Exploring the use of the RISK21 approach for Indigenous community-based human health risk assessments: two case studies
Publication: FACETS
24 February 2025
Abstract
Indigenous peoples in North America are disproportionately exposed to environmental contaminants and may face elevated health risks related to unique socio-cultural ties to the land. Conventional human health risk assessment (HHRA) methods do not account for these unique contexts. Regulators (i.e., Health Canada, US EPA) have called for the development of more ethical and efficient HHRA approaches, but to our knowledge no such approaches have been designed in consideration of Indigenous community contexts. RISK21 is a new HHRA approach gaining regulatory attention. We present two case studies piloting RISK21’s use in collaboration with three unique communities (Cold Lake First Nations, Chipewyan Prairie First Nation, Apsáalooke/Crow Nation). Our objectives are first, reflect upon the benefits and challenges of using RISK21 in these contexts; second, compare RISK21-based to conventional assessments; and third, ideate adaptations and improvements to the approach. The RISK21-based analyses had similar descriptive results to the original conventional assessments, including when using less information. We found RISK21 useful for rapid chemical assessment and visually representing data from multiple sources. We recommend areas where RISK21 (and other next-generation HHRA approaches) might be improved for Indigenous community contexts, including increasing the community relevance of communication tools and incorporating holistic and non-conventional information.
Introduction
Environmental pollution is a pervasive issue for Indigenous communities across Canada and the US. For example, approximately 4500 contaminated sites, largely driven by external industrial and mining activities, are located on First Nations reserve land in Canada (Government of Canada 2021). In a recent report from the federally led Northern Contaminants Program, exposures to several persistent organic pollutants and heavy metals were up to eleven-fold higher among Inuit men and women from northern Canada compared to the general Canadian population in the South (Government of Canada 2017). In the US, approximately 34% of Superfund sites are deemed “Native American Interest” sites (i.e., they affect Native American peoples) (Chong and Basu 2023), with an estimated 146 605 Native American people living within one mile of a Superfund site (U.S. EPA 2020). Many Indigenous communities face unique risks and impacts related to environmental pollution, including from contaminated sites, compared to the general population. This is not only due to disproportionate physical exposure, but also through impacts on culture, spirituality, language, the economy, and many other facets of life to which the land is a central element (Harris 2000; O'Neill 2000; Van Oostdam et al. 2005; Burger and Gochfeld 2011; Eggers et al. 2018; Fernández-Llamazares et al. 2020). The contamination of traditional food systems has been particularly impactful, through chemical exposures and cultural losses associated with an inability to conduct traditional harvesting, fishing, hunting, and agricultural activities, and resulting in a transition to nutrient-poor market foods (Damman et al. 2008; Chan et al. 2021). It has also contributed to a breakdown of intergenerational knowledge transfer related to procuring, processing, and preparing traditional foods (Robin et al. 2021).
Human health risk assessment (HHRA) is a method used to estimate the “nature and probability of adverse health effects in humans who may be exposed to chemicals in contaminated environmental media” (U.S. EPA 2022). HHRAs are required by many governments globally prior to the initiation of industrial or development projects, and to make decisions about remediation and risk management at existing contaminated sites, including those that impact Indigenous communities (Health Canada 2004). However, such generic and institutionalized HHRA methods have been described as inadequate for addressing the unique risk assessment needs of the Indigenous communities in which they are used. This includes, for example, that conventional HHRA methods may not adapt to community-specific needs or include Indigenous perspectives (OECD 2020), are considered costly and time consuming (Pastoor et al. 2014), and often produce risk information that is not locally useful (Harris 2000; Harper et al. 2012). Further, there is no well-established and agreed upon method for assessing combined risks of multiple chemical and non-chemical (socio-cultural) stressors (i.e., cumulative risks) (Fox et al. 2017; Tong and Zhang 2023). A recent systematic review concluded that while evidence continues to emerge on the harms associated with combined chemical and psychosocial stressors, there has been minimal progress on the integration of cumulative risk assessment into environmental management or public health policy decision-making (Tong and Zhang 2023).
Within the past few decades, communities and governments in Canada and the US have called for a need to change the way that HHRAs are conducted, especially in regulatory contexts. There is increasing emphasis on the adoption of alternatives that are cheaper, more efficient, less resource intensive, and fit for purpose. This paradigm shift is marked by the U.S. National Research Council’s “Science and Decision Report” of 2009 and the U.S. National Academy’s book, “Exposure Science in the 21st Century” of 2012. Aligned with this shift, Canada’s 2016 Chemicals Management Plan (CMP) Science Committee recommended that governing bodies, including Health Canada (HC) and Environment and Climate Change Canada (ECCC), must develop a roadmap for integrating alternative risk assessment approaches into regulatory decisions (CMP Science Committee 2016). Most recently, in February of 2022, amendments to the Canadian Environmental Protection Act (CEPA) 1999 were proposed in Bill S-5, “Strengthening Environmental Protection for a Healthier Canada” Act. Bill S-5 specifically references honoring the United Nations Declaration on the Rights of Indigenous Peoples (UNDRIP), calls for improved risk communication, and ultimately developing new approaches to risk assessment for environmental contaminants. These amendments are particularly relevant to the many First Nations, Inuit, and Métis communities who are affected by environmental pollution and are subject to risk assessment results (United Nations 2007; Government of Canada 2022).
A variety of alternative tools and approaches to toxicity testing, exposure assessment, and risk assessment are currently under development. The desire is that modernized approaches will underpin the next generation of standards on which regulatory decisions will be based. One such approach is the Health and Environmental Sciences Institute’s (HESI) RISK21, a HHRA approach that aims to optimize the use of existing information and testing resources (Embry et al. 2014). RISK21 was developed in response to recommended improvements to HHRA approaches, including by the National Academy of Sciences, to more efficiently, ethically, and transparently reach a risk or safety determination compared to conventional methods (Embry et al. 2014; risk21.org). RISK21’s components include a conceptual framework for risk assessment and a visual exposure-toxicity matrix, which allows for exposure and hazard information to be evaluated and compared effectively (Embry et al. 2014). RISK21 also includes a framework for assessing cumulative risks and has a web-based risk communication tool (Moretto et al. 2017). RISK21 has been noted as a strategy for the integration of alternative risk assessment approaches in Canada’s Chemicals Management Plan (CMP) committee report (CMP Science Committee 2016) and publications from the US EPA (Turley et al. 2019). Institutional legitimacy of RISK21 is increasing, with eight government and regulatory agencies (including Health Canada), two academic institutions, five industries, and two consulting organizations involved (HESI N.D.). A variety of case studies have been published using the RISK21 approach, for example, prioritizing chemicals in drinking water (Wolf et al. 2016; Dellarco et al. 2017) and assessing pyrethroid exposures in bed netting (Doe et al. 2016). However, RISK21 has not been used or validated in the context of Indigenous community-based risk assessments (M. Embry, personal communication, November 2022). In fact, to our knowledge, to date there has been no evidence of collaboration on the development of any next generation risk assessment tools with the Indigenous communities to which they may be ultimately applied.
Therefore, the objective of this paper is to explore and critically examine the use of the RISK21 tool in the context of Indigenous community-level HHRA data, through case studies. The expected outcomes are to first, reflect upon the use of the approach in these contexts, including its benefits and challenges; second, compare the approach (including results and conclusions) to conventional risk assessment methods; and third, to discuss and ideate adaptations or additions to the tool to improve its use in-context. This study is conducted in collaboration with three distinct communities: The Chipewyan Prairie First Nation, Cold Lake First Nations, and Apsáalooke/Crow Nation. The cases use data from risk assessments that have been previously conducted in these communities (Eggers et al. 2018; McAuley et al. 2018). The critical evaluation and discussion of RISK21’s use in these contexts includes perspectives of the community members and their collaborators, who are co-authors of this work.
Methods
The RISK21 approach
HESI describes RISK21 as ‘a flexible framework whereby both exposure and hazard are evaluated using all relevant sources of information, with a goal of reducing unnecessary resource utilization while providing sufficient precision and accuracy to make decisions protective of human health’ (Embry 2018). The approach is centred on four principles: (1) focus on problem formulation; (2) utilize existing information; (3) start with exposure rather than toxicity; and (4) use a tiered approach to data development and decision-making. These principles are organized into a “roadmap”, which provides a basic framework for conducting risk assessments (supplemental materials S1) (Pastoor et al. 2014). Under this framework, risk assessments are conducted beginning at the lowest tier (tier 0) and progressing to higher tiers based on the assessment’s results and available information. Exposure tiers and toxicity tiers can be combined at various levels. For example, if there is a high level of toxicity data available for a chemical, but minimal exposure data available, the two can be used to characterize risk. The decision on what level of precision is required for assessment and decision-making is based upon the problem formulation stage (Pastoor et al. 2014). The case studies presented here will utilize various modalities of the RISK21 roadmap and matrix to explore its potential use in-context, which will be discussed in relation to the above objectives. The RISK21 approach followed in these case studies draws on publications from HESI, including methodology papers on the basic roadmap for HHRA (Embry et al. 2014; Pastoor et al. 2014), and its use demonstrated in a case study (Doe et al. 2016), and two methodology papers on problem formulation and use of the framework to conduct a cumulative risk assessment (Solomon et al. 2016; Moretto et al. 2017) (supplemental materials S2).
