Sociodemographic disparities and metal concentrations in US drinking water supplies
Year: 2022
Type: conference-abstract
Abstract: Background & Aims: Few drinking water studies have examined metal contamination in the context of environmental justice and fewer have investigated drivers. This study explores associations between sociodemographic disparities and metal concentrations in US public water systems (PWS) and whether these associations are mediated by the availability of best available technologies (BAT). Methods: We assessed arsenic, total chromium, manganese, and selenium concentrations for over 380,000 water samples collected between 1975-2017 from 13,000+ PWS in 15 states from a national drinking water database curated by the Environmental Working Group. We queried PWS characteristics, including BAT use, from the US EPA and averaged sociodemographic data from the 2017 US Census and environmental factors from the USGS and PRISM Climate Group within PWS service area boundaries. We developed a two-step Hurdle model to first predict the probability of metal detection and secondly, among detections, predict chemical concentrations. Results: We found significant relationships between sociodemographic variables and metal concentrations, although directionality varied by metal. Higher percentages of non-English speakers were associated with increasing detection probability for arsenic (0.17, 95% confidence interval [0.06, 0.28]), but decreasing detection probabilities for chromium (-0.18 [-0.28, -0.07]) and manganese (-0.43 [-0.55, -0.32]). Arsenic (binomial: 0.18 [0.08, 0.28]; regression: 0.07 [0.02, 0.12]) and manganese (binomial: 0.33 [0.02, 0.33]; regression: 0.10 [0.04, 0.16]) indicated uniform positive associations with respect to increases in Black population percentages. We did not find evidence that these associations were mediated through BATs. Conclusions: We found some evidence that linguistically isolated communities and communities of color are associated with higher metal concentrations in drinking water. Since we did not find BAT to be a mediator, other local policies, such as historic zoning, may explain these associations. Future work may explore disparities at other spatial levels and consider how financial and operational procedures affect drinking water metal pollution.
Source: ISEE Conference Abstracts
Authors: Mona Dai, Xindi Hu, Brent Coull, Chris Campbell, David Andrews, Olga Naidenko, Elsie Sunderland
Institution: Harvard University
Language: English
Subfield: Sociology and Political Science
Field: Social Sciences
Domain: Social Sciences
Open Access status: diamond
OpenAlex ID: W4317178403