SEPA
United States
Environmental
Protection Agency
EPA/100/B-24/001 \ January 2025 | www.epa.gov/research
GUIDELINES FOR CUMULATIVE RISK ASSESSMENT
PLANNING AND PROBLEM FORMULATION
January 2025
Risk Assessment Forum
U.S. Environmental Protection Agency
Washington, DC 20460
-------
vvEPA
EPA/100/B-24/001
January 2025
Guidelines for Cumulative Risk Assessment
Planning and Problem Formulation
Risk Assessment Forum
U.S. Environmental Protection Agency
-------
DISCLAIMER
This document has been reviewed in accordance with U.S. Environmental Protection Agency (EPA)
policy. Mention of trade names or commercial products does not constitute endorsement or
recommendation for use. Any opinions, findings, conclusions, or recommendations do not change or
substitute for any statutory or regulatory provisions. While the statutory provisions described in this
document contain legally binding requirements, this document does not impose legally binding
requirements, nor does it confer legal rights, impose legal obligations, or implement any statutory or
regulatory provisions.
Preferred citation: U.S. EPA (U.S. Environmental Protection Agency). (2025). Guidelines for Cumulative
Risk Assessment Planning and Problem Formulation. (EPA/100/B-24/001). Washington, D.C.: Risk
Assessment Forum, U.S. EPA.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
ii
-------
CONTENTS
DISCLAIMER ii
GLOSSARY OF KEY TERMS v
ACRONYMS AND ABBREVIATIONS vii
PREFACE viii
CHAPTER 1. INTRODUCTION 1
1.1. Early Considerations for Planning and Scoping in Cumulative Risk Assessment 4
1.2. Organization of This Document 6
CHAPTER 2. CUMULATIVE RISK ASSESSMENT PLANNING AND SCOPING 8
2.1. Decision Context and Initiating Factors 8
2.2. Participant and Stakeholder Involvement 10
2.3. Statement of Purpose 11
2.4. Scoping Cumulative Risk Assessment Objectives, Boundaries, and Constraints 12
2.5. Tiering and Phasing the Assessment 16
2.6. Data Quality, Needs, Availability 19
2.7. Project and Risk Management Considerations 21
2.8. Peer Review 22
CHAPTER 3. PROBLEM FORMULATION 23
3.1. Examining Risk Management Options Based on the Initiating Factor 23
3.2. Conceptual Model 23
3.2.1. Consideration of Stressors 28
3.2.2. Receptors of Potential Interest 29
3.2.3. Adverse Effect and Exposure Stressor Groups 30
3.2.4. Integration of Data for Examining Stressor-Response Relationship(s) 31
3.3. Analysis Plan 33
3.4. Uncertainties and Variability 36
3.5. Next Steps in Cumulative Risk Assessment 37
REFERENCES 38
APPENDIX A . EVOLUTION OF CUMULATIVE RISK ASSESSMENT AT THE EPA A-l
A.l. Background and History A-l
A.2. Examples of CRA at the EPA A-5
A.3. Looking Ahead toward CRA Advances A-6
APPENDIX B . CUMULATIVE RISK ASSESSMENT FOUNDATIONAL DOCUMENTS B-l
APPENDIX C . EXAMPLE OF A RESPONSE-BASED CONCEPTUAL MODEL C-l
APPENDIX D . EXPOSURE-RESPONSE MODIFIERS D-l
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025 iii
-------
TABLES
Table 1. Considerations for Cumulative Risk Assessment 14
Table 2. Outline for a Stepwise Approach to Cumulative Risk Assessment Planning and
Problem Formulation 28
Table B-l. EPA Publications Relevant to Cumulative Risk Assessment Planning and Problem
Formulation B-l
Table B-2. National Research Council and Other Publications Relevant to Cumulative Risk
Assessment Planning and Problem Formulation B-8
FIGURES
Figure 1. Framework for Human Health Risk Assessment 3
Figure 2. Timeline of CRA Publication Milestones [Selected Milestones] 4
Figure 3. Elements of a Conceptual Model' 25
Figure 4. Example of a Generalized Conceptual Model Evaluating Cumulative Risk and the
Exposure-Disease Paradigm 27
Figure C-l. Conceptual Model for Factors Influencing the Risk of Cardiovascular Disease C-2
Figure D-l. Vulnerability Factor Categories, Interactions, and Pathways in Exposure-Response
Relationships D-2
Figure D-2. Key Intrinsic Events in Exposure-Response Relationships D-3
Figure D-3. A Framework for Incorporating Nonchemical Stressors into Risk Assessments D-4
TEXT BOXES
Text Box 1. Cumulative Risk Assessment Planning Milestones 7
Text Box 2. Hypothetical Cumulative Risk Assessment Statement of Purpose 11
Text Box 3. Tiers for Cumulative Risk Assessment Analysis 17
Text Box 4. Elements in a Cumulative Risk Assessment Phased Approach 18
Text Box 5. Exposure-Response Modifiers and Stressors 29
Text Box 6. Factors to Consider in Characterizing Receptors 30
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025 iv
-------
GLOSSARY OF KEY TERMS
The terms in this glossary were selected to ensure consistent meaning when used in these Guidelines. The
EPA recognizes that definitions continue to evolve and may be updated in future EPA resources. This
glossary was created for the Guidelines for Cumulative Risk Assessment Planning and Problem
Formulation, is not a comprehensive list of risk assessment terms, and might not represent how the terms
are defined in specific EPA programs.
Adverse outcome pathway. A conceptual framework that portrays existing knowledge about biological
events that could lead to an adverse outcome in health effects in human populations and ecosystems (U.S.
EPA, 2022a).
Agent. A chemical, physical, or biological entity that contacts a receptor (U.S. EPA, 2019).
Aggregate exposure. The sum of exposures to a single stressor from all sources by multiple routes over
multiple periods (U.S. EPA, 2019).
Conceptual model. A diagram or written description of the predicted key relationships (e.g., known,
predicted, and assumed causal relationships) between the stressor(s) (or agents) and the assessment
endpoint(s) for a risk assessment (Linder & Sexton, 2011; Suter, 1999).
Cumulative exposure. An accounting of exposures to multiple stressors and sources by multiple
pathways and routes over multiple periods (Zartarian & Schultz, 2010).
Cumulative impacts. The totality of exposures to combinations of chemical and nonchemical stressors
and their effects on health, well-being, and quality-of-life outcomes (U.S. EPA, 2022c).
Cumulative impacts assessment. The process of accounting for cumulative impacts in the context of
problem identification and decision-making. Cumulative impacts assessments consider exposures to both
chemical and nonchemical stressors at each life stage throughout the life course and apply to individuals,
geographically defined groups, or definable population groups (U.S. EPA, 2024).
Cumulative risk assessment. An analysis, characterization, and possible quantification of the combined
risks to health and/or the environment from multiple agents and/or stressors (U.S. EPA, 2003b).
Exposure. The contact between an agent and the external boundary (exposure surface) of a receptor for a
specific duration (U.S. EPA, 2019).
Environmental justice. The fair treatment and meaningful involvement of all people regardless of race,
color, national origin, or income with respect to the development, implementation, and enforcement of
environmental laws, regulations, and policies (U.S. EPA, 2010).
Exposure-response modifier. Any condition or state that could alter a receptor's (individual- or group-
level) exposure to a chemical or nonchemical agent (single or multiple) and/or could alter a physiological
response following contact with these agents.
Health impact assessment. A systematic process using an array of data sources and analytical methods
to determine the potential effects of proposals on the health of a population and the distribution of those
effects within the population. HIA provides recommendations on monitoring and managing those effects
(NRC, 2011b).
Initiating factor. A condition involving more than one chemical or agent that prompts a cumulative risk
assessment, such as (1) multiple sources/releases, (2) measured or inferred chemical concentrations, or
(3) illness in a given population (U.S. EPA, 2007a).
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
v
-------
Lifestage. Temporal stages of life that have distinct anatomical, physiological, and behavioral or
functional characteristics that contribute to potential differences in vulnerability to environmental
exposures (U.S. EPA, 2006c).
Nonchemical stressor. A stressor that is not based on chemical exposure, which could include biological
or physical factors and activities that directly or indirectly adversely affect health or increase vulnerability
to chemical stressors. The term is often used to refer to psychological or social stressors that might also
act as an exposure-response modifier to other stressors (Tulve et al., 2016; U.S. EPA, 2003b).
Receptor. Any biological entity (e.g., human, human population, lifestage within a human population)
that receives an exposure or dose (U.S. EPA, 2019).
Stressor. Any physical, chemical, biological, or psychosocial agent that can induce an adverse response.
Stakeholder. (1) Individuals or representatives from organizations or interest groups that have a strong
interest in the Agency's work and policies (U.S. EPA, 2014b) and (2) anyone who has a "stake" in a risk
assessment or risk management decisions.1
Vulnerability. Characteristics of individuals or populations that place them at increased risk of an
adverse health effect (U.S. EPA, 2019). Any conditions that increase the likelihood and/or consequences
of exposure to a stressor(s) in an identifiable group of receptors; the intrinsic predisposition or extrinsic
condition of an exposed receptor (person, community, population, ecologic entity) to suffer from external
stresses and perturbations. Vulnerability is based on variations in disease susceptibility, psychological and
social factors, exposures, and adaptive measures to anticipate and reduce future harm and to recover from
an insult (adapted from NRC, 2009).
1 Adapted from PCCRARM (1997), Vol. 1, p. 15, text for stakeholder engagement: "A stakeholder is anyone who has a 'stake' in
a risk management situation."