Case studies
To respond to our objective, the case studies presented involve the collaboration of three Indigenous communities who have conducted risk assessment activities for environmental contaminants (Table 1). Members and partners of the communities of focus in the case studies are co-authors of this work and have experience with risk assessments, and thus the discussion stems from their critical consideration of the use of the RISK21 approach and ideas for its adaptation. The case studies represent a “pilot test” of RISK21’s use in community contexts, with two distinct risk assessment scenarios: (1) to characterize human health risks from contaminant exposure; and (2) to assess cumulative risks (i.e., the combined risks posed by multiple chemical and non-chemical stressors). All data used in the case studies are publicly available and published online. The risk assessment questions in these cases are those posed at the community level and capture a range of data contexts, through which we may identify the potential limitations and/or benefits of using RISK21 in this context. The results are not meant to be prescriptive, nor are they meant to be interpreted as actionable information on potential health risks.
Table 1.
| Case No. | 1 | 2 |
|---|---|---|
| Community name(s) and location | Chipewyan Prairie First Nation and Cold Lake First Nations, Alberta, Canada | Apsáalooke Nation (Crow Tribal community), Montana, USA |
| Case study name | Assessing human health risks from cadmium exposure through moose tissues in the Chipewyan Prairie First Nation and Cold Lake First Nations | Cumulative risk assessment for contaminated well water and the Apsaálooke/Crow Nation, Montana |
| Risk assessment focus | Characterizing risk | Cumulative risk (chemical and non-chemical) |
| Exposure media of focus | Moose meat (kidney, liver, muscle) | Well water |
| Chemicals of interest | Cadmium | Arsenic, manganese, zinc, uranium, nitrate |
| Reference publications | McAuley et al. 2018. Cadmium tissue concentrations in kidney, liver, and muscle in moose (Alces Alces) from First Nations Communities in Northern Alberta. Frontiers in Sustainable Food Systems, 2(69). Available from https://doi.org/10.3389/fsufs.2018.00069. | Eggers et al. 2018. Community engaged cumulative risk assessment of exposure to inorganic well water contaminants, crow reservation, Montana. International Journal of Environmental Research and Public Health, 15(1). https://doi.org/10.3390/ijerph15010076 |
| Chan et al. 2016. First Nations Food, Nutrition and Environment Study (FNFNES): Results from Alberta 2013. University of Ottawa, Ottawa, ON. 2016:155. Available from https://www.fnfnes.ca/docs/FNFNES_Alberta_Regional_Report_ENGLISH_2019-10-09.pdf |
Results
This section summarizes two case studies on human health risk assessments using the RISK21 approach, following a tiered exposure and/or toxicity assessment depending on the availability of data for each case. While we had access to all data at the outset, for demonstrative purposes we progressed from the use of less “precise” to more “precise” data following RISK21, where “precision” of exposure and toxicity values is defined as “the degree of accuracy in the data… proportional to the quality and quantity of data used to generate the estimate” (Embry et al. 2014). The case studies are organized as two separate results and discussion sections. The discussion sections detail our observations gleaned from each case in relation to the objectives outlined, specific to each community’s perspective. The separation of the case discussions is intended to highlight the uniqueness of the two contexts. We discuss our third objective, ideating potential future adaptations and areas for research, in consideration of both cases.
Case study 1—assessing human health risks from cadmium exposure through moose tissues in the Chipewyan Prairie First Nation and Cold Lake First Nations
Step 1: problem formulation
Scenario: The harvest and consumption of moose is essential to the dietary, social, and cultural well-being of many members of the Chipewyan Prairie First Nation (CPFN) and the Cold Lake First Nations (CLFN) in Alberta, Canada. CPFN and CLFN traditional territory is in Alberta’s oil sands region, and there are concerns about the impacts on moose meat and subsequent human health concerns for consumers in these communities (McAuley et al. 2016, 2018). Of particular concern is the community’s potential exposure to cadmium, a heavy metal that can accumulate in the tissues of moose, particularly in the kidneys and liver (McAuley et al. 2018). This case study will use the basic RISK21 framework and matrix to conduct an HHRA, utilizing data from an original community-based risk assessment published by McAuley et al. (2018).
Purpose: The purpose of this risk assessment is to characterize risks associated with the consumption of cadmium in moose tissues (muscle, kidney, and liver) by community members. The results will be compared to the original human health risk assessment conducted to estimate benchmark consumption quantities in CPFN and CLFN (McAuley et al. 2018).
Step 2: assembly of information
Following the RISK21 framework, we first assemble information relevant to the risk assessment, including chemical properties, biological activity, adverse effects, and use and exposure information (Doe et al. 2016). Cadmium is a well-characterized heavy metal (U.S. EPA 2023a; 2023b) with risk assessment information available through the US EPA’s CompTox Database and Integrated Risk Information System (IRIS) and Health Canada (Health Canada 1986; U.S. EPA 2023a; 2023b). Evidence supporting the information on cadmium is considered the highest possible according to CompTox (Level 1: expert curated, highest confidence in accuracy and consistency of unique chemical identifiers) (U.S. EPA 2023a). Cadmium’s mechanism of action involves effects on cell proliferation, differentiation, and apoptosis. Adverse effects of cadmium include cancer and organ system toxicity (skeletal, urinary, reproductive, cardiovascular, central/peripheral nervous, reproductive) (Rahimzadeh et al. 2017).
The First Nations Food, Nutrition, and Environment study (FNFNES) report from Alberta provides regional consumption rates (average and maximum) of moose meat, kidney, and liver for First Nations adults in Alberta (n = 603) and cadmium concentrations (average and maximum) found in a sample of each tissue (moose meat n = 9; kidney n = 8; liver n = 7) (Chan et al. 2016). Further, community-specific data are available on average and maximum cadmium concentrations from a 2018 study by the CLFN and CPFN, which involved the harvesting of 35 moose (26 from CPFN, 9 from CLFN) and analysis of cadmium concentrations in the tissues (muscle, liver, kidney) (McAuley et al. 2018).
Step 3: tiered risk assessment
While all the data were available at the outset of our case study, we conducted the assessment in two tiers, beginning with lower-tier exposure estimates (tier 0) based upon regional exposure data and progressing to more “precise” exposure estimates (tier 1) based on local community data, demonstrating the stepwise collection of data following the RISK21 approach (Pastoor et al. 2014). In this case study, both steps used the same level of toxicity information (i.e., RfD from US EPA and Health Canada based on mode of action information, which is the highest tier/tier 3).
First assessment—Tier 0 exposure assessment (minimal information)
The first assessment involved the use of lower-tier data (i.e., regional data from the FNFNES study as opposed to community-specific exposure information), simulating the initial availability of minimal data and demonstrating the stepwise use of the RISK21 framework.
a.
Exposure estimate: Using the data from FNFNES on Moose consumption rates for First Nations in Alberta and mean and maximum cadmium concentrations in moose meat, we calculate average and maximum chronic daily intake (CDI) values, providing an exposure range (Table 2).
b.
Toxicity estimate: The US EPA’s CompTox dashboard catalogs cadmium toxicity values for oral exposures at the highest level of evidence (including mechanistic information). EPA’s IRIS provides the No-Observed Adverse Effects Level (NOAEL) and reference doses (RfDs) for cadmium ingested via food. Thus, we will use a toxicity estimate range that spans from the NOAEL (10 ug/kg/day) to the RfD (1 ug/kg/day) (Health Canada 1986; U.S. EPA 2023a).
c.
d.
Assessment of safety: Examining the visual matrix (Fig. 1a), maximum exposure estimates for cadmium in kidney tissue approach the yellow area, beyond which the risks are considered unacceptable to human health by US EPA standards. This assessment has a relatively high degree of uncertainty, as the exposure data used is regional as opposed to community-specific, and there was a small sample size of moose tissues in which cadmium concentrations were measured in the FNFNES study (moose meat n = 9; kidney n = 8; liver n = 7). Also, the assessment does not account for background levels of cadmium (ex: in other foods, soil, water, cigarette smoke). These limitations may prompt us to proceed to the next tier of the RISK21 approach, using community-specific exposure information.
Fig. 1.
Table 2.
| Tier 0 exposure assessment (minimal information) | Tier 1 exposure assessment (deterministic information) | ||||||
|---|---|---|---|---|---|---|---|
| Regional data (FNFNES) | Chipewyan Prairie First Nation | Cold Lake First Nations | |||||
| Tissue type | Daily moose meat consumption (g/kg/day) | Cadmium concentration in tissues (ug/g) | Total CDI (ug/kg/day) | Cadmium concentration in tissues (ug/g) | Total CDI (ug/kg/day) | Cadmium concentration in Tissues (ug/g) | Total CDI (ug/kg/day) |
| Muscle (avg) | 0.165 | 0.009 | 0.001 | 0.039 | 0.006 | 0.004 | 0.001 |
| Muscle (max) | 0.826 | 0.016 | 0.013 | 0.064 | 0.053 | 0.007 | 0.006 |
| Liver (avg) | 0.006 | 1.161 | 0.007 | 2.582 | 0.016 | 0.860 | 0.005 |
| Liver (max) | 0.028 | 3.78 | 0.107 | 24.5 | 0.697 | 1.900 | 0.054 |
| Kidney (avg) | 0.006 | 13.17 | 0.082 | 11.948 | 0.074 | 4.600 | 0.029 |
| Kidney (max) | 0.020 | 31.1 | 0.633 | 78.9 | 1.607 | 11.000 | 0.224 |
Notes: Daily moose meat consumption values are from the FNFNES report on Alberta (Chan et. al. 2016). Tier 0 uses FNFNES exposure estimate values and regional moose meat contaminant measurements (Chan et al. 2016; McAuley et al. 2018). Tier 1 uses exposure estimates from moose harvested by the Chipewyan Prairie First Nation (CPFN) and Cold Lake First Nations (CLFN) (McAuley et al. 2018). The calculation of exposure estimates used standard human body weight (70.7 kg) measures for an average Canadian recommended by Health Canada (Health Canada, 2019), as used in the original risk assessment by McAuley et al. (2018).
