Justice: Cumulative Impacts Addendum." January 2023. http://www.epa.gov/system/files/documents/2022-12/bh508-Cumulative%20Impacts%20
Addendum%20Final%202022-11-28.pdf

FDA (US Food and Drug Administration). n.d. "Sterilization for Medical Devices." Accessed December 16, 2022. http://www.fda.gov/medical-devices/general-hospital-devices-and-supplies/sterilization-medical-devices

IARC (International Agency for Research on Cancer). 2012. Chemical Agents and Related Occupations. Lyon, France: World Health Agency, International Agency for Research on Cancer. http://monographs.iarc.who.int/wp-content/uploads/2018/06/mono100F.pdf

IRIS (US Environmental Protection Agency, Integrated Risk Information System). n.d. "Basic Information about the Integrated Risk Information System." Accessed December 20, 2022. http://www.epa.gov/iris/basic-information-about-integrated-risk-information-system

Lerner, Stephen. 2010. Sacrifice Zones: The Front Lines of Toxic Chemical Exposure in the United States. Cambridge, MA: MIT Press.

MCAP (Memphis Community Against Pollution). n.d. "We Deserve to Breathe Clean Air! Cancer-Causing Toxin Ethylene Oxide (EtO) Released Into South Memphis Air." Accessed January 6, 2023. http://www.memphiscap.org/eto

NCEA (National Center for Environmental Assessment). 2016. Evaluation of the Inhalation Carcinogenicity of Ethylene Oxide. Washington, DC: Office of Research and Development, US Environmental Protection Agency. http://cfpub.epa.gov/ncea/iris/iris_documents/documents/toxreviews/1025tr.pdf

NRC (US House of Representatives, Natural Resources Committee). n.d. "Environmental Justice: The Environmental Justice for All Act." Accessed December 16, 2022. http://naturalresources.house.gov/environmental-justice

NTP (US Department of Health and Human Services, National Toxicology Program). 2021. "Report on Carcinogens, Fifteenth Edition: Ethylene Oxide." Accessed December 10, 2022. http://ntp.niehs.nih.gov/ntp/roc/content/profiles/ethyleneoxide.pdf

OIG (Office of Inspector General). 2021. EPA Should Conduct New Residual Risk and Technology Reviews for Chloroprene- and Ethylene Oxide-Emitting Source Categories to Protect Human Health. Washington, DC: US Environmental Protection Agency. http://www.epa.gov/sites/default/files/2021-05/documents/_epaoig_20210506-21-p-0129.pdf

Parker, Halle, and Gordon Russell. 2021. "Louisiana Chemical Corridor Is the Country's Largest Hot Spot for Toxic Air, Cancer Risk." Nola.com, November 5, 2021. http://www.nola.com/news/environment/louisiana-chemical-corridor-is-the-countrys-largest-hot-spot-for-toxic-air-cancer-risk/article_852d98d0-3d81-11ec-a61f-0bf8e82339a8.html

Pradelli, Chad, and Cheryl Mettendorf. 2022. "Federal Agency Fails to Notify Allentown Residents about Their Increased Cancer Risk." 6ABC Action News, February 11, 2022. http://6abc.com/bbraun-allentown-cancer-lawsuit-ethylene-oxide-eto-environmental-protection-agency/11552914

Reed, Genna. 2022. "What is Ethylene Oxide? Answers to Your Questions about the Cancer-Causing Chemical." The Equation (blog),October 6. http://blog.lkgear.com/genna-reed/what-is-ethylene-oxide-answers-to-your-questions-about-the-cancer-causing-chemical

SEC (US Securities and Exchange Commission). 2021. "Form 10-K Annual Report: Sotera Health Company." Accessed December 10, 2022. http://www.sec.gov/ix?doc=/Archives/edgar/data/1822479/000182247922000014/shc-20211231.htm

Shahbandar, Lena. 2018. "Alternatives to Ethylene Oxide." Stop Sterigenics, November 11, 2018. http://www.stopsterigenics.com/post/alternatives-to-ethylene-oxide

Stop Sterigenics—Georgia. n.d. Facebook. Accessed December 13, 2022. http://www.facebook.com/StopSterigenicsGA