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
vi
-------
CEQ
CERCLA
CIA
CMG
CRA
CVD
DAG
DBP
DQO
EFSA
EPA
FQPA
HAP
HI
HIA
HQ
IOM
IPCS
NAAQS
NEJAC
NRC
OAR
OAR-TTN
OP
OPP
PCCARM
RAF
RAGS
RME
SARA
WHO
WoE
ACRONYMS AND ABBREVIATIONS
Council on Environmental Quality
Comprehensive Environmental Response, Compensation, and Liability Act
cumulative impact assessment
Common Mechanism Group
cumulative risk assessment
cardiovascular disease
Directed Acyclic Graph
disinfection by-product
data quality objective
European Food Safety Authority
U.S. Environmental Protection Agency
Food Quality Protection Act
hazardous air pollutant
hazard index
Health Impact Assessment
hazard quotient
Institute of Medicine
International Programme on Chemical Safety in the World Health Organization
National Ambient Air Quality Standards
National Environmental Justice Advisory Council
National Research Council
Office of Air and Radiation
Office of Air and Radiation-Technology Transfer Network
organophosphate
Office of Pesticide Programs
Presidential Congressional Commission on Risk Assessment and Risk Management
Risk Assessment Forum
Risk Assessment Guidance for Superfund
reasonable maximum exposure
Superfund Amendments and Reauthorization Act
World Health Organization
weight of evidence
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
vii
-------
PREFACE
Cumulative risk assessment (CRA) is an analytical approach to understanding the health effects of
exposure to multiple environmental agents and/or stressors and other stressors that could confer or
exacerbate vulnerabilities in an affected population. Risk assessment is a dynamic practice with
established procedures and methods that evolve with experience and the advance of new scientific
capability. Risk assessment informs risk management but is distinct from risk management decisions at
the U.S. Environmental Protection Agency (hereafter referred to as the "EPA"), which are guided by
statutes and regulations. To successfully use CRA, an understanding is needed of when and how the
approach can be applied. This document (hereafter referred to as the "Guidelines") describes steps for the
planning and problem formulation of CRAs and offers guidance for when such assessments might be
appropriate. It updates and supersedes the 1997 Guidance on Cumulative Risk Assessment, Part 1,
Planning and Scoping and builds on the 2003 Framework for Cumulative Risk Assessment. Emphasis is
placed on providing a uniform, yet flexible, CRA planning and problem formulation approach for risk
assessment at the EPA. These Guidelines are not prescriptive and do not impose any requirement for use
by the EPA or any federal, state, or Tribal agency.
The need to characterize human health and environmental risks posed by exposures to multiple stressors
was first articulated as EPA-wide policy in a 1997 memorandum (U.S. EPA, 1997a). The memorandum
directed "each office to take into account cumulative risk issues in scoping and planning major risk
assessments." Subsequent CRAs conducted on drinking water disinfection by-products (U.S. EPA,
2000a) and classes of pesticides (U.S. EPA, 2006e, 2006f, 2007b) illustrate major risk assessments
incorporating CRA methods. Statutes like the 1996 Food Quality Protection Act (FQPA)2 need to be
taken into consideration when preparing CRAs because they stipulate provisions that delineate assessment
requirements for certain chemicals.
These Guidelines build on the experience and knowledge of the EPA in conducting human health and
ecological risk assessments and have also been shaped by recommendations from the National Research
Council (NRC), the National Environmental Justice Advisory Council (NEJAC), and EPA's Children's
Health Protection Advisory Committee,3 among others. The Guidelines address public health and
environmental health concerns arising from sequential and multiple exposures to multiple stressors. NRC
articulates the importance of more generally incorporating CRA in EPA decision-making, citing two
primary justifications: (1) "consideration of other compounds and other factors may be necessary to
inform the decision," even if the regulatory decision of interest addresses a single chemical with a single
route of exposure; and (2) "the types of questions that are increasingly being asked of the EPA require the
tools and concepts of cumulative risk assessment" (NRC, 2009). NRC observes that CRA will provide the
EPA with a "broader and more comprehensive understanding of the complex interactions between
chemicals, humans, and the environment" (NRC, 2012).
These Guidelines lay the foundation for considering current and anticipated future cumulative risk
analytical methods. They provide recommendations to aid Agency risk assessors in developing a CRA
analysis plan but do not provide direction on which analytical methods to use for conducting CRAs.
These Guidelines are intended for use with other EPA guidelines, such as the Guidelines and
Supplementary Guidance for Assessment of Chemical Mixtures (U.S. EPA, 1986b, 2003c), the
Framework for Human Health Risk Assessment to Inform Decision Making (U.S. EPA, 2014b), and the
2 The statutory requirement for pesticide cumulative risk assessment is specifically defined by the 1996 Food Quality Protection
Act (FQPA) as a consideration of potential human health risks from all pathways of dietary and nondietary exposures to more
than one pesticide acting through a common mechanism of toxicity.
3 See two letters from J. Routt Reigart, MD, Chair of the Children's Health Protection Advisory Committee, dated September 26,
2000 ("https://www.regulations.gov/document/EPA-HO-OA-2022-0552-0Q22 "1 and August 9,2002
("https://www.regulations.gov/document/EPA-HO-OA-2022-0554-00Q6'l.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
viii
-------
Guidelines for Human Exposure Assessment (U.S. EPA, 2019). The Guidelines were prepared by senior
risk assessors and managers from across the EPA and coordinated through EPA's Risk Assessment
Forum.
These Guidelines are not prescriptive, instead offering considerations and recommendations to EPA risk
assessors regarding best practices for the CRA's initial phase of planning and problem formulation. The
risk analysis and characterization phases of the assessment are not addressed in these Guidelines and are
typically guided by the statute or authority under which they are conducted. The Guidelines do not
impose any requirement for their use by the EPA or any federal, state, or Tribal agency.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
ix
-------
RISK ASSESSMENT FORUM (RAF) TECHNICAL PANEL ON CUMULATIVE RISK
ASSESSMENT
[WRITING TEAM in bold]
Wendy O'Brien, D.V.M., Ph.D. (Chair)
Lawrence Martin, Ph.D. (retired RAF Staff)
Olivia Anderson, M.P.H. (RAF Staff)
Carolyn Acheson, Ph.D. (retired)
Katherine Anitole, Ph.D.
Bob Benson, Ph.D. (retired)
George Bollweg, Ph.D. (retired)
Carole Braverman, Ph.D.
Krista Christensen, Ph.D.
Bruce Duncan (retired)
Stiven Foster, M.S.
Audrey Galizia, Dr.P.H.
Kevin Garrahan (retired)
David Herr, Ph.D.
David Hrdy
Anna Lowit, Ph.D.
Gregory Miller, M.S.
Onyemaechi Nweke, Dr.P.H.
Marian Olsen, Dr. P.H.
Glenn Rice, Sc.D.
Jason Sacks, M.P.H.
Jane Ellen Simmons, Ph.D. (deceased)
Mark Sprenger (retired)
Cynthia Stahl, Ph.D.
J. Michael Wright, Sc.D.
EPA REVIEWERS
Daniel Axelrad (retired)
Michael Breen, Ph.D.
Chris Dockins
Kathryn Gallagher, Ph.D.
Greg Miller, M.S.
Jason Mills
Edward Ohanian, Ph.D.
Monique Perron, Ph.D.
Carolyn Persoon, Ph.D.
Solomon Pollard (retired)
Kristin Riha
Chris Sarsony
Region 8
Office of Research and Development
Office of Research and Development
Office of Research and Development
Office of Chemical Safety and Pollution Prevention
Region 8
Region 5
Region 5
Office of Research and Development
Region 10
Office of Land and Emergency Management
Office of Research and Development
Office of Research and Development
Office of Research and Development
Office of Chemical Safety and Pollution Prevention
Office of Chemical Safety and Pollution Prevention
Office of Children's Health Protection
Office of Environmental Justice and External Civil
Rights
Region 2
Office of Research and Development
Office of Research and Development
Office of Research and Development
Office of Land and Emergency Management
Region 3
Office of Research and Development
Office of Policy
Office of Research and Development
Office of Policy
Office of Water
Office of Children's Health Protection
Office of Land and Emergency Management
Office of Water
Office of Chemical Safety and Pollution Prevention
Region 5
Region 4
Office of Air and Radiation
Office of Air and Radiation
EXTERNAL PEER REVIEWERS
Nicole C. Deziel, Ph.D., M.H.S., Yale School of Public Health
Amy D. Kyle, Ph.D., M.P.H., University of California, Berkeley School of Public Health
Stephen H. Linder, Ph.D., University of Texas School of Public Health
Devon Payne-Sturges, Dr.P.H., M.P.H., M.Engr., University of Maryland School of Public Health
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
x
-------
CHAPTER 1. INTRODUCTION
Cumulative risk assessment (CRA) is an analysis, characterization, and possible quantification of the
combined risks to health or the environment from multiple agents or stressors (U.S. EPA, 2003b). In
describing CRA, we begin with a focus on the purposes for which CRA may be useful. All environmental
risk assessments endeavor to estimate a likelihood of harm arising from exposure to stressors. The
primary distinguishing factor between CRA and other risk assessments is that CRA is an assessment of
the probability of harm from exposure to multiple agents cmd/or stressors. CRA is responsive to the
EPA's mission, as well as national policy, regulations, and statutes, including several executive orders.4
Together, they direct agencies of the federal government, as appropriate and consistent with applicable
law, to address multiple and cumulative exposures, recognize effects on vulnerable populations, recognize
the value of Indigenous Knowledge, and evaluate disproportionate and adverse human health and
environmental effects (including risks).
These Guidelines describe considerations for evaluating when CRA is both suitable and feasible, as well
as steps to plan a CRA when those conditions are met, consistent with applicable law. These Guidelines
were designed to adapt to the differing and evolving needs of risk managers at EPA program offices and
regions based on differing initiating factors, statutory authority, program objectives, and availability of
data and analytical methods. The Guidelines provide a strategy that emphasizes planning the CRA to
consider the need(s) of the risk management decision. Depending on the purpose, problem, or question,
CRAs can take many different forms and may be conducted under various environmental statutes.
Examples include community-scale risk assessments, national-scale standards setting and chemical
reviews, and prioritization of actions. Therefore, no single set of criteria can be applied to determine when
a CRA is appropriate. That determination is necessarily tailored to the purpose for which a CRA is
proposed, considering statutory authority, and decided by risk assessors and decision-makers. These
Guidelines are not prescriptive, instead offering considerations and recommendations to EPA risk
assessors regarding best practices for the C R A' s initial phase of planning and problem formulation. The
risk analysis and characterization phases of the assessment are not addressed in these Guidelines and are
typically guided by the statute or authority under which they are conducted. As with other risk
assessments, a CRA should not be conflated with a risk management decision but is appropriately used as
scientifically defensible evidence to inform decisions. However, other factors also come into play in
decision-making. Depending on the situation, a CRA may not always be the most appropriate or suitable
approach. When a CRA helps address a risk management question and is within statutory mandates, it
may be appropriate to consider conducting a CRA.