Second assessment—Tier 1 (further investigation of exposure)
a.
Estimate of exposure: The second-tier exposure estimate used community-specific data from CPFN and CLFN (i.e., 35 locally harvested moose, 26 from CPFN, and 9 from CLFN). Community-specific consumption data are unavailable, so the same consumption values from the FNFNES were used (Chan et al. 2016) to calculate the CDI values (Table 2).
b.
Estimate of toxicity: As the toxicity information available for cadmium is at the highest level of evidence, the same values will be used as in the first assessment.
c.
d.
Assessment of safety: The community-level data provides a more accurate estimation of exposure and differentiates assessments between the two communities. Compared to the Tier 0 exposure assessment, exposure values for CPFN slightly exceed regional-level values, moving the assessment towards the “red-zone” i.e., the area of potential concern (Fig. 1b). The exposure values for CLFN were slightly lower than values calculated using regional data, moving the assessment towards the “green zone” (i.e., safe level of exposure) (Fig. 1c).
Step 4: outcome and conclusions
A conclusion that may stem from the assessment is that moose kidney and liver should be consumed with caution, especially in the context of CPFN. Certainty of this assessment may increase at higher exposure data tiers (ex: human biomonitoring data).
Case study 2—cumulative risk assessment for contaminated well water and the Apsaálooke/Crow Nation, Montana
The RISK21 approach includes a framework for conducting a cumulative risk assessment (CRA) (i.e., assessing risks of combined exposure to multiple chemical and non-chemical stressors) (Moretto et al. 2017). This approach involves identifying common chemical assessment groups and non-chemical modulating factors. Unlike the first case, the framework for CRA includes a “gatekeeper step” prior to problem formulation. This step assesses the evidence for co-exposure and common toxicity of the contaminants in question, which (under the RISK21 framework) determines the need for a CRA.
Scenario: The Crow Reservation in Montana is home to the Apsáalooke/Crow people, and is rich in water resources, including several mountain-fed rivers. Water quality began to visibly deteriorate in the 1960s with regional expansion of agriculture and mining. This resulted in the installation of costly well water infrastructure in homes, subsequently leading to metals contamination from the wells (Eggers et al. 2015, 2018; Martin et al. 2021). Today, about half of the families on the Reservation rely on home well water for cooking and drinking, although quality remains poor (Eggers et al. 2018). Using a community-based participatory research approach, members from the Apsáalooke/Crow Tribal community and researchers from Montana State University conducted a CRA of exposure to inorganic contaminants (uranium, zinc, manganese, arsenic, and nitrate) in well water, integrating multiple knowledge sources (Eggers et al. 2018). The study involved the quantification of exposures to well water contaminants, a CRA for multiple chemicals, and a discussion of tribal members’ knowledge of the contamination based on surveys and interviews, incorporating secondary health and economic data. Drawing data from the original CRA conducted by Eggers et al. (2018), this demonstrative case will explore the use of the RISK21 CRA framework in-context and compare it to the original assessment.
Step 1: gatekeeper step
For multi-chemical CRA to occur (prior to considering non-chemical stressors), the RISK21 framework requires evidence of both co-exposure of receptors to chemicals and common toxicity (i.e., assuming dose additivity occurs when chemicals have a similar mode of action) (Moretto et al. 2017). However, RISK21 publications acknowledge a lack of methodological consensus on the criteria to determine when it is toxicologically appropriate to include chemicals in a CRA, consistent with literature on the topic (Callahan and Sexton 2007; Moretto et al. 2017). Criteria for conducting a CRA may range from evidence of co-exposure only, to in vitro and in silico methods, to common apical effects or a common mechanism of action (U.S. EPA 2011; Moretto et al. 2017). The contaminants considered in this case have evidence of co-exposure (i.e., are found in the same well water source), and some of them have potentially common endpoints (i.e., diabetes has been found to be associated with U, Mn, and As). In addition, a CRA was justified, and conducted, in the original study by Eggers et al. (2018) (Table 4). Following US EPA and other relevant risk assessment guidance (U.S. EPA 2011; OECD 2018), and aligned with the original community-based study, the present case study thus considers all chemicals within a common chemical assessment group.
Step 2: problem formulation
This risk assessment will aim to answer: “What are the cumulative risks associated with Apsáalooke Tribal community members’ exposures to inorganic well water contaminants?” The contaminants of interest are uranium (U), manganese (Mn), arsenic (As), zinc (Zn), and nitrate (NO3–). The assessment will characterize non-cancer risks associated with contaminants (both cumulatively and independently), and subsequently consider the effects of modulating factors (i.e., economic, cultural, social, legal, and physical factors).
Step 3: assembly of information (exposure and toxicity)
Exposure information (i.e., contaminant measures of 164 wells, 197 participant surveys from 165 Tribal families) for well water at the community level is summarized (Table 3) (Eggers et al. 2018). This information represents a relatively high tier of exposure data (environmental monitoring data—Tier 2 of the RISK21 framework) and provides evidence of common exposure as the contaminants were found in the same well water samples.
Table 3.
| Chemical name | Average exposure (ug/kg/day) | Reference dose (ug/kg/day) | Hazard quotients |
|---|---|---|---|
| Arsenic (As) | 0.034 | 0.3 | 0.114 |
| Manganese (Mn) | 2.914 | 46 | 0.063 |
| Zinc (Zn) | 3.143 | 300 | 0.01 |
| Nitrate (Ni) | 46 | 1600 | 0.029 |
| Uranium (U) | 0.229 | 0.2 | 1.143 |
| SUM (Hazard Index) | 1.356 | ||
Notes: Exposure calculations assume a standard human body weight of 70 kg, and an average well water consumption rate of 2 L/day per the original assessment. The RfDs are from the US EPA IRIS database through US EPA CompTox (https://comptox.epa.gov/dashboard). The right column displays the Hazard Index (HI) method for CRA, which involves summing chemical HQs. Given that the HI is >1, the combined exposure to the chemicals of concern is at a level that poses a human health risk greater than that of individual chemicals. IRIS, Integrated Risk Information System; RfDs, reference doses; CRA, cumulative risk assessment.
Toxicity information available in the Integrated Risk Information System (IRIS) from the US EPA (Table 3) represents the highest level of evidence for chemical toxicity data used for regulatory decision making on human health risks in both the US and Canada (Health Canada 2021; U.S. EPA 2023c; U.S. EPA 2023d). An evidence/concordance table was created to compare details of the toxicity information amongst the chemicals (supplemental materials S3). While it is established that As exposures are associated with carcinogenic risks (International Agency for Research on Cancer 2012), for demonstrative and simplistic purposes, and as the case is not meant to be interpreted as real risk assessment results, the assessment here presented considers non-carcinogenic risks only.
Table 4.
| Modulating factor | EXP or TOX | Description | Direction/ magnitude (increase or decrease) |
|---|---|---|---|
| Income | EXP and TOX | Purchasing bottled water and water treatment is an economic hardship for lower-income families; those that can afford bottled water use it for cooking/drinking (Doyle et al. 2018; Eggers et al. 2015, 2018). Lower socio-economic status families have decreased access to healthcare services, increasing toxicity risks (Mitchell 2019). | Increase and decrease (all contaminants) |
| Access to and adequacy of health services | TOX (?) | Lack of adequate health care services leads to inadequate treatment of potential toxic effects of exposures to chemicals in well water (Eggers et al. 2015). | Increase (all) |
| Physical health | EXP | Poor physical health decreases one’s ability to seek out and transport clean water home, reliance on contaminated well water increases (Eggers et al. 2015; Martin et al. 2021). | Increase (all) |
| Environmental health literacy | EXP | 21% of community members surveyed felt confident they understood well stewardship measures to protect well water from contamination (Doyle et al. 2018; Eggers et al. 2018). No significant difference in exposure for “environmentally literate” participants (Eggers et al. 2018). This has been noted in other studies in Indigenous communities (Dellinger et al. 2022). | Unknown |
| Access to information and education | EXP | Reservation members feel they lack access to information about water quality concerns (Martin et al. 2021). No community-based opportunities for homeowners to learn how to take care of wells or plumbing (Martin et al. 2021). | Increase (all) |
| Environmental regulation | EXP | Inadequate environmental enforcement and complex legal and regulatory issues contribute to drinking water disparities (Eggers et al. 2018). Delay in addressing contamination and establishing clean water infrastructure—under-recognition of Local Knowledge as water quality data needed to secure funding, institutionalization of funding boards, intersecting legal responsibilities (state vs. federal), poor environmental enforcement, lack of tribal authority (Doyle et al. 2018). | Increase (all) |
| Diabetes prevalence rates | TOX | Diabetes prevalence double the statewide average. Diabetes increases vulnerability to nephrotoxic effects of U (Eggers et al. 2018; Menke et al. 2016). U in urine increases likelihood of diabetes (Menke et al. 2016). Physicians in the community health center have noticed that people with diabetes are losing kidney function more rapidly (Eggers et al. 2018). Diabetes has also been associated with both Mn (Shan et al. 2016) and As exposures (Rahman et al. 1998; Tseng et al. 2000; Liu et al. 2016). | Increase (U, Mn, As) |
Step 4: using the RISK21 matrix to assess risk
a.