US Census Bureau. 2020. "2016--2020 American Community Survey 5-Year Estimates." Accessed October 5, 2022. http://www.census.gov/programs-surveys/acs/technical-documentation/table-and-geography-changes/2020/5-year.html

UCS (Union of Concerned Scientists). 2022. "Experts Tell EPA: Follow the Science, Protect Communities from Ethylene Oxide." March 15, 2022. http://jhe6.lkgear.com/about/news/experts-tell-epa-follow-science-protect-communities-ethylene-oxide

WVDEP (West Virginia Department of Environmental Protection). n.d. "Ethylene Oxide (EtO)." Accessed December 13, 2022. http://dep.wv.gov/key-issues/Pages/EtO.aspx

Younes, Lylla, Ava Kofman, Al Shaw, and Lisa Song. 2021. "Poison in the Air." ProPublica, November 2, 2021. http://www.propublica.org/article/toxmap-poison-in-the-air

Methodology

The analysis included 104 Commercial Sterilizer and Miscellaneous Organic Chemical Manufacturing (MON) facilities total.

Data for commercial sterilizers were gathered from the US Environmental Protection Agency's (EPA) webpage of Ethylene Oxide Commercial Sterilization Facilities (EPA 2022a). Twenty-three of these commercial sterilizers were also identified as contributing to elevated cancer risk by EPA risk assessments in 2022 (EPA 2022b). Information about these facilities, including lifetime cancer risk estimates attributable to ethylene oxide emissions, were provided on individual webpages for each facility, accessed on November 25, 2022. EPA's threshold for "elevated" cancer risk is that which exceeds 100 additional cancer cases per 1 million people, assuming 24 hours of exposure over a 70-year lifetime (EPA 2022b). These risk estimates were based solely on ethylene oxide emissions from each facility and do not consider other exposures or stressors that may impact one's risk of developing cancer. Data for MON facilities were drawn from Appendix 10 of the EPA's 2019 "Residual Risk Assessment for the Miscellaneous Organic Chemical Manufacturing Source Category" (EPA 2019).

Location coordinates, compliance, enforcement, and emissions data for each facility were gathered from the EPA's Enforcement and Compliance History Online (ECHO) database, accessed on November 25, 2022 (ECHO, n.d.). It is important to note that the ECHO database is inconsistently updated and is missing data. We were unable to locate 19 facilities in ECHO, and 72 facility pages (nearly 70 percent) did not include ethylene oxide releases and transfers data for 2021.

US population data, as well as data on people of color, people with low income, and people with limited English language proficiency were gathered from the US Census Bureau, 2016-2020 American Community Survey, 5-Year Estimates. "People of color" was defined as the population that did not identify as "White Alone, Not Hispanic or Latino" (the variable value was subtracted from the total population). "People with low income" was defined as the population living below 200 percent of the federal poverty level. "People with limited English language proficiency" was defined as the population that identifies as speaking English "not well" or "not at all."

The number of "schools" includes both public and private schools. Data on public and private schools and childcare centers were gathered from the Homeland Infrastructure Foundation-Level Data (HIFLD) Geoplatform (DHS 2022).

AirToxScreen data were gathered from EPA's AirToxScreen 2018, which was the latest version available at the time of the analysis (EPA 2022c). The EPA's AirToxScreen database integrates emissions, ambient concentrations, and exposure estimates for 181 of the 189 Clean Air Act air toxics, plus diesel particulate matter. Cancer risks are included for nearly 140 of these. According to the EPA, AirToxScreen "helps assess which air toxics and emission source types may pose health risks" (EPA 2022c).

The facilities were mapped as points in ArcGIS using geographic coordinates. For each point, the one-, three-, and five-mile buffer circles were added. While the distance that gases can travel from a facility will vary greatly based on weather, geography, and other factors—and ethylene oxide emissions will generally most impact those living on the fenceline or working in these facilities—we selected the five-mile buffer around each facility as the study area because the EPA's 2022 risk assessment of 23 commercial sterilizers demonstrated elevated cancer risks more than four miles away from some facilities.