The elements of CRAs discussed here are: the initiating factor, iterative scoping of the study, problem
formulation, and steps for preparation of the analysis plan (U.S. EPA, 2014b). The subsequent analysis
plan will specify the methods and data to estimate and characterize risks to the target receptor(s) from
multiple agents of concern. These Guidelines focus on human health outcomes, but many of the planning
recommendations are generally applicable to CRAs that integrate human health and ecology.5 Evaluation
of both human health and ecological endpoints is important for understanding risk for groups who may
have a greater focus on integrating ecological resources (e.g., cultural and subsistence users of ecological
resources including Tribes and Indigenous Peoples). The EPA has a policy of considering and
incorporating Indigenous Knowledge, as appropriate, per the Guidance for Federal Departments and
4 Some executive orders and memorandums include: Executive Order 12898 — Federal Actions to Address Environmental Justice
in Minority' Populations and Low-Income Populations (EOP, 1994); Executive Order 13985 —Advancing Racial Equity' and
Support for Underserved Communities Through the Federal Government (EOP, 2021); Executive Order 14091— Further
Advancing Racial Equity and Support for Underserved Communities Through the Federal Government (EOP, 2023a); Executive
Order 14096 — Revitalizing Our Nation 's Commitment to Environmental Justice for All (EOP, 2023b); Executive Order 13045 —
Protection of Children From Environmental Health Risks and Safety' Risks (EOP, 1997); Implementation of Guidance for Federal
Departments and Agencies on Indigenous Knowledge (CEQ, 2022b).
5 See EPA's web page on Ecological Risk Assessment (https://www.epa.gov/risk/ecological-risk-assessment').
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
1
-------
Agencies on Indigenous Knowledge.6 Tribes should be, and are, considered subject matter experts of their
own lifeways and environment, and their science should be considered an aspect of "best available
science."
A defining feature of CRAs, given the current state of practice, is the approach for combining exposures
to multiple stressors that can result in common adverse health outcomes. The National Research Council
(NRC), in Science and Decisions: Advancing Risk Assessment (2009), notes that many risk assessment
applications in the EPA and elsewhere "are often centered on evaluating risks associated with individual
chemicals in the context of regulatory requirements or isolated actions...NRC states that there is
concern "among stakeholder groups (especially communities affected by environmental exposure) that
such a narrow focus does not accurately capture the risks associated with the exposure(s) driving the
CRA, given simultaneous exposure to other multiple chemical and nonchemical stressors and other
factors that could influence vulnerability." This concern can potentially be addressed by CRAs, which
may be implemented when there is adequate information to identify the relevant exposure-risk
relationships for multiple chemical and nonchemical stressors, as well as the means to identify those most
important to health outcomes. This approach is consistent with the existing risk assessment paradigm that
relies upon scientifically rigorous methods for assessing hazard and exposure and the resulting adverse
outcomes. The wide range of conditions and circumstances that might result in consideration of a CRA
necessitates that planning and problem formulation guidelines be equally broad in consideration of the
factors that may be relevant to the assessment. The diversity of initiating factors may result in the call for
a CRA under various federal authorities or from other governmental entities or the public.
Because of the complexity of properly identifying and evaluating multiple agents in a CRA, a "systems
approach" for planning and problem formulation is useful. The systems approach seeks to examine "a
problem holistically, include[ing] all the drivers and stressors that affect it and the dimensions that frame
it, and integrate^] information from human health and ecological sciences and the social sciences to
formulate sustainable solutions..." (Burke et al., 2017). The systems approach Burke et al. described for
problem formulation is consistent with existing best practices and includes as examples the conceptual
model of Suter (2006) and information important to consider during problem formulation, which the EPA
describes in its Framework for Human Health Risk Assessment to Inform Decision Making (U.S. EPA,
2014b). Also important in the systems approach is engaging assessment users and stakeholders early in
the process so all useful information sources are identified and all parties fully understand the problem or
question and the system boundaries under investigation.7 The systems approach is also described in the
total environment framework (Tulve et al., 2016). An application of this framework illustrates the
interrelationships between inherent characteristics, activities, and behaviors, as well as stressors from the
built, natural, cultural, and social environments that influence children's health and well-being as they
progress through various stages of development (Barros et al., 2018a; Barros et al., 2018b; Lichtveld et
al., 2018; Nilsen et al., 2020a; Nilsen et al., 2020b; Nilsen & Tulve, 2020; Ruiz et al., 2016).
A second defining feature of CRAs is that the problem formulation can start with either the stressor(s) or
the adverse human health or environmental outcomes of interest in the receptor(s). Traditionally, EPA
risk assessments start with an examination of exposure to stressors, with the intent to estimate the effect
of such exposures on human health or the environment. However, a CRA problem formulation can also
begin with an adverse outcome (e.g., cancer, neurotoxicity) experienced by a receptor (e.g., organism,
community8) and then seek to examine exposures to possible stressors causing the adverse outcome (see
0 See Indigenous Traditional Ecological Knowledge and Federal Decision Making (CEQ, 2021); Guidance for Federal
Departments and Agencies on Indigenous Knowledge (CEQ, 2022a); Implementation of Guidance for Federal Departments and
Agencies on Indigenous Knowledge (CEQ, 2022b).
7 See EPA's web page on its draft Meaningful Engagement Policy (https://www.epa.gov/environmentaliustice/epas-meaningful-
engagement-policv).
8 Receptors, as used in these Guidelines, do not include molecular receptors and are focused on higher levels of organization.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
2
-------
APPENDIX C, Figure C-l. Conceptual Model for Factors Influencing the Risk of Cardiovascular
Disease).
CRA is one among a variety of assessment methods undertaken to inform risk and other management
decisions protective of environmental and public health. In addition to risk, other management decision
contexts may include prioritizing stressors for
attention and program design, identifying
communities disproportionately affected by
pollution and other stressors, determining
enforcement priorities, or setting regulatory
standards. Understanding the purpose of an
assessment is fundamental to its design. Clarifying
the management decision context and the
information necessary to support it, as well as data
quality, is the first step in determining whether a
CRA may be suitable, and if so, determining an
assessment design. This step is described as "fit for
purpose," in EPA's Framework for Human Health
Risk Assessment to Inform Decision Making (U.S.
EPA, 2014b) (see Figure 1). This step is common to
all assessment processes because the assessor should
understand the strengths and limitations of the
assessment frameworks potentially being used to
address the risk management decision (NRC, 2009).
Some agencies use the National Academies of
Sciences, Engineering, and Medicine (NASEM)
recommended phrase "intended purpose and context
of use" to communicate the fit for purpose concept
(NASEM, 2023). In the current practice of risk
management at the EPA, analysis relating stressor
exposures to specific effects or changes in primary or target exposure-response relationships for the CRA
is often employed to support decisions on environmental and human health standards and regulations.
These Guidelines largely conform to this practice and provide approaches for considering multiple
chemical and nonchemical exposure and response modifiers as part of CRA problem formulation and
planning. The background and history of CRA development at the EPA, as well as CRA applications, are
discussed in APPENDIX A. APPENDIX B describes key Agency documents informing CRA, with tables
highlighting contributions from the EPA (Table B-l) and other source documents (Table B-2), to improve
CRA planning and problem formulation. Together, they provide useful background for planning a CRA.
Figure 2 highlights some of the CRA publication milestones.
— —-»i -i m
i
Problem Formulation
Conceptual Analysis
Model Plan
1
Figure 1. Framework for Human Health Risk
Assessment
Source: U.S. EPA (2014b).
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
3
-------
1980 1983
1986
1994 1995 1997 2000 2002
2003 2004
2007 2009 2014
2016
EPA, Guidelines
for the Health
Risk Assessment
of Chemical
Mixtures
NRC, Risk
Assessment in the
Federal
Government:
Managing the
Process
EPA, Cumulative
Risk Assessment
Guidance-Phase
1 Planning &
Scoping
EPA, Guidance on
Cumulative Risk
Assessment of
Pesticide Chemicals
That Have a
Common
Mechanism of
Toxicity
EPA, EPA
Environmental
Justice Strategy EPA, Supplemental
Guidance for
Conducting Health
Risk Assessment of
Chemical Mixtures
NRC, Principles of
Toxicological
Interactions
Associated with
Multiple Chemical
Exposures
Executive Order 12898,
Federal Actions to
Address Environmental
Justice in Minority
Populations and Low
Income Populations
EPA, Framework
for Cumulative
Risk Assessment
NEJAC, Ensuring
Risk Reduction
in Communities
with Multiple
Stressors: EJ &
Cumulative
Risks/Impacts
EPA, Concepts,
Methods, and
Data Sources for
Cumulative Health
Risk Assessment
NRC, Science and
Decisions:
Advancing Risk
Assessment
EPA, Cumulative
Risk Webinar
Series: What We
Learned
NRC, Phthalates
and Cumulative
Risk Assessment
EPA, Pesticide Cumulative
Risk Assessment:
Framework for Screening
Analysis
Figure 2. Timeline of CRA Publication Milestones [Selected Milestones]
1.1. Early Considerations for Planning and Scoping in Cumulative Risk Assessment
Early considerations in scoping a CRA could include public health needs and the concerns of
communities or Tribes and Indigenous Peoples and statutory or other requirements; an anticipated time
frame for addressing community concerns; and risk management needs. These considerations may need to
be approached in an iterative way, allowing for possible reassessment of some aspects of the CRA
throughout the process.
Whether CRA is a suitable and feasible approach for informing a risk management decision should be
determined as early in the assessment process as possible. The determination rests on whether CRA is a
good fit for the risk management question (fit for purpose) and whether the available data, analytical
methods, and resources will support the needs of the assessment. The decision, which can involve
stakeholder input, can be considered a two-step process:
1. Suitability - Examining whether a CRA is a good match for the risk or other management
question (i.e., is it suitable? see Sections 2.3 and 2.4).
2. Feasibility - Determining whether data, analytical methods, time, and other resources are
sufficient to conduct a CRA (i.e., is it feasible? see Section 2.6).
Determining whether CRA is a suitable approach begins with considering the initiating factors and risk
management question relative to the resources and data available to address them. There may be various
initiating factors for a CRA. Potential examples include possible exposure to multiple pollutants,
chemical concentrations above levels of concern (individually or collectively), or a documented
community illness with a known or suspected connection to pollution, which have been linked to
numerous variables that can contribute to adverse health outcomes (U.S. EPA, 2007a).