Plotting information on the RISK21 matrix: The exposure and toxicity information were plotted on the RISK21 matrix (Fig. 2). Lines on the matrix define three areas: area of potential concern for individual compounds (right of the solid black line), area in which chemicals are considered for a CRA (between solid and dashed black lines), and area of no concern (left of the dashed black line). The lines represent cut-off points for exposure/toxicity ratios, which may be determined by policy or the scenario in question (for demonstrative purposes we use 1:1 and 1:1000 margins of exposure for the lower and higher thresholds, respectively). The matrix shows that U exposures alone are at a level which may pose a risk to human health. Depending on the priorities outlined in the problem formulation stage, the risk assessment may stop here until the risks posed by U are addressed. The remaining chemicals lie between the dashed and solid black lines, so do not pose human health risks independently, but we may assess them as a chemical group (Moretto et al. 2017). Notably, as this plot used average exposure estimates, it may not represent “high consumers” (ex.: those exceeding 2 L of water consumption per day or using highly contaminated wells) and does not account for background exposures (ex.: water used in food preparation). Furthermore, oral RfDs are used to estimate toxicity in this assessment, which produces different results than would the use of other reference values. For example, maximum contaminant level goals (MCLGs) for drinking water as defined by the US EPA are set at a value of zero for both As and U (U.S. EPA 2024). Thus, the risks as calculated from RfDs identify U alone as being at a level of concern, but if the EPA MCLGs for drinking water were applied, As would also be identified. The determination of which guidelines should be used in an assessment would be community-specific and dependent upon the problem formulation stage.
b.
Applying CRA method: Based on the above assessment of the RISK21 matrix, we conduct a multi-chemical CRA using an existing methodology. Aligned with the original assessment by Eggers et al. (2018), we used a Hazard Index (HI) method, which involves summing the hazard quotients (HQs) (i.e., the ratios between the exposure and toxicity values of each chemical) (Table 3). An HI > 1.0 indicates that there are non-carcinogenic health risks for well water consumption over a lifetime (Eggers et al. 2018). The five chemicals in this case study produce an HQ of 1.36, indicating an increased risk to human health compared to exposure to each individual chemical. According to Moretto et al. (2017), the HQ cannot be accurately visualized on the RISK21 matrix due to the application of safety factors to reference doses prior to comparing them to exposure estimates. Thus, this assessment will consider individual chemicals and how non-chemical stressors may impact their exposure and toxicity.
c.
Consideration of modulating factors: In the study by Eggers et al. (2018), 30 key informant interviews described economic, cultural, and behavioral factors within the community that may be important when assessing the contaminant data. Several other papers authored by members of the Crow Environmental Health Steering Committee and Tribal Research Group further highlight factors that may modulate contaminant exposure and/or toxicity (Table 4). For example, a study that conducted 97 well water tests and homeowner surveys described physical health (inability to seek out and transport clean water home), location (proximity to U sources), spirituality, and adequacy of health services as factors potentially contributing to contaminant exposure (Eggers et al. 2015). A survey of community members described that the lack of access to education and information about contamination, minimal financial and technical resources, and no routine testing or communication of results, contributed to perceived contaminant-related risks (Martin et al. 2021). Finally, complex jurisdictional and legal issues prevent the Apsáalooke/Crow Tribal community from controlling and owning contaminant monitoring, challenging their efforts to repair and replace water infrastructure and potentially contributing to ongoing exposure (Doyle et al. 2018).
Under RISK21’s framework, modulating factors are understood quantitatively (i.e., effects on biological processes and key events within a mode of action for a given chemical), falling into the categories of host, lifestyle, and environmental factors (Moretto et al. 2017; Simon et al. 2014). However, many of the factors considered in the original community-based assessment (Eggers et al. 2018) are qualitative descriptive measures of importance to the community, including survey information and local knowledge from key informants. Thus, we adapted the evidence table as proposed in the RISK21 framework to display hypothesized relationships between modulating factors and exposure/toxicity. While the magnitude of these factors cannot be precisely determined, their directional effect on either exposure or toxicity were estimated, which was based on qualitative descriptions of their effects observed in the Apsáalooke/Crow Tribal community population (Table 4).
Plotting modulating factors on the RISK21 matrix: Based on the above information on modulating factors, we add directional arrows to the RISK21 matrix to represent hypothesized changes to exposure and/or toxicity that these factors may precipitate (Fig. 2). While there is evidence from the literature to support a potential relationship between modulating factors on exposure to and toxicity of the metals found in well water (both as a group and individually), the effects of these factors (socio-cultural and socio-economic) are variable and often qualitative, and thus it is not possible to accurately estimate the magnitude of increased or decreased risk. Rather, the figure presents a visual representation of the hypothesized potential relationship of these factors to exposure and toxicity.
d.
Conclusions from plotting of modulating factors: Based on the visual representation of modulating factors in relation to individual chemical exposure and toxicity, it is apparent that these factors have the potential to bring some chemicals (i.e., As and Mn) from an area of “no expected risk” to an area of “potential human health risk”, while also magnifying the “potential human health risk” of U.
Fig. 2.
Step 5: outcomes and conclusions
Conclusions that come from the above assessment may be that exposures to well-water contaminants, in combination with one another and in the context of non-chemical modulating factors, potentially pose health risks to Apsáalooke/Crow Tribal community members. Variations amongst individual community members’ physical and social factors may be either protective or further increase contaminant-related risks. We may also use the above assessment to prioritize specific chemicals of concern (i.e., U) that exceed regulatory limits on their own.
Discussion
Objective 1: conventional versus RISK21-based HHRA
Case study 1
The first case study pilots the use of the RISK21 approach for dietary risk characterization of cadmium in moose muscle, kidney, and liver in the CLFN and CPFN. The case used a tiered exposure assessment, initially using regional-level data to estimate exposure values and progressing to the use of community-specific information. We compare the original community-based assessment (McAuley et al. 2018), which examined exposures to cadmium and determined benchmark consumption quantities, to the RISK21-based assessment (tier 0, regional-level exposure estimates, and tier 1, community-level exposure estimates), both quantitatively and descriptively. First, from a quantitative perspective, the original assessment indicates that the risk value (HQ) for high consumers of moose kidney is equal to one. In the RISK21-based assessment, we calculated HQ < 1 for high consumers of cadmium in the tier 0 assessment (regional data). In the tier 1 assessment using community-specific data, in CLFN we also calculated HQ < 1, though in CPFN we calculated HQ > 1 (supplemental materials S4). Notably, the original assessment considered background exposures, with an assumption of a background cadmium exposure of 0.22 ug/kg-bw/day, and thus used an “allowable cadmium exposure” of 0.78 ug/kg-bw/day (2018), while the RISK21-based HQ calculations used a reference dose of 1 mg/kg-bw/day, which does not account for background exposures. Furthermore, the original assessment combined data across three communities, while the RISK21-based assessments were community-specific in tier 1. Lastly, the original assessment determined benchmark consumption quantities, while the RISK21-based assessment did not. From a descriptive or decision-making perspective, the original risk assessment published by McAuley et al. (2018) concluded that moose muscle was safe, while kidney and liver consumption should be limited. In comparison, the results of our RISK21 case study here concluded that kidney should be consumed with caution (i.e., high consumers land in the “red zone” on the matrix). Muscle tissues were in the “green zone” (i.e., no concern), while liver was plotted in the “yellow zone” (i.e., potential concern) for CPFN for the second-tier assessment, while all tissues fell into the “green zone” for CLFN. The use of this information would be at the discretion of decision-makers. In the context of RISK21, the less “precis”, lower-tier estimates (i.e., generated from regional-level data) might be used for preliminary decisions when community-level data are unavailable.
Case study 2
The second case study used the RISK21 CRA framework to demonstrate a quantitative CRA of multiple chemical stressors, and subsequently visualize potential relationships between non-chemical modulating factors and contaminant exposure and toxicity in the context of quantitative exposure and toxicity data, survey data, and secondary demographic information from the Apsáalooke/Crow Tribal community. The assessment may be used to highlight the need for measures to directly reduce exposure (i.e., through environmental health education, remediation, and legal and policy action) and reduce the impact of modulating factors that potentially contribute to disproportionate exposure and toxicity (i.e., addressing social determinants of health linked to socioeconomic status such as housing and water security, better access to primary and tertiary healthcare, etc.).
