Low-income and minority communities often face disproportionately high pollutant exposures. The lead crisis in Flint, Michigan, has sparked concern about broader socioeconomic disparities in exposures to drinking water contaminants. Nitrate is commonly found in drinking water, especially in agricultural regions, and epidemiological evidence suggests elevated risk of cancer and birth defects at levels below U.S. EPA’s drinking water standard (10 mg/L NO3-N). However, there have been no nationwide assessments of socioeconomic disparities in exposures to nitrate or other contaminants in U.S. drinking water. The goals of this study are to identify determinants of nitrate concentrations in U.S. community water systems (CWSs) and to evaluate disparities related to wealth or race/ethnicity. We compiled nitrate data from 39,466 U.S. CWSs for 2010–2014. We used EPA’s Safe Drinking Water Information System (SDWIS) to compile CWS characteristics and linked this information with both city- and county-level demographic data gathered from the U.S. Census Bureau. After applying multiple imputation methods to address censored nitrate concentration data, we conducted mixed-effects multivariable regression analyses at national and regional scales. 5.6 million Americans are served by a CWS that had an average nitrate concentration ≥ 5 mg/L NO3-N between 2010 and 2014. Extent of agricultural land use and reliance on groundwater sources were significantly associated with nitrate. The percent of Hispanic residents served by each system was significantly associated with nitrate even after accounting for county-level cropland and livestock production, and CWSs in the top quartile of percent Hispanic residents exceeded 5 mg/L nearly three times as often as CWSs serving the lowest quartile. By contrast, the percent of residents living in poverty and percent African American residents were both inversely associated with nitrate. Epidemiological evidence for health effects associated with drinking water above 5 mg/L NO3-N raises concerns about increased risk for the 5.6 million Americans served by public water supplies with average nitrate concentrations above this level. The associations we observed between nitrate concentrations and proportions of Hispanic residents support the need for improved efforts to assist vulnerable communities in addressing contamination and protecting source waters. Future studies can extend our methods to evaluate disparities in exposures to other contaminants and links to health effects.
Hundreds of different disinfection byproducts form in drinking water following necessary treatment with chlorine and other disinfectants, and many of those byproducts can damage DNA and increase the risk of cancer. This study offers the first side-by-side comparison of cancer risk assessments based on toxicological and epidemiological studies of disinfection byproducts using a comprehensive contaminant occurrence dataset for haloacetic acids and trihalomethanes, two groups of disinfection byproducts that are regulated in drinking water. We also provide the first analysis of a new occurrence dataset for unregulated haloacetic acids that became available from the latest, fourth round of the U.S. EPA-mandated unregulated contaminant monitoring program (UCMR4). A toxicological assessment indicated that haloacetic acids, and in particular brominated haloacetic acids, are more carcinogenic and are associated with a greater number of attributable cancer cases than trihalomethanes. Based on the toxicological analysis, cumulative lifetime cancer risk due to exposure to trihalomethanes and haloacetic acids for community water systems monitored under UCMR4, estimated with standard default parameters for body weight and water intake, corresponds to 7.0 × 10−5 (3.5 × 10−5–1.3 × 10−4). The same analysis conducted with age sensitivity factors to account for elevated risk in infants and children yielded a cumulative risk estimate of 2.9 × 10−4 (1.7 × 10−4–6.2 × 10−4). Epidemiological data suggest that lifetime cancer risk from disinfection byproducts for the U.S. population served by community water systems is approximately 3.0 × 10−3 (2.1 × 10−4–5.7 × 10−3), or a lifetime cancer risk of three cases per thousand people. Overall, this analysis highlights the value of using human data in health risk assessments to the greatest extent possible.
Research studies analyzing the geospatial distribution of air pollution and other types of environmental contamination documented the persistence of environmental health disparities between communities. Due to the shortage of publicly available data, only limited research has been published on the geospatial distribution of drinking water pollution. Here we present a framework for the joint consideration of community-level drinking water data and demographic data. Our analysis builds on a comprehensive data set of drinking water contaminant occurrence for the United States for 2014–2019 and the American Community Survey 5-year estimates (2015–2019) from the U.S. Census Bureau. Focusing on the U.S. states of California and Texas for which geospatial data on community water system service boundaries are publicly available, we examine cumulative cancer risk for water served by community water systems of different sizes relative to demographic characteristics for the populations served by these water systems. In both California and Texas, greater cumulative cancer risk was observed for water systems serving communities with a higher percentage of Hispanic/Latino and Black/African American community members. This investigation demonstrates that it is both practical and essential to incorporate and expand the drinking water data metrics in the analysis of environmental pollution and environmental health. The framework presented here can support the development of public policies to advance environmental health justice priorities on state and federal levels in the U.S.