If it is determined that CRA is a suitable assessment approach, several screening considerations are
recommended for evaluating whether CRA is feasible:
• Sufficient data exist or may be reasonably obtained to inform the assessment with adequate
consideration of uncertainty (see Sections 2.6 and 3.4).
• Methods are adequate to analyze the data and to integrate them into the risk characterization with
consideration of uncertainty.
• Time, staff, and financial resources are sufficient to successfully conduct the assessment.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
4
-------
The current practice of CRA seeks quantitative estimates of risk that are based on an understanding of
how multiple stressors jointly contribute to one or more common adverse health outcomes, indicated by
common mechanisms of toxicity, common key events, or converging adverse outcome pathways, and to
account for uncertainties related to the relationship between stressors and adverse outcomes. These
quantitative methods increase our understanding about toxicological action and resulting adverse health
outcomes from multiple stressors.
CRA is appropriate when the decision context is well matched to an assessment that is informed by this
kind of information. Chemical class (or group) reviews, for example, are good candidates for this kind of
analysis. Community-based CRAs may also be undertaken when the stressors of concern are well
characterized, and sufficient data can be obtained to provide for an analysis of the relationship between
stressors, exposure, and health outcomes.
Some decision contexts may present questions that extend beyond what can be answered with dose-
response stressor and exposure information. Examples include concentrated burdens of multiple and
different stressors lacking health effects data or concerns of disproportionate and adverse human health
and environmental effects among populations. There are also analytical challenges stemming from when
populations of concern are subject to less well-understood nonchemical stressors (e.g., poor nutrition or
limited access to health care) that might exacerbate an adverse response to known stressors and result in
greater vulnerability (or less resilience) than the general population. In these cases, there may be
uncertainty about which stressors are important and how to incorporate them into a risk analysis. A
qualitative or semiquantitative characterization of suspected stressors that lack data needed for
quantitative risk analysis can nonetheless improve understanding of the full range of potential stressors
contributing to adverse health outcomes. An evaluation of when and how to incorporate them into an
assessment is case dependent and should follow a determination by the CRA team using an appropriate
weight-of evidence (WoE) evaluation (see WoE discussion in Section 3.2.4).
To ensure CRA is an appropriate fit for purpose tool, it is important to understand when and how it can
be productively applied relative to other approaches, including but not limited to cumulative impact
assessment (CIA) and health impact assessment (HIA). Each approach may employ a variety of methods
to provide information to decision-makers about health outcomes associated with exposures to multiple
chemical and nonchemical stressors. Methods employed are not exclusive to an assessment approach and
can be used across different approaches. Determinations about which approaches and analytical methods
to use depend on factors such as statutory requirements, the scope of an assessment, types of data needed
and available, and applicability for the evaluation and needs of the decision-maker. Risk assessment is
specifically used to inform risk management decisions. However, any of these approaches could be used
to help address environmental management questions related to disproportionate and adverse human
health and environmental effects, community action plans or other community-based needs, and other
environmental management considerations if the analytical method is warranted given the decision-
making context. More than one approach may be used in combination. Assessors need to exercise
judgment in determining when to use CRA, CIA, or another approach for evaluating exposures to
multiple stressors for a specific purpose. Assessors can refer to EPA's Interim Framework for Advancing
Consideration of Cumulative Impacts for additional information on the matter (U.S. EPA, 2024).9
Risk assessments have traditionally been conducted on specific stressors with known adverse health
outcomes. CRA may be planned around stressors but can also focus on receptors and adverse outcomes as
the risk management question. NRC's report on phthalates (NRC, 2008a) and the European Union's
NoMiracle project (Lokke, 2010) illustrate the use of CRA to evaluate receptors of concern. A major
9 See EPA's website for the Interim Framework for Advancing Consideration of Cumulative Impacts:
https://www.epa.gov/cumulative-impacts/interim-framework-advancing-consideration-cumulative-impacts.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
5
-------
conclusion of the NoMiracle project highlighted the importance of CRA when focused on receptors: "... it
gradually became clear that the current approach in coping with chemical mixtures focusing on 'the
chemical' and 'chemical cocktails' should be replaced by a focus on the biological organismal receptor,
e.g., on the organism (humans or other species), the population or the ecosystem being exposed to a more
definite cocktail of stressors" (Lokke. 2010). Similarly, NRC's Phthalates and Cumulative Risk
Assessment (NRC, 2008a) report was initiated because "studies show widespread human exposure to
multiple phthalates and indicate that effects on the development of the reproductive system of laboratory
animals occur at much lower doses than were predicted in earlier studies." In this context, the EPA asked
NRC to review independently the health effects of phthalates.
As the practice of CRA continues to evolve, an increasing number of toxicological and epidemiological
studies link nonchemical stressors to population vulnerabilities in many contexts that may be relevant to a
risk management decision (Payne-Sturges et al., 2018). When there is evidence of a relationship between
the stressor and the health outcome, that evidence can be incorporated into the assessment in multiple (or
a variety of) ways (e.g., when the suspected stressors demonstrate stressor estimates of risk are difficult to
quantify but show other qualitative or semiquantitative relationships). These Guidelines provide
recommendations for documenting quantitative and qualitative information in the risk characterization
(see Sections 2.6 and 3.5). Additionally, methods for merging quantitative and qualitative information are
discussed in the Framework for Cumulative Risk Assessment (U.S. EPA, 2003b) and elsewhere in peer-
reviewed literature (e.g., Schafer et al., 2023).
1.2. Organization of This Document
These Guidelines provide considerations and strategies for CRA planning and problem formulation,
which are necessary steps in the advancement of CRA from a concept to a decision-relevant tool (Sexton,
2015). These Guidelines comprise three sections: this introduction, a second section on CRA planning
and scoping, and a third section on CRA problem formulation. The scoping step in the planning phase
identifies the significant factors in the assessment and should determine whether CRA is the right
approach for conducting the assessment. The steps for scoping the assessment presume a risk
management question suited to CRA. The problem formulation section describes steps for developing the
conceptual model that provides an initial understanding of relationships among factors in the assessment.
The problem formulation statement provides a precise description of how the proposed risk assessment
serves the risk management decision. Problem formulation results in an analysis plan, which describes the
methods necessary to conduct the analysis.
Although steps are defined sequentially for ease of communication, the process can involve simultaneous
steps or a different sequence. Steps also might be repeated in an iterative approach. The scoping and
problem formulation phases are mutually informative, as greater detail and more information advance the
assessment. Factors important to consider are listed in many of the sections. Strategic questions
associated with some sections are highlighted in shaded boxes.
These Guidelines recommend a tiered or phased process for matching the design of a CRA to the level of
risk management needed (see Section 2.5). The intent of tiering or phasing the analysis is to tailor the
level of effort to the purpose of the risk assessment. The incremental process of gathering information
typical of tiering and phasing reinforces the recursive process between problem formulation and the other
steps in planning the CRA.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
6
-------
Text Box 1.
Cumulative Risk Assessment Planning Milestones
1. Initiating factors (Section 2.1.)
2. Identification of stakeholders (Section 2.2.)
3. Statement of purpose (Section 2.3.)
4. Evaluation of "fit for purpose" (Section 2.4.)
5. Scoping summary statement (Section 2.4.)
6. C onceptual model (Section 3.2.)
7. Weight of evidence evaluation (Section 3.2.4.)
8. Analysis plan (Section 3.3.)
These Guidelines identify eight milestones in
planning the CRA (Text Box 1). Each
milestone is an appropriate point at which the
CRA team may confirm its planning progress
and consider whether any changes would be
appropriate (perhaps on the basis of new
information).
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
7
-------
CHAPTER 2. CUMULATIVE RISK ASSESSMENT
PLANNING AND SCOPING
CRA design involves defining and preparing a risk-related problem for evaluation. CRA planning and
scoping generate the initial definition of the problem for assessment. Planning and scoping help to
organize issues and participants in advance of problem formulation and the development of a technically
oriented conceptual model and the preliminary analysis plan for CRA.
Problem formulation and development of a conceptual model and preliminary analysis plan for CRA are
considered parts of the planning process but are organized separately in these Guidelines (Chapter 3). In
practice, however, problem formulation occurs iteratively with other planning and scoping considerations.
This chapter discusses eight topics for CRA planning and scoping:
• Decision context and initiating factors (Section 2.1.)
• Participant and stakeholder involvement (Section 2.2.)
• Statement of purpose (Section 2.3.)
• Scoping objectives, boundaries, and constraints (Section 2.4.)
• Tiering and phasing the assessment (Section 2.5.)
• Data quality, needs, and availability (Section 2.6.)
• Project and risk management considerations (Section 2.7.)
• Peer review (Section 2.8.)
The approach to CRA planning in these Guidelines is consistent with the vision for risk assessment that
NRC's Committee on the Institutional Means for Assessment of Risks to Public Health advances in its
report, Risk Assessment in the Federal Government: Managing the Process (NRC, 1983).
2.1. Decision Context and Initiating Factors
Risk assessments, including CRAs, are generally performed within a decision-making context to inform
regulatory actions, to address a health concern, or to comply with legal requirements or align with broad
EPA priorities, such as children's health, environmental justice,1" and sustainability (EOP, 1997; U.S.
EPA, 2014b 2021a). This context and the information or technical factors leading to a decision to
consider CRA are termed "initiating factors." Identification of CRA initiating factors is generally the
beginning of the CRA planning phase. Consistent with explanations in EPA's Framework for Human
Health Risk Assessment (U.S. EPA, 2014b) and NRC's Science and Decisions (NRC, 2009), initiating
factors provide the rationale for CRA. In addition to supporting EPA regulatory requirements and policy
commitments, the decision to conduct a CRA may stem from the input of stakeholders, including
government and Tribal representatives, academia, industry, and concerned citizens or organizations.
Initiating factors can relate to environmental justice concerns, specific stressors (sources), or population
exposures (receptors); they can also be specific to a location or situation. Initiating factors help define the
assessment in terms of geographic area(s) and population(s) or receptor(s) to be evaluated, relevant
stressors and effects, and the time frame in which results are needed. Clarity on initiating factors is
helpful for determining whether CRA will be a suitable approach to inform the management decision.