We compare the assessment conducted on the Crow Reservation by Eggers et al. (2018) to the presented RISK21-based case. Quantitatively, both the original study and RISK21-based assessment used the same CRA methodology (Hazard Index), and thus the results were comparable. Descriptively, conclusions stemming from the RISK21 case study here (i.e., that multiple chemical exposures, and other contextual factors, may increase health risks related to well water contamination for Apsáalooke/Crow Tribal community members) is aligned with those reported in the original assessment by Eggers et al. (2018). Methodologically, a notable difference was the requirement of a “gatekeeper step” in RISK21 for chemicals to have both evidence of common toxicity and co-exposure (Moretto et al. 2017), while in contrast, there is a general lack of agreement from the scientific community on when it is toxicologically appropriate to group chemicals for CRA (Callahan and Sexton 2007; Moretto et al. 2017), and criteria can range from co-exposure only to a variety of toxicity information. US EPA published guidelines and current literatures on CRA were followed in the original community-based case and in the present case study (US EPA 2011; Stoiber et al. 2019). Another key difference was that the RISK21 framework favors quantitative information input on modulating factors, while the information of importance to the original community assessment is largely qualitative and descriptive. Despite these differences, we were able to plot the information on the RISK21 matrix. In this sense, we found the tool to be flexible for a range of data availability contexts and found the matrix useful for integrating and visualizing multiple information sources.
Objective 2: challenges and benefits
Case study 1
In the first case study, there were several data limitations to consider in the interpretation of the RISK21 assessment results. First, cadmium concentrations in moose organ tissues are related to age (Arnold et al. 2006), and the assessment does not account for this. The animals harvested by CLFN were juveniles, therefore the cadmium organ concentrations were much lower than those in animals harvested by CPFN (i.e., average liver cadmium concentrations were 2.582 ug/g in CPFN harvested moose versus 0.860 ug/g in CLFN harvested moose; and similarly, 11.948 versus 4.6 ug/g for kidney and 0.039 versus 0.004 ug/g for muscle, respectively). A second limitation was the use of standardized assumptions to calculate exposure. For example, the use of standard human body weight and regional-level exposure data may miss important outliers related to country food consumption at the community level (ex: children and older adults). While the lower-tier exposure estimates may be useful for an initial assessment, the variations between the first and second assessments’ RISK21 matrices demonstrates that obtaining community-level exposure information can assist with understanding variations between exposures in individual communities. Finally, the RISK21-based case studies presented here did not account for background exposures, potentially leading to an underestimate of exposure values.
We note some challenges with the use of the RISK21 approach in this context. First, the matrix was limited in its application to community-level risk communication. The authors of the original risk assessment did not foresee the tool to be able to help communities to understand risk in a different way. While the matrix is likely simple to use for those with experience in risk assessment and provides a risk visualization tool that can facilitate discussion between risk assessment professionals, it may not be interpretable by the general community or easy to communicate, which is a priority for the CLFN and CPFN. This perspective is consistent with literature on the persistent challenges with risk communication in many other Indigenous communities globally (Boyd and Furgal 2019, 2022; Fernández-Llamazares et al. 2020). The original publication by McAuley et al. (2018) aimed to estimate benchmark consumption quantities associated with minimal health risks for community members (i.e., amounts that can be safely consumed, in servings per month). While the tool can assist with these estimations, the matrix itself is not widely accessible as a risk communication tool.
Furthermore, an important element of risk communication for CPFN and CLFN is the ability to account for potential food insecurity (for example, through the loss of confidence in traditional food systems resulting from consumption advisories) (McAuley and Knopper 2011), which is not inherent to this approach. Previous writings on risk communication in Indigenous communities evidences the potentially negative impacts of miscommunicating or misunderstanding risks related to country foods. For example, an assessment of traditional food alternatives for coastal First Nations communities in BC found that switching to market foods may pose greater health risks than continuing to eat potentially contaminated traditional food (Wiseman and Gobas 2002). Incorporating different units of measure into the RISK21 matrix, such as servings per month as opposed to mg/kg/day, may be more appropriate for community-wide risk communication. Consumption guidance regarding reducing the size of meals from older animals may help to reduce exposure. The National Research Council (1996) writes that decision-making must recognize uncertainties in risk assessments, including both the magnitude and its sources, as well as fundamental biases. RISK21’s visual output may also be made more transparent as a decision-making tool by including a component indicating the precision of the assessment, data tiers used, and uncertainties.
The regional data from the recent FNFNES study was helpful for estimating exposures in the absence of community-specific consumption data and was used in both the RISK21-based and original community HHRAs. In combination with other databases (ex.: US EPA CompTox dashboard), the FNFNES exposure data (i.e., consumption information and average regional concentrations of contaminants found in a variety of traditional foods) can be used for preliminary risk assessment. It is important to emphasize that the use of regional data (when available) would be based upon the community-level risk assessment needs and may not be appropriate to address all risk assessment goals, and thus detailed problem formulation is an essential first step (Pastoor et al. 2014). In the context of a community-based assessment, initial problem formulation needs to be community-led and community perspectives on the data needed for decision-making should be prioritized.
Case study 2
There were several potential challenges with the RISK21 approach in the context of cumulative risk assessment for well water contamination in the Apsaálooke/Crow Nation. First, within-community variations in exposures and modulating factors are difficult to communicate through the visual matrix (for example, socio-economic status and spatial variations in U concentrations) (Eggers et al. 2015). The integration of this information into the visual output may be an area for adaptation. Second, as mentioned above, RISK21’s framework favors quantitative information to estimate the effects of modulating factors on exposure and toxicity (Solomon et al. 2016), though this does not well-capture the historical nature (i.e., over several decades, and in a format that is descriptive and observational) of how water contamination has affected community members (Doyle et al. 2018). For example, Apsáalooke/Crow Tribal community members have noted that the loss of water security has contributed to poor physical health outcomes, deterioration of ecosystems and species vital to spiritual and cultural practices (Doyle et al. 2018), and a loss of connection to the land (Eggers et al. 2015). This is consistent with writings from other communities and Indigenous scholars on the importance of cumulative and holistic risk assessment that includes qualitative information. For example, the Mohawk community of Akwesasne writes that a holistic risk assessment approach must consider multi-disciplinary and qualitative information sources (Arquette et al. 2002). Harris and Harper's (2000) work on eco-cultural dependency webs in tribal health risk assessment describes that elements of health (ecological, human, socio-cultural, and socio-economic) interact in complex ways and must be factored into assessments. These and other “non-conventional” information sources need to be considered in a CRA. Though there is minimal consensus to date on how this might be achieved (Callahan and Sexton 2007; Sexton 2012, 2015), writings on CRA methods have defined that qualitative assessments are possible, depending on the circumstances, and in some cases may be the only way to understand complex cumulative risks (ex.: sociocultural stressors accompanying the known presence of contaminants in traditional foods) (Callahan and Sexton 2007). In this case study, we adapted the RISK21 approach to visualize hypothesized relationships between qualitative data and exposure and toxicity estimates. Development and adaptation of methods to account for qualitative information in risk assessment is an area for further research. Lastly, RISK21 is a human health (as opposed to ecological) risk assessment tool, which is consistent with regulatory standards in which these are conventionally considered as separate assessments. Considering both human and ecological risks, as well as interacting components, is important to include in modernized risk assessments, especially in Indigenous communities. This need was recently highlighted in a statement from the Assembly of First Nations in response to proposed amendments to the Canadian Environmental Protection Act (CEPA) through Bill S-5 (AFN 2022). Further adaptations to the RISK21 tool might consider adding an integrated ecological risk assessment step that is factored into an overall improved understanding of cumulative risk. The need for a shift from single-chemical risk assessment approaches to CRA is recognized in Bill S-5 and globally (National Research Council 2009; WHO 2009; Government of Canada 2022). This case study supports that new CRA methods, such as RISK21 be developed for a variety of data contexts, including in situations with disparate evidence on modes of action for chemicals (which is not readily available for the tens of thousands of chemicals available on the global market), and for a diversity of non-chemical information of importance to communities.
Objective 3: potential adaptations and areas for further research
Conventional risk assessment work in Indigenous communities faces several inherent challenges (Harris 2000; Arquette et al. 2002; Harper et al. 2012; Chong and Basu 2023), and as we consider modernizing risk assessment practices, these challenges must be accounted for. In the current study, we piloted the use of the RISK21 approach in the context of two case studies involving three Indigenous communities and discussed the cases amongst co-authors from the communities of focus. In doing so, we uncovered some considerations for the improvement of next-generation risk assessment approaches such as RISK21.
First, the RISK21 approach follows a “hierarchy” of evidence, in which certain sources of evidence and knowledge are privileged over others through the tiering of information. This approach prioritizes the initial chemical assessment over non-chemical stressors and does not well incorporate non-quantitative (i.e., qualitative, descriptive) information. Similarly, for communities who engage in risk assessment work, the privileging of western-scientific information over Indigenous knowledge systems is an ongoing challenge, as cited in a recent systematic review on the braiding of Indigenous knowledge systems and western sciences in the Alberta oil sands region (Wilcox et al. 2023). These findings and ours highlight the need for a better understanding of biases towards the value of certain evidence sources within risk assessment approaches, and adaptations that best meet community-specific contexts. The US National Research Council (NRC) recommends that cumulative risk assessment should be included in regulatory decision-making, instead of as a part of a general background description of context (NRC 2009; Sexton 2015). Further, writings on Indigenous processes of decision-making emphasize that Traditional Knowledge and political systems must be considered in addition to conventional risk assessment findings to make well-informed environmental management decisions, such as in the Inuvialuit Final Agreement (Indian and Northern Affairs Canada 2005). Similarly, we found that non-quantitative evidence of importance to communities should be considered as primary as opposed to secondary, contextual information. This may include the consideration of non-conventional information sources (ex., local and traditional knowledge, descriptive and qualitative information). The meaningful use of such information in risk assessment, and its potential incorporation into the RISK21 tool is an area for further research.