Any facility with a buffer area that overlapped with the buffer of another facility was deemed a "hotspot." Hotspots are areas where a facility is 10 or fewer miles from at least one other facility. Communities between facilities in a hotspot may be exposed to higher levels of ethylene oxide. In this analysis, all hotspots included commercial sterilizers only.

For each facility, we estimated the population; number of schools (public and private) and childcare centers; and number of people of color, people with low income, and people with limited English language proficiency within the five-mile buffers. These estimates were determined in R-4.2.2, where each facility was geolocated in space and a 5-mile radius was drawn around the point using the sf package. A fraction of the area of the circles within each geography (block group, tract, county, and state) was calculated using the tigris package. Next, the census variable and total population counts or estimates for each geography were extracted using the tidycensus package. The final "buffer population" for the census variables was estimated by multiplying the census geography population by the fraction of the facility buffer area compared to the area of the total census geography. If the buffer overlapped more than one census geography, these population estimates were summed to reflect the whole buffer.

To understand whether there were local and regional disparities in who is most impacted by ethylene oxide pollution, we compared the sociodemographic data within five miles of each facility to county averages. To calculate county averages, the population estimates for total geographies were extracted as described above. If a buffer overlapped more than one county, a mean was calculated for the population estimates for each overlapped county.

For each facility, the proportion of people of color, people with low income, and people with limited English language proficiency within the five-mile buffer were compared with the county averages. The proportion within the five-mile buffer was only deemed greater than the county average if there was a 1 percent or greater difference.

To determine the estimated air toxics cancer risk attributable to ethylene oxide, we divided the "Ethylene Oxide Risk" by the "Total Risk" for each census tract where the facilities are located. Note that AirToxScreen assumes that "exposures to multiple carcinogens can be added together to estimate risks" (EPA 2022d). According to the EPA, "This approach has drawbacks: Effects from multiple chemicals may be greater or less than additive, and statistical limitations exist. But this straightforward calculation is widely used to estimate cumulative risks" (EPA 2022d). While this approach may not fully capture the synergistic interactions between chemicals, it provides an estimate of ethylene oxide's contribution to the total cancer risk from nearly 140 air toxics. In the StoryMap, we included the estimated air toxics cancer risk attributable to ethylene oxide if it exceeded 10 percent. If it is not provided, it means that ethylene oxide is estimated to contribute to less than 10 percent of the overall air toxics cancer risk in the census tract where the facility is located. Finally, while the EPA has not yet released a Technical Support Document for AirToxScreen 2018, the map indicates that ethylene oxide risks have decreased since 2018 due to emissions controls; however, at the time of this analysis, AirToxScreen 2018 was the latest version available.

DHS (US Department of Homeland Security). 2022. "Homeland Infrastructure Foundation-Level Data (HIFLD). http://hifld-geoplatform.opendata.arcgis.com/

ECHO (US Environmental Protection Agency, Enforcement and Compliance History Online). n.d. "Enforcement and Compliance History Online." Accessed November 25, 2022. http://echo.epa.gov

EPA (US Environmental Protection Agency). 2019. "Residual Risk Assessment for the Miscellaneous Organic Chemical Manufacturing Source Category in Support of the 2019 Risk and Technology Review Proposed Rule." Accessed December 13, 2022. http://www.regulations.gov/document/EPA-HQ-OAR-2018-0746-0011

EPA (US Environmental Protection Agency). 2022a. "Ethylene Oxide Commercial Sterilization Facilities." Accessed December 11, 2022. http://www.epa.gov/hazardous-air-pollutants-ethylene-oxide/ethylene-oxide-commercial-sterilization-facilities

---------. 2022b. "Ethylene Oxide Risk from Commercial Sterilizers." Accessed December 11, 2022. http://www.epa.gov/hazardous-air-pollutants-ethylene-oxide/forms/ethylene-oxide-risk-commercial-sterilizers

---------. 2022c. "2018 AirToxScreen." Accessed December 20, 2022. http://www.epa.gov/AirToxScreen/2018-airtoxscreen

---------. 2022d. Technical Support Document, EPA's Air Toxics Screening Assessment, 2017 AirToxScreen TSD. Accessed January 16, 2023. http://www.epa.gov/system/files/documents/2022-03/airtoxscreen_2017tsd.pdf