However, the variability of initiating factors and decision-making contexts does not allow for a definitive
111 Environmental justice means the just treatment and meaningful involvement of all people, regardless of income, race, color,
national origin, Tribal affiliation, or disability, in agency decision-making and other Federal activities that affect human health
and the environment so that people: are fully protected from disproportionate and adverse human health and environmental
effects (including risks) and hazards, including those related to climate change, the cumulative impacts of environmental and
other burdens, and the legacy of racism or other structural or systemic barriers; and have equitable access to a healthy,
sustainable, and resilient environment in which to live, play, work, learn, grow, worship, and engage in cultural and subsistence
practices. See the EPA web page on Environmental Justice definitions for more information:
https://www.epa.gOv/environmentaliustice/leam-about-environmental-iustice#definitions.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
8
-------
decision criterion to initiate a CRA. A determination to initiate a CRA builds on early planning steps and
an evaluation of the match of available information for risk analysis to inform the decision context.
Examples of initiating factors include one or more of the following:
Statutory Provisions. Statutes might require or pertain to CRA. For example, the Food Quality Protection
Act (FQPA) of 199611 specifies CRA for certain pesticide assessments.
Community Concern. Communities can play a role in initiating CRA by identifying ecological or human
health conditions of concern, such as exposure or vulnerabilities from multiple factors (e.g., nonchemical
stressors, age) and exposure to combined (local or national) multiple pollutant sources or releases not
addressed by single-pollutant or single-source assessments.
Future Changes in the Community/Environment. Permitting. Upfront Assessment. Anticipated changes in
a community or environment might be a precondition for a project or activity (e.g., environmental review
requirements associated with rezoning proposals).
Evidence of Human Illness/Ecosystem Stress. Increased or elevated prevalence of human illness (or the
observation of personal or local increases) in a community or an ecosystem could motivate consideration
of CRA.12 Available data supporting multiple chemical exposures leading to specific disease outcomes
could also motivate CRA. An example is highlighted in an NRC report stating: "... EPA could evaluate
combined exposures to lead, methylmercury, and polychlorinated biphenyls because all contribute to the
cumulative risk of cognitive deficits associated with IQ reductions in children, although the deficits are
produced by different mechanisms of action.../' (NRC, 2008a).
Elevated Stressor Levels. Measured or potential elevations in pollutant concentrations could motivate
CRA (e.g., metrics such as those included in EPA's Report on the Environment (U.S. EPA, 2020c) or
America's Children and the Environment (U.S. EPA, 2013)).13 Research suggests an association between
proximal human exposures to chemicals and the production of high-volume chemicals and accidental
releases and that "where chemical production volumes are so high (i.e., they are produced or imported
into the U.S. in quantities of 500 tons per year or greater)... human exposures should be expected'' and
should "trigger additional scrutiny and potential interventions" (Vandenberg et al., 2023).
Evidence of Widespread Exposure to Chemical(s) of Concern in a Population. NRC recommends
conducting a CRA when there is substantial evidence supporting multiple chemical exposures in the
human population (e.g., biomonitoring data) (NRC, 2008a). In some instances, coordination with other
entities (e.g., other federal agencies, state and local government, other organizations) may be appropriate.
Social and Economic Concerns about Natural Resource Conservation. Environmental conservation
concerns, such as sustaining cultural and subsistence practices, could lead to requests for CRA.
Useful descriptions of CRA initiating factor(s) are concise statements of what the initiating factor(s) are:
how, when, and by whom they were brought forward for evaluation; human or ecological
population(s)/receptor(s) potentially affected; any specific contaminants or other stressors of concern;
temporal scales and geographic areas of concern; and anticipated environmental impacts or health effects
that might need evaluation.
11 See the EPA's Summary of the Food Quality Protection Act fhttps://www.epa.gov/laws-regulations/summarv-food-aualitv-
protection-act).
12 Community concerns that environmental exposures are contributing to illness and that single-agent and/or single-source
assessments do not adequately reflect risks from the combination of stressors and sources (as well as vulnerabilities) are
conditions that can lead to a community's request for a CRA.
13 Updates to America's Children and the Environment occurred in 2015, 2017,2018, and 2019. See
https://www.epa.gov/americaschildrenenvironment/basic-infonnation-about-ace.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
9
-------
2.2. Participant and Stakeholder Involvement
The EPA's Framework for Human Health Risk Assessment to Inform Decision Making defines
stakeholders as .. individuals or representatives from organizations or interest groups that have a strong
interest in the Agency's work and policies." It also identifies internal stakeholders, such as EPA offices
and external stakeholders, which can include the public, affected industries, public health or
environmental organizations, and other governmental agencies such as Tribes, states, and municipalities
(U.S. EPA, 2014b).
The Presidential/Congressional Commission on Risk Assessment and Risk Management (PCCRARM)
uses another definition also worth noting: "A stakeholder is anyone who has a 'stake" in a risk
management situation" (PCCRARM, 1997). This broad stakeholder definition recognizes that even those
unaware of risk assessment/risk management activities such as CRA might have a "stake" in or be
affected by such work. The Commission considered the "Problem/Context" stage of its process—roughly
equivalent to planning, scoping, and problem formulation in CRA—to be crucial. The Commission
emphasized "active stakeholder involvement at this particular stage [as] the most critical element of the
decision-making process." Determining who is "at the table" during CRA design is a consideration that
has consequences for all subsequent steps in the CRA process.
The levels of public participation, stakeholder negotiation, community outreach, Tribal consultation,14
and capacity building vary depending on the activity and program requirements (U.S. EPA, 2001b,
2014b). Applicable laws, regulations, and the domain of other agencies affect what may and may not be
done in participatory processes and how agencies with authority to act may use the results of such
processes. Even when process flexibility is constrained by legal mandates, the CRA team can have a
robust discussion of stakeholder participation and involvement (NRC, 2008b). When engaging with
federally recognized Tribes, the EPA uses the Tribal consultation process to provide an opportunity for
meaningful dialogue and input when EPA actions or decisions may affect Tribes (U.S. EPA, 2023c).15
The range of circumstances motivating or initiating CRA can be broad, and stakeholder involvement can
be greatly determined by initiating factor(s). For example, stakeholders concerned about local
environmental conditions could seek to initiate a CRA process. In such a case, stakeholder involvement
and participation can influence EPA's risk assessment or CRA design; however, the extent and timing of
such involvement can vary.16 As stated in the Agency's Framework for Human Health Risk Assessment to
Inform Decision Making (U.S. EPA, 2014b), the appropriate stakeholder involvement process will depend
on the specifics of the situation: "... The timing, frequency and level of community involvement will
depend on several factors, including regulatory requirements, the nature of the decision and community
interest." Public, stakeholder, and community involvement, when appropriate, should be considered part
of the risk assessment and decision-making processes. A key principle of community engagement is to be
clear about the purposes or goals of the engagement and the populations or communities to be engaged
(ATSDR, 2011). Other examples of how to engage communities and public stakeholders can be found in
Appendix2 of the EPA's Meaningful Involvement Policy^1 and in the EPA's Stakeholder Involvement
14 https: //www, epa. go v/tribal/consultation-tribes
15 Tribal consultation is a process to ensure meaningful and timely input by Tribal officials. In addition, the Agency's network of
Tribal Partnership Groups ("https: //www, epa. go v/tribal/partnerships-tribes) facilitates the exchange of technical information and
communication between Tribes and the EPA. For example, the National EPA-Tribal Science Council works to integrate and
increase tribal involvement in the EPA's scientific activities, while the National Tribal Toxics Council provides Tribal input on
issues related to toxic chemicals and pollution prevention.
16 A request for a CRA can be made by anyone. However, for the EPA to initiate a CRA in response, it would need to conform
with statutory or program guidelines and procedures and have management approval. While this document is written in particular
for EPA use and application under regulatory authorities, the principles may be useful to other authorities, such as Tribal, state,
and local governments.
17 See the EPA's web page on its draft Meaningful Engagement Policy fhttpsV/www.epa.gov/environmentaliustice/epas-
meaningful-engagement-policv'l.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
10
-------
Public Participation at the U.S. EPA: Lessons Learned, Barriers, & Innovative Approaches (U.S. EPA,
200Id). Key questions and considerations may include:
• Who are the stakeholders?
• What and when are the opportunities for stakeholder involvement?
• What communication materials are needed to effectively engage stakeholders?
• What mechanisms, including data-sharing agreements, will be most effective for engaging
different stakeholders?
Those conducting the CRA (e.g., the EPA, contractors, subject matter experts) can be called CRA
participants or the CRA team. The composition of the CRA team depends on the nature of the problem.
The level of complexity of a risk assessment and the context for its conduct often dictate a
multidisciplinary approach. Legal experts, Tribal program managers, and policymakers also might be
called on to contribute to risk assessment planning and scoping. Depending on the context and the process
by which the risk assessment is conducted, specific expertise might be needed to develop specialty tools,
data, or analyses. Other team participants can provide information on project management issues,
including funding levels and sources, staffing, contractor requirements, and any need for interagency
agreements (U.S. EPA, 2003c).
2.3. Statement of Purpose
Following identification of initiating factors
and stakeholders/participants, a purpose
statement (see Text Box 218) should be
confirmed with all involved parties.
Depending on the context of the assessment,
this statement could be a formal or informal
step, but written documentation is
recommended. A statement of purpose could
also take the form of a relevant framework for
a conceptual model, which also could become
an important part of the analysis plan. In
general, the purpose should be stated in
concise and direct terms, clearly identifying
what the CRA is intended to accomplish or
produce. This statement, which may be
written as a specific risk question, is used to
identify what will be needed to accomplish the
desired outcome or product of the CRA. Topics such as resource and data availability, CRA scope, and
risk management options are relevant to the CRA purpose statement and should be discussed before it is
completed. Because the statement of purpose provides direction for subsequent steps in the CRA and
likely is referred to repeatedly, the discussion should be as precise as possible and clearly understood. At
a minimum, the statement should be expected to clarify the initiating factor(s) and the risk management
decision the CRA is intended to inform. It can be sufficiently detailed to provide an initial framework for
construction of a conceptual model. This step provides an initial indication of the suitability and
feasibility of the CRA and the extent to which it can address the risk management decision.
18 Tribal lifeways are inclusive of, but not limited to, economic, cultural, ceremonial, recreational, and subsistence practices. It is
characterized by an EPA Partnership Group, the National Tribal Toxics Council.
https://nttc.sfo3.cdn.digitaloceanspaces.com/Docs/NTTC-Understanding Tribal Exposures to Toxics-2015-06-19.pdf
Text Box 2.