Second, while the definition of “health” in conventional risk assessment guidance tends to focus on physical health risks related to chemical exposures (Health Canada 2019; U.S. EPA 2022), Indigenous communities and leaders have called for a more holistic perspective (Chong and Basu 2023). For example, the Mohawk community of Akwesasne (Arquette et al. 2002) and the Confederated Tribes of the Umatilla Indian Reservation (Harris and Harper 2000) have published papers with approaches to risk assessment that consider interconnected economic, environmental, political, and social factors. The RISK21 CRA framework may assist in providing a way to visualize and incorporate multiple sources of information related to physical health risks, and we recommend that new approaches such as RISK21 continue to be adapted to consider diverse definitions of “health” and a holistic understanding of health risks. This may include re-defining and increasing the parameters of key terminology used in RISK21 to facilitate its adaptation to specific contexts.
Third, we recommend that risk assessment approaches such as RISK21 continue to strive to improve the communication of results to communities in ways that are meaningful and relevant. In our case studies, it was deemed that the visual communication output (i.e., RISK21 matrix) remained most accessible to those with an expertise with conventional risk assessment work, but less so for community members. For example, axes labeled with exposure and toxicity estimates in ug/kg/day are not readily interpretable into tangible applications for risk management at the community or individual level. The risk communication matrix presented here may be most helpful within a “western-scientific” understanding of risk but may still not be suitable for communities themselves. This is consistent with previous writings on the limitations of conventional risk communication in Indigenous community contexts, especially related to traditional foods (McAuley and Knopper 2011; Robin et al. 2021). Further research may explore how to bridge this gap, including through the consideration of community-level communication from the beginning of risk assessment processes that are led by communities themselves. One idea is to use social media and new technology, which has been evidenced to improve health risk messaging in the Arctic, notably by replacing traditional top-down approaches with two-way communication that in turn can be used to better understand the perceptions and interpretations of messages from the community at large (Krummel and Gilman 2016; Basu et al. 2022). We also note that the communities of focus in the presented case studies, and many others who conduct community-based risk assessment work, have strong communication strategies and ongoing efforts to make risk assessment results relevant to community members. We recommend that newly designed risk assessment approaches consider ways in which their outputs could be made more harmonious with community-level communication efforts.
Finally, and fourth, for many Indigenous communities, risk assessment work requires holistic, community-specific approaches, as well as time to establish trusting relationships between communities and external risk assessors. “21st Century Risk Assessment” of increasing interest to regulatory bodies prioritizes efficiency in decision-making, time, and resource use (National Research Council 2009; Embry et al. 2014; CMP Science Committee 2016; Stucki et al. 2022), which is a needed response to the current paucity of information on contaminants and contaminated sites globally. We recommend that community-based cases such as ours that explore the use of RISK21 and other new approaches may help to gather a diversity of experiences and perspectives and ensure the harmonization of community and regulatory priorities in next-generation risk assessments, and there is a need for more research in this area. Leveraging transdisciplinary approaches such as user experience and design thinking may assist in ideating further improvements to next-generation risk assessments.
Conclusions
Two case studies were conducted to explore the use of the RISK21 HHRA framework and matrix in the context of three Indigenous communities. The first case study involved risk characterization for cadmium exposures through moose tissues consumed by the Cold Lake First Nations and the Chipewyan Prairie First Nation in Alberta. The second case was a cumulative risk assessment (i.e., assessed multiple chemical and non-chemical factors) for exposures to well water contaminants for the Apsáalooke/Crow Tribal community in Montana. Overall, the descriptive results from the RISK21-based analysis performed here were similar to results from the original assessments. This similarity held when we performed RISK21-based analysis with lower data tiers (i.e., using less information than a conventional assessment). We also found RISK21 potentially useful in these contexts for rapid assessment of chemicals, and for visually representing data from multiple sources. However, overall we found that the current RISK21 approach may require adaptations to adequately address the HHRA needs of the Indigenous community contexts presented here. Thus, within the two cases, we identified and discussed several areas in which the RISK21 approach may be adapted for improved relevance and utility within the specific Indigenous community contexts studied here, and these ideas may be utilized by other communities who wish to conduct similar work. These include (1) dismantling hierarchies of evidence that favor quantitative information; (2) broadening to more holistic definitions of health in HHRAs; (3) improving community-level risk communication outputs; and (4) producing more context-specific case studies to understand the use and adaptation of the tool in a variety of contexts.
Limitations
The two case studies presented represent unique risk assessment situations in three Indigenous communities across North America. However, as is the nature of case studies, the findings reflect a snapshot in time in the communities of focus. In addition, they were purposefully selected here for demonstrative purposes. There is a wide diversity of communities and purposes for which risk assessments are used, as demonstrated in the FNFNES study, which reported large differences in exposure estimates according to seasonal, agricultural, and traditional food patterns and use among First Nations communities across Canada (Chan et al. 2021). This diversity extends internationally, as described in a recent review of Indigenous peoples and pollution (Fernandez-Llamazares et al. 2020). Thus, the presented cases using RISK21 here may not be widely generalizable. Further, assumptions inherent to risk assessment increase uncertainty in the context of Indigenous communities. For example, the standard body weight from Health Canada is derived from the general population, although it is well-known that nutrition and food security in Indigenous communities contributes to differences in average weight. Finally, though the data collection methods used in the originally cited assessments were rigorous, and every effort was made to minimize biases, researcher bias is inherent to case study methodology, and is present in our reported results and interpretations.
Acknowledgements
This research project was supported by the Canada Research Chairs program (N.B.), the Natural Sciences and Engineering Research Council of Canada, NSERC Alliance Program (ALLRP 578553) for a project titled “Indigenization of new approach methods for contaminated sites assessments in Indigenous communities: A community-based case study in the Kanien'kéha:ka community of Kanesatake” (N.B.), and fellowship support to KC from the PURE CREATE program (funded by NSERC, with supporting monies from McGill University) and the Sustainable Agricultural Fellowship from McGill University. Funding for the CLFN and CPFN moose studies was provided through a contribution agreement by the First Nations Environmental Contaminants Program (FNECP) of the First Nations and Inuit Health Branch (FNIHB) at Indigenous Services Canada (ISC). We would like to thank members of the ECHIP and ALLIANCE Project (especially Eugene Nicholas, Mary Nicholas, Gabrielle Lamouche, Valerie Gabriel, and Isabelle-Anne Bisson) in the Kanien'kehà:ka (Mohawk) Community of Kanesatake in Quebec, Canada for inspiring this work and engaging in fruitful discussions. We also thank Dr. Michelle Embry for her expertise on the use of RISK21 at various stages throughout the project’s progress, and Jenny Eng for technical and managerial support.
References
Arnold S.M., Zarnke R.L., Lynn T.V., Chimonas M.-A.R., Frank A. 2006. Public health evaluation of cadmium concentrations in liver and kidney of moose (Alces Alces) from four areas of Alaska. Science of the Total Environment, 357(1–3): 103–111.
Arquette M., Cole M., Cook K., LaFrance B., Peters M., Ransom J., et al. 2002. Holistic risk-based environmental decision making: a native perspective. Environmental Health Perspectives, 110: 259–264.
Assembly of First Nations. 2022. Submission to the House of Commons Committee on Environment and Sustainable Development. Study on Bill S-5 an Act to amend the Canadian Environmental Protection Act, 1999, to make related amendments to the Food and Drugs Act and to repeal the Perfluorooctance Sulfonate Virtual Elimination Act. Available from https://www.ourcommons.ca/Content/Committee/441/ENVI/Brief/BR12143095/br-external/AssemblyOfFirstNations-e.pdf [accessed 29 December 2023].
Basu N., Abass K., Dietz R., Krummel E, Rautio A, Weihe P 2022. The impact of mercury contamination on human health in the Arctic: A state of the science review. Science of the Total Environment 831: 154793.
Boyd A.D., Furgal C.M. 2019. Communicating environmental health risks with indigenous populations: a systematic literature review of current research and recommendations for future studies. Health Communication, 34(13): 1564–1574.
Boyd A.D., Furgal CM. 2022. Towards a participatory approach to risk communication: the case of contaminants and Inuit health. Journal of Risk Research, 25(7): 892–910.
Burger J., Gochfeld M. 2011. Conceptual environmental justice model for evaluating chemical pathways of exposure in low-income, minority, Native American, and other unique exposure populations. American Journal of Public Health, 101(S1): S64–S73.
Callahan M.A., Sexton K. 2007. If cumulative risk assessment is the answer, what is the question? Environmental Health Perspectives, 115(5): 799–806.
Chan H.M., Fediuk K., Batal M., Sadik T., Tikhonov C., Ing A., Barwin L. 2021. The First Nations Food, Nutrition and Environment Study (2008–2018)—rationale, design, methods and lessons learned. Canadian Journal of Public Health, 112: 8–19.
Chan L., Receveur O., Batal M., David W., Schwartz H., Ing A., et al. 2016. First Nations Food, Nutrition and Environment Study (FNFNES): results from Alberta 2013. University of Ottawa, Ottawa, ON, p. 155. Available from https://www.fnfnes.ca/docs/FNFNES_Alberta_Regional_Report_ENGLISH_2019-10-09.pdf [accessed 17 November 2023].
