Invited Perspective: Advancing Cumulative Approaches in Regulatory Decision Making

Despite longstanding calls for a shift toward cumulative approaches that better capture real-world exposures,^1–5 most environmental regulations in the United States address individual chemicals. As part of the regulatory process, population exposure and dose–response information for a single chemical is incorporated into a health impact function (HIF) to quantify the health effects avoided by reducing exposure.^6,7 This information can be subsequently combined with valuation estimates to conduct benefit–cost analyses and help choose between policy options.^6,7 In contrast, a cumulative approach to regulation would entail analyzing exposures to multiple chemical and nonchemical stressors and considering the ways in which these exposures may combine or interact to worsen health and associated economic costs.^1,8

Cumulative approaches are key to assessing the multitude of exposures present in our environments. A growing literature indicates that failure to consider environmental exposures holistically may result in underestimating risks, particularly in disadvantaged and marginalized communities that are disproportionately exposed to chemicals and nonchemical stressors.^9–11 Chemical mixtures may elicit effects that would be missed using a single-chemical approach. A recent systematic review found that the majority of identified studies on mixtures found evidence of additivity or synergy (65% and 20%, respectively), with the remaining 15% demonstrating antagonism.¹⁰ There is compelling evidence of synergy among certain ubiquitous classes of chemicals, such as pesticides and heavy metals.¹² Moreover, many complex mixtures present in our environments remain unstudied.¹⁰ Nonchemical stressors (e.g., poverty, racism, comorbidities, climate change) are also important considerations, given that these coexposures may increase susceptibility to chemicals.^1,9,11,13,14

In this issue of Environmental Health Perspectives, Coffman et al. demonstrate a developmental version of the US EPA’s Environmental Benefits Mapping and Analysis Program—Community Edition (BENMAP-CE)—a computer program for conducting benefit–cost analyses—that allows for consideration of the health risks and associated economic burdens of exposures to multiple chemicals.¹⁵ By facilitating quantification of potential interactions among chemicals, the developmental version of BENMAP-CE represents an important advance in the US EPA’s capability for benefit–cost analyses of cumulative risks. Coffman et al. illustrate this advance by simulating changes in concentrations of multiple air pollutants in Atlanta, Georgia, between 2011 and 2025 and quantifying the associated changes in pediatric emergency department visits for asthma. The authors quantified these changes using three approaches: a multipollutant HIF without interactions, a multipollutant HIF with interactions, and single-pollutant HIFs summed across common groupings of air pollutants.

Each approach employed by Coffman et al. used concentration–response functions from an analysis published by Winquist et al.,¹⁶ in part because the statistical methods used therein were less computationally complex and more interpretable as compared with many other epidemiological studies of mixtures. This highlights an opportunity for improved communication between epidemiologists and risk assessors; epidemiological studies tend to lack sufficient methodological detail for use in a quantitative risk assessment framework.¹⁷ Improving methodological reporting recommendations for epidemiological studies would allow more of the literature to be used in support of risk assessment. As a result, a broader base of epidemiologic evidence would be amenable to support the assessment of chemical interactions needed for cumulative approaches to environmental regulation. In addition, certain statistical models for mixtures analysis may be unable to be parameterized, thereby precluding their use in an HIF. Increased collaboration across the disciplines would help to bridge the gap between hypothesis-driven epidemiological studies and the outputs required to support a benefit–cost analysis that accounts for cumulative risks.

In comparing multipollutant HIFs with and without interactions, Coffman et al. found that inclusion of interaction terms tended to result in higher estimates of avoided emergency department visits for pediatric asthma. In general, there was consistency between the results of multipollutant HIFs and single-pollutant HIFs summed across chemicals. For certain scenarios of air pollutant exposures, results of the summed single-pollutant HIFs indicated greater reductions in pediatric asthma emergency department visits than the multipollutant HIFs.

Based on their results, Coffman et al. suggest that the value added by a multipollutant approach should be considered alongside the time and resource investments required. We believe that replication of this approach using different epidemiological studies is warranted to increase confidence that the three approaches reliably yield similar results. In addition, the multipollutant model may be suitable for future modifications that account for factors that render populations more susceptible to health effects, aligning with stated agency priorities of better characterizing and addressing cumulative population burdens.^18,19 Additional case studies of the BENMAP-CE model should explore assessment of potential interactions among chemical and nonchemical stressors. These modifications are needed both to better quantify the impacts of real-world exposures and to address longstanding environmental injustices.

Refers to https://doi.org/10.1289/EHP12969

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.