Hypothetical Cumulative Risk Assessment
Statement of Purpose
A Tribe experiences ongoing air and water pollution released
from two nearby facilities. The purpose of a cumulative risk
assessment would be to evaluate whether these exposures are
affecting the health of the community through estimating
emission rates of air and water pollutants from the two facilities.
As part of this assessment, potential concerns that would be
particularly relevant for the Tribe could be considered:
contaminated fish, which makes up half of their normal dietary
protein intake; contaminated water, which is not just a dietary
intake exposure but used for cultural practices such as steam
baths and sweat lodges; and higher rates of asthma or respiratory
issues among Tribal children. Additional consideration could
also include nonchemical stressors. If the Tribe agrees to share
further information pertaining to unique lifeways, this could also
be considered in the evaluation.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
11
-------
2.4. Scoping Cumulative Risk Assessment Objectives, Boundaries, and Constraints
Scoping of a risk assessment's general objectives, boundaries, and constraints provides a foundation for
problem formulation (U.S. EPA, 2014b). Scoping represents the CRA team's initial "big picture," first-
draft statement of the key considerations, topics, or elements the assessment will address and begins with
consideration of whether the CRA is fit for purpose—the suitability and feasibility of CRA to address the
risk management decision. Scoping incorporates as many issues and concerns as the team believes
relevant to the assessment's success. It describes the risk problem and how the CRA will address and
inform that problem. In scoping the assessment, decision managers, stakeholders, and relevant technical
experts work as a team informed by stakeholder input to discuss key considerations of the subject CRA.
Because the scope of a CRA may cut across the legal domains of multiple statutes and regulatory
agencies, it can involve issues and topics that are outside the expertise of the EPA risk assessors and
regulators (e.g., HIPAA) (Sexton, 2015). The CRA team should seek appropriate additional expertise as
necessary. Scoping often concludes with a scoping summary statement.
Fit for Purpose. A well-designed scoping process ensures the CRA will meet the intended purpose(s) as
specified in the statement of purpose and will inform decision-making. As discussed in Section 1.1, the
evaluation early in the assessment planning process of whether a CRA is a suitable and feasible
assessment approach that will inform its purpose is referred to as fit for purpose. The EPA introduced the
concept of fit for purpose for human health risk assessment to help ensure assessment products were
properly designed to inform the risk management decisions for which they were conducted (U.S. EPA,
2014b). Consideration of fit for purpose is an articulation of the risk management needs early in the
scoping process and the questions to be informed by the risk assessment. Attention is given to this
concept through focused planning and problem formulation, and its confirmation throughout the process
to ensure the assessment is generating the information that will inform the risk management decision. An
initial evaluation of fit for purpose follows the statement of purpose and constitutes a determination of
whether a CRA can inform a risk management decision.
Overarching questions when evaluating the CRA's suitability and feasibility using a fit for purpose
approach—the answers to which will improve scientific understanding of the problem—include:
• How will the assessment inform choices among risk management options?
• Will the risk assessment need to change or expand to differentiate among risk management
options?
• Does the risk assessment design meet the objectives, and will it have the attributes identified in
the problem formulation step?
• Does the assessment address the initial objectives, and is it consistent with the attributes
identified in problem formulation? Or, if the initially identified objectives or attributes have been
modified, does the assessment incorporate the modifications?
• If the assessment requires peer review, is the review consistent with the current EPA peer-review
policy (e.g., Peer Review Handbook, 4th Edition), and have the issues raised during peer review
been addressed adequately?
• How will the results of the risk assessment be communicated to the risk managers and
stakeholders (from U.S. EPA, 2014b)?
Other important scoping considerations in this section (below and in Table 1) may also guide the
evaluation of whether the CRA is fit for purpose.
Potential Statutory and Regulatory Provisions. Risk assessment could be affected by statutory or
regulatory mandates administered by the EPA or local, state, Tribal, and other federal agencies. Such
mandates can affect how the CRA is conducted (e.g., by constraining the assessment's scope or the way
appropriated funds can be spent). The CRA team should refer to the relevant EPA authority other federal
agency, or Tribal, state, and local authorities for statute and regulation information.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
12
-------
Supportive Policy/Executive Orders. Relevant scientific background and reasoning for a CRA can be
noted in some relevant policies and directives, which can be considered in planning and scoping any risk
assessment, including a CRA. For example, the Policy on Children's Health (U.S. EPA, 2021a) describes
the EPA's policy to protect children from environmental exposures by consistently and explicitly
considering early-life exposures and lifelong health in all human health decisions.19 More policies and
executive orders can be relevant to supporting CRA planning and scoping.20
History of the Issuc(s). The history of a particular environmental issue might be extensive and need
documentation to set context and boundaries for the CRA. Identifying stakeholders who have been
previously concerned or involved or communities that are directly affected by the issue could be
important. Event history and potential legacy pollution might be useful information in describing the
initiating factor(s), assembling a planning team, and obtaining and organizing information for the CRA.
The contributing pollutants, regulatory history (existing laws and regulations), and assessment history
(conducted by other EPA programs and other organizations) are also important to identify and consider.
Populations of Interest. As part of CRA scoping, characteristics of any population of interest in the CRA
need to be defined and would include identification of worker/occupational nonusers, consumers and
bystanders, fenceline communities, Tribal and Indigenous Peoples, and other populations of interest.
Within each population of interest could be individuals with increased exposures or vulnerabilities, which
could result in increased consequences from exposure.
It is important for a CRA to consider groups of individuals or communities within the general identified
study population who, due to vulnerabilities, may be at greater risk than the general population of adverse
health effects from exposure to stressors. Vulnerability can stem from intrinsic factors (e.g., preexisting
disease, lifestage, reproductive status, age, sex, genetic traits) or extrinsic factors (e.g., social
determinants of health such as food insecurity, geography, poverty, socioeconomic status, racism,
discrimination, cultural and subsistence practices, workplace) when identifying subpopulations of
concern.21 A goal of the Agency is to address age- and sex-specific issues, using age- and sex-
differentiated data in Agency risk assessments and risk management decisions, whenever appropriate, and
when relevant information is available (U.S. EPA, 1995b, 1997c, 2021a).
Differences in vulnerability may be explained by an analysis of toxicokinetic or toxicodynamic
differences across lifestages or populations (e.g., across humans and animals, across animal species,
across sexes or lifestages). Individual and social factors that may increase vulnerability to exposure-
19 This Policy recognizes that such considerations may be qualitative when data are unavailable to support quantitative
assessment. It states in part: "In implementing this policy, the EPA will identify and integrate data to conduct risk assessments of
children's health to inform decisions. To the extent that relevant data is available, a quantitative risk assessment will be
conducted. When quantitative information is not available but risks to children may exist, a qualitative risk assessment will be
performed. In certain circumstances, assessment of aggregate and cumulative exposures may be necessary to properly
characterize risks to children."
20 Another recent policy example that may be relevant to CRA planning and scoping is Executive Order 13985 on Advancing
Racial Equity and Support for Underserved Communities Through the Federal Government, which directs all agencies of the
federal government to "pursue a comprehensive approach to advancing equity for all, including people of color and others who
have been historically underserved, marginalized, and adversely affected by persistent poverty and inequality." A third example
of a major policy directive to consider is Executive Order 14096, which requires the EPA to specifically "take steps to address
disproportionate and adverse human health and environmental effects (including risks) and hazards unrelated to federal activities,
including those related to climate change and cumulative impacts of environmental and other burdens on communities with
environmental justice concerns."
21 EPA programs and some regions rely on the Integrated Risk Information System (IRIS) for toxicity values for human health.
IRIS noncancer reference values include the application of uncertainty factors that address interindividual differences in
variability and susceptibility (e.g., addressing differences in response due to lifestage, sex, or genetic predisposition). Similarly,
some IRIS toxicity values for cancer recommend the application of age-dependent adjustment factors to address an assumption of
increased early-life susceptibility when there is sufficient support for a mutagenic mode of action. Such decisions are dependent
on the chemical-specific evidence available when the assessment was developed.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
13
-------
related health effects include, genetic or epigenetic variability, age or developmental lifestage, health
status, behaviors or practices, and social determinants of health (see Table 1 and APPENDIX D).
Empirical data supporting such vulnerabilities may be absent, and incorporation of these considerations
may be dependent on modeling.
Nonchemical stressors are an umbrella category of factors that can independently influence the risk of
adverse health effects or may do so in combination with another stressor(s). Nonchemical stressors may
directly affect physiological stress or contribute to physiological dysregulation and damage, which can be
measured as allostatic load.22 For CRA, nonchemical stressors are most likely to be considered for their
role as potential exposure-response modifiers to chemical stressors (see Section 3.5). Limited experience
with incorporating nonchemical stressors (e.g., allostatic load) in risk analysis may challenge the CRA
team to identify appropriate methods consistent with data quality guidelines for the assessment.
Incorporating nonchemical stressors in a CRA can be complicated by uncertainties in evaluating exposure
to them, and their evaluation, potential health consequences, potential modification of the effects of or by
other stressors (e.g., chemical exposures) and variation in individual (or community-scaled)
vulnerabilities. When a lack of methods for assessing and quantifying such stressors limits their
incorporation into an analysis plan, the conceptual model can be used to flag them for further study or
note that any relevant qualitative information be included in the risk characterization for consideration by
risk managers. Methods to incorporate such information qualitatively or quantitatively (when possible)
may be available in peer-reviewed literature and could be considered during CRA scoping. The EPA may
develop additional guidance to incorporate such information in the future.
Output from CRA scoping includes a documentation of findings (e.g., a scoping summary statement) that
outlines general objectives, constraints, and boundaries of the CRA. Such documentation allows a
common understanding of the scoping results across the team and will provide the foundation upon which
further planning and problem formulation proceeds. The CRA scoping output also formalizes the earlier
determination of whether CRA is a suitable approach for informing the risk management decision in the
fit for purpose determination. Following this related discussion, the CRA team determines whether the
available information supports continuation of the CRA with possible consideration of tiering or phasing
(Moretto et al., 2017; U.S. EPA, 2016b). This phase is similar to the "gatekeeper step" implemented prior
to problem formulation described and recommended by Moretto et al. (2017). The purpose of this step is
to ensure all chemicals of potential concern are identified and grouped so the likelihood of common
toxicity and co-exposure can be given initial evaluation by the CRA team and a determination made
whether further effort for a CRA is appropriate (Embry et al., 2014; Moretto et al., 2017). The initial
evaluation can vary in scope and level of detail based on the project. If a CRA is determined to be
unsuitable, an explanation for the record of why the CRA is not considered appropriate to inform the risk
or other management decision should be prepared and communicated to the risk manager or other
decision-maker. A CRA might not be appropriate if there is not enough information available on stressors
that have been identified by the CRA team. Similarly, if the scoping indicates there is no potential for
adverse health risk, a full CRA may not be necessary (see Section 2.5).