Chong K., Basu N. 2023. Contaminated sites and Indigenous peoples in Canada and the US: a scoping review. Integrated Environmental Assessment and Management, 20: 1306–1329.
CMP Science Committee. 2016. Meeting no. 5: 16-17 November 2016 integrating new approach methodologies within the CMP: identifying priorities for risk assessment, Existing Substances Risk Assessment Program. Available from https://www.ec.gc.ca/ese-ees/default.asp?lang=En&n=172614CE-1 [accessed 17 November 2023].
Damman S., Eide W.B., Kuhnlein HV. 2008. Indigenous peoples’ nutrition transition in a right to food perspective. Food Policy, 33(2): 135–155.
Dellarco M., Zaleski R., Gaborek B., Qian H., Bellin C., Egeghy P., et al. 2017. Using exposure bands for rapid decision making in the RISK21 tiered exposure assessment. Critical Reviews in Toxicology, 47(4): 317–341.
Dellinger M.J., Pingatore N., Chelius T., Visotcky A., Sparapani R., Ripley M. 2022. Environmental health literacy for Anishinaabe (Great Lakes Native American) fish consumers: a randomized control trial. Environmental Research, 212: 113335.
Doe J.E., Lander D.R., Doerrer N.G., Heard N., Hines R.N., Lowit A.B., et al. 2016. Use of the RISK21 roadmap and matrix: human health risk assessment of the use of a pyrethroid in bed netting. Critical Reviews in Toxicology, 46(1): 54–73.
Doyle J.T., Kindness L., Realbird J., Eggers M.J., Camper AK. 2018. Challenges and opportunities for tribal waters: addressing disparities in safe public drinking water on the Crow Reservation in Montana, USA. International Journal of Environmental Research and Public Health, 15(4): 567.
Eggers M.J., Doyle J.T., Lefthand M.J., Young S.L., Moore-Nall A.L., Kindness L., et al. 2018. Community engaged cumulative risk assessment of exposure to inorganic well water contaminants, Crow Reservation, Montana. International Journal of Environmental Research and Public Health, 15(1): 76.
Eggers M.J., Moore-Nall A.L., Doyle J.T., Lefthand M.J., Young S.L., Bends A.L., Crow Environmental Health Steering Committee and Camper AK. 2015. Potential health risks from uranium in home well water: an investigation by the Apsaalooke (Crow) tribal research group. Geosciences, 5(1): 67–94.
Embry M.R., Bachman A.N., Bell D.R., Boobis A.R., Cohen S.M., Dellarco M., et al. 2014. Risk assessment in the 21st Century: roadmap and matrix. Critical Reviews in Toxicology, 44: 6–16.
Embry MR. 2018. HESI’s risk assessment in the 21st Century (RISK21) framework: overview and applications online, ACS. Available from https://www.agrodiv.org/webinars/hesis-risk-assessment-in-the-21st-century-risk21-framework-overview-and-applications/ [accessed 17 November 2023].
Fernández-Llamazares Á., Garteizgogeascoa M., Basu N., Brondizio E.S., Cabeza M., Martínez-Alier J., et al. 2020. A State-of-the-art review of indigenous peoples and environmental pollution. Integrated Environmental Assessment and Management, 16(3): 324–341.
Fox M.A., Brewer L.E., Martin L. 2017. An overview of literature topics related to current concepts, methods, tools, and applications for cumulative risk assessment (2007–2016). International Journal of Environmental Research and Public Health, 14(4): 389.
Government of Canada. 2017. Canadian arctic contaminants assessment report-human health assessment. Available from https://pubs.aina.ucalgary.ca/ncp/84294.pdf [accessed 20 November 2023].
Government of Canada. 2021. Federal contaminated sites inventory. Available from https://www.tbs-sct.gc.ca/fcsi-rscf/home-accueil-eng.aspx [accessed April 2022].
Government of Canada. 2022. Bill S-5, strengthening environmental protection for a Healthier Canada Act—summary of amendments. Ottawa, Ontario. Available from https://www.canada.ca/en/services/environment/pollution-waste-management/strengthening-canadian-environmental-protection-act-1999/bill-c-28-strengthening-environmental-protection-healthier-canada-act-summary-amendments.html [accessed 17 November 2023].
Harper B., Harding A., Harris S., Berger P. 2012. Subsistence exposure scenarios for tribal applications. Human and Ecological Risk Assessment: An International Journal, 18(4): 810–831.
Harris S.G., Harper BL. 2000. Using eco-cultural dependency webs in risk assessment and characterization of risks to tribal health and cultures. Environmental Science and Pollution Research, 2(2): 91–100.
Harris SG. 2000. Risk analysis: changes needed from a Native American perspective. Human and Ecological Risk Assessment: An International Journal, 6(4): 529–535.
Health Canada. 1986. Cadmium. Available from https://www.hc-sc.gc.ca/ewh-semt/pubs/water-eau/cadmium/index-eng.php [accessed 17 November 2023].
Health Canada. 2004. Federal contaminated site risk assessment in Canada. Available from https://publications.gc.ca/collections/Collection/H46-2-04-367E.pdf [accessed 17 November 2023].
Health Canada. 2019. Selection of body weights values for use in human health risk assessment (HHRA) conducted by Health Canada. Task Force on Scientific Risk Assessment. Available from https://www.canada.ca/content/dam/hc-sc/documents/services/health/publications/science-research-data/selection-body-weights-values-human-health-risk-assessment/27-16-1696_Body-Weight-Use-HHRA-ENG.pdf [accessed 17 November 2023].
Health Canada. 2021. Federal contaminated site risk assessment in Canada: toxicological reference values (TRVs). Government of Canada. Available from https://publications.gc.ca/collections/collection_2021/sc-hc/H129-108-2021-eng.pdf [accessed 18 December 2023].
HESI (Health and Environmental Sciences Institute). 2023. RISK 21 webtool. Available from https://risk21.org/ [accessed 27 October 2023].
HESI (Health and Environmental Sciences Institute). N.D. Fact sheet: risk assessment in the 21st Century (RISK21). Health and Environmental Sciences Institute. Available from https://hesiglobal.org/wp-content/uploads/2022/01/RISK21-2021-v2.pdf [accessed 17 November 2023].
Indian and Northern Affairs Canada. 2005. Inuvialuit final agreement. Available from https://irc.inuvialuit.com/sites/default/files/Inuvialuit%20Final%20Agreement%202005.pdf [accessed 17 November 2023].
International Agency for Research on Cancer. 2012. Arsenic, metals, fibres and dusts. IARC monographs on the evaluation of carcinogenic risks to humans. Available from https://publications.iarc.fr/Book-And-Report-Series/Iarc-Monographs-On-The-Identification-Of-Carcinogenic-Hazards-To-Humans/Arsenic-Metals-Fibres-And-Dusts-2012 [accessed 8 March 2024].
Krummel E.M., Gilman A. 2016. An update on risk communication in the Arctic. International Journal of Circumpolar Health 75: 33822.
Liu B.F., Wang J., Li Y., Han X., Hu H., Guo H., Zhang X. 2016. Association of urinary metals levels with type 2 diabetes risk in coke oven workers. Environmental Pollution, 210: 1–8.
Martin C., Simonds V.W., Young S.L., Doyle J., Lefthand M., Eggers MJ. 2021. Our relationship to water and experience of water insecurity among Apsáalooke (Crow Indian) people, Montana. International Journal of Environmental Research and Public Health, 18(2): 582.
McAuley C., Knopper LD. 2011. Impacts of traditional food consumption advisories: compliance, changes in diet and loss of confidence in traditional foods. Environmental Health, 10: 55.
McAuley C., Dersch A., Kates L.N., Sowan D.R., Koppe B., Ollson CA. 2016. Assessment of exposure to chlorinated organics through the ingestion of moose meat for a Canadian First Nation community. Frontiers in Environmental Science, 4(78).
McAuley C., Dersch A., Mouille-Malbeuf S., Koppe B., Sowan D. 2018. Cadmium tissue concentrations in kidney, liver and muscle in moose (Alces Alces) from First Nations communities in Northern Alberta. Frontiers in Sustainable Food Systems, 2(69).
Menke A., Guallar E., Cowie CC. 2016. Metals in urine and diabetes in U.S. adults. Diabetes, 65: 164–171.
Mitchell F. 2019. Water (in) security and American Indian health: social and environmental justice implications for policy, practice, and research. Public Health, 176: 98–105.
Moretto A., Bachman A., Boobis A., Solomon K.R., Pastoor T.P., Wilks M.F., Embry MR. 2017. A framework for cumulative risk assessment in the 21st century. Critical Reviews in Toxicology, 47(2): 85–97.
National Research Council. 1996. Understanding risk: informing decisions in a democratic society. National Academies Press. Available from https://nap.nationalacademies.org/read/5138/chapter/1 [accessed 17 November 2023].
National Research Council. 2009. Science and decisions: advancing risk assessment. National Academies Press, Washington, DC, US.
National Research Council. 2009. Science and decisions: advancing risk assessment. The National Academies Press, Washington, DC. Available from https://pubmed.ncbi.nlm.nih.gov/25009905/ [accessed 6 March 2024].
O'Neill C.A. 2000. Variable justice: environmental standards, contaminated fish, and “acceptable” risk to native peoples. Stanford Environmental Law Journal, 19: 3. Available from https://digitalcommons.law.seattleu.edu/faculty/633?utm_source=digitalcommons.law.seattleu.edu%2Ffaculty%2F633&utm_medium=PDF&utm_campaign=PDFCoverPages.
