Invited Perspective: The Importance of Community Involvement in Interventions to Reduce Arsenic Exposure and Improve Health Outcomes in Indigenous Communities

In this issue of Environmental Health Perspectives, George et al.¹ describe the efficacy of a multilevel intervention at reducing arsenic exposure through the provision of safe drinking water to communities impacted by contaminated water supplies. They used a community-led approach that included mobile health (mHealth) reminders, videos, and home visits to encourage usage of a point-of-use filter for all water used for drinking and cooking inside the home, as well as to remind households when to change their filter cartridges.

The need to reduce arsenic exposures is supported by work from the Strong Heart Study (SHS), a partnership between research investigators, the National Institutes of Health, and partner communities to identify risk factors for cardiovascular disease in American Indian communities. The SHS has demonstrated significant associations between arsenic exposure and chronic illnesses, such as cardiovascular disease, kidney disease, lung function, neurological disorders, diabetes, cancer, and all-cause mortality in Indigenous communities.^2–6 Many of these health outcomes occur at a disproportionately high prevalence in US Indigenous communities. More than half the Indigenous population of the country lives in 15 Western states, where the Government Accounting Office has documented $>\mathrm{160,000}$ abandoned mines where gold, uranium, vanadium, copper, and lead were once extracted.⁹ The waste associated with many of these mines is high in arsenic and associated metals and is reported by the US Environmental Protection Agency (EPA) to contaminate $>40\%$ of surface waters in the Western United States.⁷ Moreover, Indigenous communities in the US Southwest are more likely to use water contaminated with arsenic at levels exceeding safe drinking water standards due to a) the presence of both natural and anthropogenic sources of arsenic⁸ and b) a lack of infrastructure and extreme rurality, leading to reliance on unregulated sources of drinking water, such as private wells.⁹

George et al.¹ conducted a cluster randomized controlled trial in Indigenous communities involved in the Strong Heart Water Study (SHWS), a community-led intervention to reduce arsenic exposure among private well users in partnership with communities participating in the SHS. Among the SHWS participants, 29% of private wells exceeded the US EPA maximum contaminant level of $10\mathrm{\mu }\mathrm{g}/\mathrm{mL}$ for arsenic. Installation of point-of-use arsenic filters resulted in close to a 50% reduction in the geometric mean concentrations of urinary arsenic from baseline to final follow-up, with those participants relying exclusively on treated water for cooking or drinking having significantly lower urinary arsenic than participants not reporting exclusive use. This suggests that potential health benefits of intervention for these communities, where exposures are frequently chronic and lifelong, could be significant. Notably, George et al. considered only the more toxic inorganic forms of arsenic (including the metabolites monomethylarsonic acid and dimethylarsinic acid) in their total arsenic calculations and excluded less toxic organic forms (e.g., arsenobetaine and other arsenic cations) such as those predominately found in fish and other food sources.¹⁰

Moving forward, it will be important to understand specific sources of exposure within communities to optimize strategies for exposure reduction. In a recent study,¹¹ Beene et al. used a mass balance approach to estimate the extent to which different sources of exposure accounted for the variance between arsenic intake and excretion across previously studied populations. The authors found that drinking water intake alone was insufficient to achieve mass balance. However, in rural Bangladesh, accounting for intake from multiple wells, as well as from rice, improved alignment between intake and excretion, especially at low exposure; for the Navajo Nation, home dust was a major component of intake; and in northern Chile, although speciation of urinary arsenic pointed to the importance of dietary sources, imprecise measurement of drinking water sources contributed to exposure misclassification.

Therefore, multiple intervention strategies may ultimately be necessary to meaningfully reduce exposures across diverse communities owing to potential differences in exposure sources (water, dust, food), as well as the prevalence of water hauling and homes without indoor drinking water taps in Indigenous communities, which might not benefit from installing a local filter. Mechanism-based interventions to reduce biological impacts of arsenic exposures are one strategy to augment exposure reduction approaches and reduce health risk. Our ongoing clinical trial, Thinking Zinc (NCT03908736), investigates the potential for dietary zinc supplementation to mitigate arsenic and uranium toxicity. This trial is based on laboratory studies demonstrating that both arsenic and uranium can displace zinc in regulatory control proteins and block their function in multiple pathways (including the repair of exposure-induced DNA damage) and that zinc supplementation could reverse these effects.¹² Extensive community involvement from the beginning in both planning and implementation has been essential in establishing this clinical trial within impacted Navajo Nation communities. Through intensive discussions with community members, knowledge keepers, and cultural experts, investigators developed a forum to work together on developing a clinically valid and defensible intervention that was respectful of culture and was of interest to the community. This creation of community ownership of the protocol generated strong compliance and retention of participants over time. Community outreach and education events provided information about which local traditional foods are zinc rich. Frequently, at these events, community liaisons on the research team served traditionally prepared blue corn mush, a nutritious and zinc-rich food common in many Indigenous communities in this region.

Importantly, the study by George et al. points to the success of both overall compliance with methodology and the use of water filters that was accomplished with community engagement and early participation in both design and implementation, providing methods useful to decision-makers when considering other intervention strategies that may be needed. Moreover, the significant reductions achieved by this intervention will be useful in supporting policy changes that could lead to community-level strategies that include larger-scale filtration systems at distribution points. The success of initiating these community strategies before finalizing plans is a lesson applicable to implementing specific interventions for any exposure source. It is notable that, in this study, home visits did not add to the effectiveness of the intervention, demonstrating that community-led follow-up can be efficient and non–resource heavy when delivered in a way that fits the needs and expectations of the community.

Overall, the complexities in understanding arsenic exposures point to the need for future work to a) determine whether the 50% reduction in urinary arsenic achieved through this intervention is sufficient to reduce the chronic disease risks, b) understand the dose–response relationships, and c) consider future interventions that reduce exposures or interrupt the biological mechanistic pathways. Early and respectful community participation in the co-creation of interventions can result in successful strategies for risk reduction and trial implementation across a wide range of physical and biological interventions—even in communities where trust has previously been compromised—by increasing respect, trust, and ownership in the scientific process. This is especially critical when working with communities to develop interventions that may require a multipronged approach.

Refers to https://doi.org/10.1289/EHP12548

Conclusions and opinions are those of the individual authors and do not necessarily reflect the policies or views of EHP Publishing or the National Institute of Environmental Health Sciences.