Table 1 summarizes possible scoping considerations discussed throughout this document. Some of the
examples listed are aspirational in nature and may not be feasible within current CRA practice. The CRA
team should follow the data quality objective (DQO) process when scoping a CRA (see Section 2.6).
Table 1. Possible Considerations for Cumulative Risk Assessment
This table is not intended to be a checklist but provides possible considerations during the problem formulation of a CRA.
22 Allostatic load represents the physiological consequences of chronic exposure to fluctuating or heightened neural or
neuroendocrine response that results from repeated or chronic stress (Juster et al., 2010). Cardiovascular, metabolic, and
inflammatory responses are also specified (Rodriquez et al., 2019).
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
14
-------
Overall
Considerations
Description or Example
Relevant
Subsection in the
Guidelines
Overall CRA
purpose
The CRA purpose describes objectives, what is being evaluated, and limits of the
CRA. See more in the CRA statement of purpose (2.3)
Section 2.3.
General scope of
the assessment
Determine whether it is stressor- or receptor-oriented
Identify geographic scale, population(s) of interest, and temporal boundaries
Identify hazards/health outcomes and endpoints
Section 2.4.
Potential
stressors and
health outcomes
evaluated in the
assessment
Multiple stressors (i.e., chemical exposures, nonchemical stressors)
Dose-response factors and related issues (e.g., modes/mechanisms of action and
adverse outcome pathways)
Exposure routes and pathways, including spatial, temporal, "background"
exposures, pollutant modeling, or monitoring
Section 3.2.
(conceptual
model)
Potential Case-
Specific
Considerations
for CRA
Description or Example
Relevant
Subsection in the
Guidelines
Biology and
genetic
variability
Different population characteristics based on genetic differences that may affect
response to stressors (including epigenetics and genetic diseases), sex, anatomical
and physiological traits, lifestage (age or life course (e.g., women of childbearing
age, infants, fetuses, children, people >age 65+), in utero, adolescence, or
pregnancy), health status (preexisting conditions or disease, such as psychosocial
stress, elevated body mass index, frailty, nutritional status (e.g., food deserts),
chronic disease)
Section 2.4.
(Population of
Interest) and
3.2.2.
Lifestyle and
cultural practices
Diet, smoking, alcohol consumption, pica and other mouthing behaviors in children,
subsistence, or recreational hunting and fishing
Section 2.4.
(Population of
Interest) and
3.2.2.
Social
determinants of
health-related
and
vulnerabilities
Social determinants of health include economic stability (e.g., steady income),
education access and equality, health care access and equality, neighborhood and
built environment (e.g., housing conditions conducive to flooding, pests, or
allergens; housing status; access to transportation), social and community context
(e.g., social, economic, and political inequality), cultural and subsistence practices,
education, nutritional status, environmental justice factors (e.g., communities with
potentially disproportionately high exposures to stressors), legacy pollution, or other
conditions contributing to potential and disproportionate vulnerabilities in the
potentially affected population23
Examples of indicators for social determinants of health include educational status,
income status, occupation, race/ethnicity, and geography (e.g., urban/rural)
Social determinants of health could have direct impacts on health outcomes of
interest or act as exposure-response modifiers and could therefore affect the
conceptual model or analysis plan
Section 3.2.2.
(Receptors of
Potential Interest)
and Appendix D
Analyses specific
to an assessment
Topics of potential interest and accompanying rationale (e.g., effects of climate
change)
Sensitivity analyses
Previous or targeted uncertainty and variability analyses
Analytical resources required or available, such as data or models
Subsequent conceptual model and analysis plan development
Tribal lifeways (inclusive of, but not limited to, economic, cultural, ceremonial,
recreational, and subsistence practices)24
Consideration of internal/external peer review
Relationships among potential assessment endpoints and risk management options
Various sections
(e.g., Section 2.8.,
Section 3.4.)
23 Adapted from the Office of Disease Prevention and Health Promotion's Social Determinants of Health web page
(https://health.gov/healthvpeople/prioritv-areas/social-determinants-health'l.
24 Examples of Tribal lifeways and risk assessment can be found in "Paper on Tribal Issues Related to Tribal Traditional
Lifeways, Risk Assessment, and Health & Well Being: Documenting What We've Heard" (National EPA-TSC, 2006).
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
15
-------
Potential Case-
Specific
Considerations
for CRA
Description or Example
Relevant
Subsection in the
Guidelines
Risk assessment products (quantitative and qualitative) for risk management
decision-making
Economic analyses
Alternative assessment analyses
Project
management
Source(s) of funding
Availability of CRA team members
Legal constraints
Other resource and data limitations
Schedule for completion
Community/stakeholder consultation board
Data sharing agreements
Section 2.7.
Adapted from U.S. EPA (1997c. 2014b).
2.5. Tiering and Phasing the Assessment
Tiering and phasing are complementary processes that can contribute to a flexible CRA design and may
be employed to assist with planning and scoping the assessment. They are optional strategies that can help
"to balance resources against the desire to reduce uncertainty in the assessments" (Menzie et al., 2007).
Both tiering and phasing are iterative process considerations that begin with planning and involve a
sequential execution of the analysis plan. Tiering is focused on a stepwise process to evaluate risk and the
adequacy of data to meet the purpose of the CRA, whereas phasing is focused on the identification and
prioritization of stressors to address the risk management decision. Both processes use an iterative,
stepwise methodology to prioritize stressors and exposures and to determine an appropriate level of
analysis commensurate with the risk decision being informed (Meek et al., 2011). Tiering and phasing are
complementary, as they consider different aspects of how the iterative inquiry can be conducted.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
16
-------
Text Box 3.
Tiers for Cumulative Risk Assessment Analysis
Tier 0 - This tier is advisable when screening-level assumptions of exposure and hazard indicate very low or no risk, and
simple semiquantitative estimates of summed combined exposures for the stressor group may provide sufficient analysis.
Semiquantitative estimates are based on limited data and a few very simple conservative (i.e., more health-protective)
assumptions. Any information on the stressor group, such as biological mode of action and estimates of exposure, would be
compared with more robust quantitative estimates for chemicals with similar profiles to provide a crude quantified estimate
of risk. With the use of conservative scenarios if the result is no or low risk, Tier 0 may provide sufficient risk
characterization. Alternatively, further assessment at the Tier 1, 2, or 3 level may be indicated.
Tier 1 - This tier also provides an estimate typically based on conservative scenarios (i.e., more health-protective), layering
on additional information about the stressor group, such as deterministic exposure estimates for all components of that group
(this can be based on measured or modeled data). Results might suffice for comparison with a measure of hazard to
determine whether further assessment is necessary.
Tier 2 - In this tier, deterministic estimation of exposure* is refined by incorporating increasing numbers of measured
values. Exposure scenarios will be more tailored to the specific situation to provide better detail. Models might incorporate
additional parameters, and although estimates are still considered conservative (i.e., more health-protective), they are
believed to be more realistic, incorporating more data. Multiple sources often are taken into account by summation.
Tier 3 - This tier is the most thorough and will likely employ probabilistic estimates* of exposure, taking into account all
available exposure data and applying exposure factors. This approach requires representative information on exposure for the
scenarios of interest for the relevant populations and across populations. Models at this level of complexity often include
multiple-source exposures.
*Deterministic assessments use single values or point estimates as inputs to the exposure equation. Health-protective assessments typically
use a deterministic approach with default high-end point estimates (conservative). Deterministic assessments might also use central tendency
values to estimate "typical" exposure. A probabilistic assessment uses distributions of data from which multiple points are selected as inputs
to the exposure equation over the course of multiple simulations. As a result, the output of a probabilistic assessment is a distribution of
potential exposure values. Probabilistic approaches are generally used only for higher-tier assessments. See: Exposure Assessment Tools by
Tiers and Types - Deterministic and Probabilistic Assessments (https://www.epa.gov/expobox/exposure-assessment-tools-tiers-and-tvpes-
deterministic-and-probabilistic-assessmentsV
As described by Meek et al. (2011).
The objective of tiering is to optimize the efficiency of the risk analysis by first assessing apparent
margins of low hazard or exposure based on conservative (i.e., more health-protective) assumptions or
scenarios.25 Subsequent tiers incorporate more rigorous analysis with additional data that are based on
initial indications of possible hazard and risk from exposure (see Text Box 3). The Tier 0 assessment
requires the least labor, data, and analysis. It is an accounting of semiquantitative estimates of hazard and
exposure under conservative (i.e., more health-protective) and relatively simple assumptions. The EPA's
Guidance on Cumulative Risk Assessment of Pesticide Chemicals That Have a Common Mechanism of
Toxicity supports a tiering strategy and recognizes that not all CRAs will require the same scope or depth
and that some groups "will require only screening level assessment to decide whether to invest resources
in collecting and analyzing data for a more extensive cumulative risk assessment" (U.S. EPA, 2002a).
The purpose of this initial analysis tier is to assess whether initiating factors reveal cause for concern and
justify the need for a CRA. Care is necessary at Tier 0 to include all identified stressors for which there is
some rationale to combine effects that could result in a common adverse health outcome. If the results of
the initial tier analysis indicate a risk level that is potentially of concern or the need for additional
information to clarify potential risk, subsequent analytical tiers are conducted as necessary—each
progressively more refined, reducing uncertainty, but increasingly more resource and data intensive. The
interplay between the CRA planning phase and the execution of the risk analysis phase in tiers
demonstrates how the tiering process informs the level of effort associated with the risk analysis. Higher
tiers are executed as necessary to discover and demonstrate potential relationships between stressors and
outcomes, including when additional research is necessary . At any tier, the outcome of analysis can be:
• There is sufficient information to proceed with risk characterization;
25 Screening-level assumption of exposure refers to upper-bound limits of exposure and duration. When used in risk assessment,
they are intended to ensure that as much risk as possible is taken into account.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
17
-------
• Additional data are needed but not available; decision for no further action; or
• Additional data are needed and possible to obtain; decision for further assessment (a higher tier).