OECD. 2018. Considerations for assessing the risks of combined exposure to multiple chemicals, series on testing and assessment No. 296, Environment, Health and Safety Division, Environment Directorate. Available from https://www.oecd.org/chemicalsafety/risk-assessment/considerations-for-assessing-the-risks-of-combined-exposure-to-multiple-chemicals.pdf [accessed 17 November 2023].
OECD. 2020. Linking indigenous communities with regional development in Canada. [accessed 17 November 2023].
Pastoor T.P., Bachman A.N., Bell D.R., Cohen S.M., Dellarco M., Dewhurst I.C., et al. 2014. A 21st century roadmap for human health risk assessment. Critical Reviews in Toxicology, 44(Suppl 3): 1–5.
Rahimzadeh M., Rahimzadeh M., Kazemi S. 2017. Cadmium toxicity and treatment: an update. Caspian Journal of Internal Medecine, 8(3): 135.
Rahman M.T., Ahmad SA., Axelson O. 1998. Diabetes mellitus associated with arsenic exposure in Bangladesh. American Journal of Epidemiology, 148: 198–203.
Robin T., Burnett K., Parker B., Skinner K. 2021. Safe food, dangerous lands? Traditional foods and indigenous peoples in Canada. Frontiers in Communication, 6: 749944.
Sexton K. 2012. Cumulative risk assessment: an overview of methodological approaches for evaluating combined health effects from exposure to multiple environmental stressors. International Journal of Environmental Research and Public Health, 9(2): 370–390.
Sexton K. 2015. Cumulative health risk assessment: finding new ideas and escaping from the old ones. Human and Ecological Risk Assessment: An International Journal, 21(4): 934–951.
Shan Z.C., Sun T., Luo C., Guo Y., Yu X., Yang W., et al. 2016. U-shaped association between plasma manganese levels and type 2 diabetes. Environmental Health Perspectives, 124: 1876–1881.
Simon T.W., Simons S.S., Preston R.J., Boobis A.R., Doerrer N.G., Fenner-Crisp P.A., McMullin T.S., McQueen C.A., Rowlands J.C. 2014. The use of mode of action information in risk assessment: Quantitative key events/dose-response framework for modeling the dose-response for key events. Critical Reviews in Toxicology 44: 17–43.
Solomon K.R., Wilks M.F., Bachman A., Boobis A., Moretto A., Pastoor T.P., et al. 2016. Problem formulation for risk assessment of combined exposures to chemicals and other stressors in humans. Critical Reviews in Toxicology, 46(10): 835–844.
Stoiber T., Temkin A., Andrews D., Campbell C., Naidenko O.V. 2019. Applying a cumulative risk framework to drinking water assessment: a commentary. Environmental Health, 18: 37.
Stucki A.O., Barton-Maclaren T.S., Bhuller Y., Henriquez J.E., Henry T.R., Nagy E.G., et al. 2022. Use of new approach methodologies (NAMs) to meet regulatory requirements for the assessment of industrial chemicals and pesticides for effects on human health. Frontiers in Toxicology, 4.
Tong R., Zhang B. 2023. Cumulative risk assessment for combinations of environmental and psychosocial stressors: a systematic review. Integrated Environmental Assessment and Management, 20: 602–615.
Tseng C.H., Tai T.Y., Chong C.K., Tseng C.P., Lai M.S., Lin B.J., et al. 2000. Long-term arsenic exposure and incidence of non-insulin-dependent diabetes mellitus: a cohort study in arseniasis-hyperendemic villages in Taiwan. Environmental Health Perspectives, 108: 847–851. Available from http://ehpnet1.niehs.nih.gov/docs/2000/108p847-851tseng/abstract.html.
Turley A.E., Isaacs K.K., Wetmore B.A., Karmaus A.L., Embry M.R., Krishan M. 2019. Incorporating new approach methodologies in toxicity testing and exposure assessment for tiered risk assessment using the RISK21 approach: case studies on food contact chemicals. Food and Chemical Toxicology, 134: 110819.
U.S. EPA. 2020. Population surrounding 1857 superfund remedial sites. Available from https://www.epa.gov/sites/default/files/2015-09/documents/webpopulationrsuperfundsites9.28.15.pdf [accessed 17 November 2023].
U.S. EPA. 2022. Human health risk assessment. Available from https://www.epa.gov/risk/human-health-risk-assessment [accessed 6 April 2023].
U.S. EPA. 2023a. Cadmium. Comptox chemicals dashboard. U.S. Environmental Protection Agency. Available from https://comptox.epa.gov/dashboard/chemical/details/DTXSID1023940 [accessed 1 March 2023].
U.S. EPA. 2023b. Cadmium. Interated Risk Information System (IRIS). Available from https://iris.epa.gov/ChemicalLanding/&substance_nmbr=141 [accessed 1 March 2023].
U.S. EPA. 2023c. Basic information about the integrated risk information system. Available from https://www.epa.gov/iris/basic-information-about-integrated-risk-information-system [accessed 20 March 2023].
U.S. EPA. 2023d. Manganese. Integrated Risk Information System. Available from https://iris.epa.gov/ChemicalLanding/&substance_nmbr=373 [accessed 10 March 2023].
U.S. EPA. 2024. National primary drinking water regulations. Ground water and drinking water. Available from https://www.epa.gov/ground-water-and-drinking-water/national-primary-drinking-water-regulations#Inorganic [accessed 8 March 2024].
United Nations. 2007. United Nations declaration on the rights of indigenous peoples. Available from https://www.un.org/development/desa/indigenouspeoples/wp-content/uploads/sites/19/2018/11/UNDRIP_E_web.pdf.
US EPA. 2011. Technical support document—EPA’s 2011 National-scale air toxics assessment. Office of Air Quality, Planning and Standards. Available from https://www.epa.gov/sites/default/files/2015-12/documents/2011-nata-tsd.pdf [accessed 29 February 2024].
Van Oostdam J., Donaldson S.G., Feeley M., Arnold D., Ayotte P., Bondy G., et al. 2005. Human health implications of environmental contaminants in Arctic Canada: a review. Science of the Total Environment, 351-352: 165–246.
Wilcox A.A., Provencher J.F., Henri D.A., Alexander S.M., Taylor J.J., Cooke S.J., et al. 2023. Braiding indigenous knowledge systems and Western-based sciences in the Alberta oil sands region: a systematic review. Facets, 8(1): 1–32.
Wiseman C.L.S., Gobas F. 2002. Balancing risks in the management of contaminated First Nations fisheries. International Journal of Environmental Health Research, 12(4): 331–342.
Wolf D., Bachman A., Barrett G., Bellin C., Goodman J.I., Jensen E., et al. 2016. Illustrative case using the RISK21 roadmap and matrix: prioritization for evaluation of chemicals found in drinking water. Critical Reviews in Toxicology, 46(1): 43–53.
World Health Organization (WHO). 2009. International Programme on Chemical Safety & Inter-Organization Programme for the Sound Management of Chemicals, Assessment of combined exposures to multiple chemicals: report of a WHO/IPCS international workshop on aggregate/cumulative risk assessment. Available from https://www.who.int/publications/i/item/9789241563833 [accessed 6 March 2024].
Supplementary material
Supplementary Material 1 (DOCX / 916 KB).
- Download
- 915.24 KB
Information & Authors
Information
Published In
FACETS
Volume 10 • January 2025
Pages: 1 - 16
Editor: Candace Nykiforuk
History
Received: 6 May 2024
Accepted: 21 November 2024
Version of record online: 24 February 2025
Copyright
© 2025 Copyright remains with the authors Chong, Chavez, Doyle, Eggers, LaFrance, Lefthand, MacDermid, McAuley, Simonds, Young, and Basu; and Chipewyan Prairie First Nation. This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.
Data Availability Statement
Data generated or analyzed during this case study are provided within the article. The original published articles are referenced and publicly available online.
Key Words
Sections
Subjects
Authors
Author Contributions
Conceptualization: KC, NB
Funding acquisition: NB
Investigation: KC
Project administration: KC
Resources: MC, AD, JD, ME, JL, MJL, FM, CM, VS, SLY
Supervision: NB
Visualization: KC
Writing – original draft: KC
Writing – review & editing: MC, AD, JD, ME, JL, MJL, FM, CM, VS, SLY, NB
Competing Interests
The authors declare there are no competing interests.
Funding Information
Canada Research Chairs program
NSERC Alliance Program: ALLRP, 578553
First Nations Environmental Contaminants Program (FNECP) of the First Nations and Inuit Health Branch (FNIHB) at Indigenous Services Canada (ISC)
McGill University: Sustainable, Agriculture, Fellowship
PURE CREATE program (funded by NSERC, with supporting monies from McGill University)
Metrics & Citations
Metrics
Other Metrics
Citations
Cite As
Katherine Chong, Madisan Chavez, Ave Dersch, John Doyle, Margaret Eggers, JoRee LaFrance, Myra J Lefthand, Findlay MacDermid, Claire McAuley, Vanessa Simonds, Sara L Young, and Niladri Basu. 2025. Exploring the use of the RISK21 approach for Indigenous community-based human health risk assessments: two case studies. FACETS.
10: 1-16.
https://doi.org/10.1139/facets-2024-0085
Export Citations
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.
There are no citations for this item