The phased approach described by Menzie et al. (2007) emphasizes examination of stressor and pathway
combinations thought likely to have the most significant outcomes (see Text Box 4). In this phased
approach, a conceptual model is initially developed to prioritize the relative importance of the various
stressors. This conceptual model for the CRA documents the various exposure pathways and
interrelationships and can be used to indicate the relative importance of various stressor/pathway
combinations. Inclusion of all identified stressors, to the extent possible at this phase, is necessary both
for identifying sources within the EPA's authority, as well as for consideration in tiering the risk
assessment. While all identified stressors and pathways can be represented in the conceptual model,
resources should be directed to understanding the stressor and pathway combinations considered to have
the greatest potential effects. "In this way, it is possible to simultaneously capture the breadth of the
problem and [focus] on its key aspects" (Menzie et al., 2007). The phased approach to prioritizing the
evaluation of combined effects from multiple stressors is useful for bounding the scope of the CRA,
which is particularly important when the assessment includes exposure-response modifier interactions
with the primary stressors. Importantly, while the prioritization of stressors, exposure pathways, and
interrelationships is vital for bounding the CRA analysis plan, the full constellation of considerations
remains documented in the conceptual model and can be referenced to guide additional analysis as
required (e.g., vulnerable populations are identified). Stressors or other agents that might be important,
but for which there are insufficient data for the analysis or for which risk management options are not
effective (see Section 2.7), can be documented and forwarded to the risk characterization step for
discussion as uncertainties or qualitative considerations. Those descriptions may be used to inform risk
management decisions or to identify areas for further research.
The phased approach begins with a simple but comprehensive analysis of key stressors identified in the
conceptual model. Central to the phased approach is a focus on the key considerations in the evaluation
(see Table 1). This focus requires an initial effort to prioritize the relative importance of the various
stressors. Exposure pathways and interrelationships identified in the conceptual model can be used to
indicate the relative importance of stressor and pathway combinations and point toward situations in
which risk management interventions might have the greatest potential for risk reduction.
Text Box 4.
Elements in a Cumulative Risk Assessment Phased Approach
• Develop a conceptual model sufficient to delineate the problem; include all relevant stressors and describe how they
might act in combination.
• Screen stressors to arrive at an appropriate and manageable number for the problem; this step is a focusing exercise.
Other stressors and pathways can be represented in the conceptual model, but resources are directed to understanding
the stressor and pathway combinations considered to have the greatest potential effects. Retain screened stressors on a
watch list for subsequent checks after more information is developed.
• Evaluate the individual effects of individual stressors to determine whether any predominantly contribute to (or could
contribute to) the effect(s) of interest.
• Evaluate the collective effects of stressors without yet considering the potential for interactions (e.g., synergism or
antagonism) and identify the potential for stressor or effect overlap (e.g., based on common properties or temporal and
spatial links).
• Evaluate the combined effects of stressors, taking into account potential interactions and considering qualitative to
quantitative methods, depending on the information available. Key to the iterative process is revisiting these steps at
intermediate stages throughout the assessment to ensure that contributing stressors, influencing factors and effect
endpoints are integrated so that combined effects and primary risk contributors can be well characterized to the level
that existing knowledge allows.
Adapted from Menzie et al. (2007).
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
18
-------
A focus on the interactions emphasized in the phasing process can also be applied to a CRA that is
receptor focused (e.g., vulnerable communities). The granularity of the process steps identified in
Text Box 4, which depict the phased approach, may also be useful for informing the tiered approach
described in Text Box 3.
Tiering, phasing, or both may be helpful process steps when a screening-level effort is indicated as an
initial scoping step or when the envisioned CRA might be so complex that identifying priority focus areas
would help bound the study scope and make the analysis more manageable.26,27
Depending on the circumstances, the initial scoping of the CRA and fit for purpose evaluation may serve
as part of an initial screening-level step (comparable to Tier 0) for evaluating the suitability of CRA to
address the risk management question, although the approaches used to evaluate the suitability of CRA
may differ somewhat across EPA programs (see Text Box 3). In all cases, the screening-level step
examines the available data to determine the extent to which they would inform the CRA's intended
purpose (the management decision) and whether they would warrant a CRA. If the CRA team makes a
positive determination—that the available data suggest the potential to sufficiently address the fit for
purpose determination and support a cumulative risk analysis approach—further tiering or phasing of the
CRA may be appropriate.
2.6. Data Quality, Needs, Availability
The decision to use CRA is based on a purposeful evaluation of whether it is an appropriate strategy to
inform the risk management question. DQOs are established for EPA projects and are used to evaluate
suitability of data or information to inform management decisions.28 Consistent with the DQO process,
planning and scoping for a CRA should consider the data, analytical methods, and available resources to
conduct the CRA to determine whether they are sufficient to meet the risk management goals. The
decision to proceed with a CRA should ultimately be contingent on meeting DQO criteria.
During planning and problem formulation, an initial assessment of the availability of assessment methods,
data, and resources is assembled (discussed further below) within an outline of the general science
approaches considered for the various evidence streams supporting the risk evaluation (i.e., chemistry,
fate, release, exposure, hazard), along with a description of the reasonably available information and
conceptual models. This evaluation process begins in the scoping stage through development of a scoping
summary statement (see Section 2.4). The evaluation of data, methods, and resources is refined and
further clarified during problem formulation and development of the conceptual model and then finalized
for the risk analysis plan.
26 An example of tiering can be found in the Office of Pesticide Program's Pesticide Cumulative Risk Assessment: Framework
for Screening Analysis, which describes such an initial tiering step (U.S. EPA, 2016b): "The screening analysis for CRA
described in this guidance begins with an evaluation of the toxicological knowledgebase available on a particular group of
pesticides derived from experimental toxicology studies submitted for pesticide registration and from the scientific literature. If
the toxicological characterization of potential for common mechanism suggests a candidate common mechanism group (CMG)
may be established, then a screening-level toxicology and exposure analysis may be conducted to provide an initial screen for
multiple pesticide exposure" (U.S. EPA, 2016b).
27 The EPA's Pesticide Cumulative Risk Assessment: Framework for Screening Analysis (U.S. EPA, 2016b) also cites the Meek
et al. (2011) description of the World Health Organization (WHO) International Programme on Chemical Safety (IPCS)
framework for assessment of exposure to multiple chemicals, which describes a screening approach involving tiered analysis
with increasing levels of refinement (see Text Box 3). The WHO/IPCS screening approach is more generic than that described in
the EPA's Framework.
28 The DQO process provides a standard working tool for risk assessors, project managers, and planners to develop DQOs for
determining the type, quantity, and quality of data needed to reach defensible decisions or make credible estimates (U.S. EPA,
2006d).
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
19
-------
Availability of Assessment Methods. Methods for chemical additivity are widely accepted (e.g., U.S.
EPA, 1986b, 2000b, 2023a); therefore, CRAs with chemicals that are toxicologically similar29 have been
successfully conducted and subsequently used by the EPA in decision-making. At the time of this
publication, there were few CRAs demonstrating how to incorporate nonchemical stressors or chemicals
that are not necessarily toxicologically similar but act on a common outcome. However, when there are
data on nonchemical stressors that will support their inclusion in the risk analysis, then they should be
incorporated into the CRA plan. If the toxicological action is not understood, then alternative approaches,
such as response addition, may capture the possible risk modifier. When methods are not available to
incorporate data into the risk analysis directly, two alternative approaches are available to the CRA team:
• Include this information as relevant context during risk characterization; or
• Include this information in an alternative assessment framework provided to the risk manager.
Availability of Information. To conduct a CRA, the team needs to determine the data required for the
specific needs of the assessment (e.g., monitoring data or modeling results, dose- or exposure-response
information), the availability of the data, and the data quality. If the preferred data are unavailable, then
consideration is given to whether new data can be collected, and it may be appropriate for the CRA team
to relay the need for data to an appropriate authority so such needs can be included in future Agency
projects or grant programs. In some instances, the absence of necessary data, such as exposure data, might
preclude the assessment or necessitate data collection (or modeling) before the assessment can proceed.
Selecting the analysis methods for a risk assessment then entails choosing the appropriate methods for
using the available or anticipated new data.
Availability of Resources. Because CRAs might be designed to capture complex assessments of multiple
chemical and nonchemical stressors, they can be resource intensive. The CRA team, in consultation with
risk managers, needs to consider required resources for the CRA (e.g., budget, technical support, time) in
planning and scoping.
The information available to determine whether to proceed with the CRA might be insufficient until the
scoping of the assessment is fully conducted. As stated above, the determination might need to be
revisited during problem formulation—for example, during analysis plan development—to confirm that
data and analytical methods are sufficient to meet the required level of confidence to inform the risk
management decision. Always central to the decision is whether the adequacy of the analysis will serve
the purpose for which the assessment is performed (WHO/IPCS, 2009c). Following the provisional
decision to proceed with CRA planning, new information might necessitate adjustments in the scope,
conceptual model, or analysis plan. This possibility is built into the CRA planning process through two
strategies:
• Recursive and iterative consideration of the relationships among stressors, and among stressors
and receptors, can be appropriate during the planning, analysis, and characterization phases of the
overall assessment. This process can help revise and refine the CRA team's understanding of the
significant stressors and their relationship to the receptors of concern. When appropriate, iteration
enables previous decisions to be revisited when new information suggests reconsideration.
• A phased or tiered approach to the risk analysis is recommended to manage resources efficiently
when informing the risk management decision (see Section 2.5).
29 Toxicological similarity infers a general knowledge about the action of a chemical or a mixture and is used as an overarching
term with a wide range of reference, including mechanism of action, target organ (e.g., enzyme changes in the liver), adverse
outcome pathways, and in silico tools such as structure-activity or read-across analyses (Williams et al., 2021). Similarity
judgments can be tailored to both the specific goals of the risk assessment and the availability of information (U.S. EPA, 2000d,
2023a).
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
20
-------
The scope of a CRA should consider all relevant pathways and routes, aggregate and background
exposures, data quality, and nonchemical stressors that are exposure-response modifiers—including
vulnerabilities that are due to social and economic factors. The EPA's well-established Quality System
(U.S. EPA, 2020d) emphasizes that a critical analysis of available and relevant environmental data is
crucial, including monitoring and modeling results (U.S. EPA, 2002b), and discusses important aspects of
data quality relevant to the CRA. Important considerations for data quality include the extent to which the
scientific and technical approaches are reasonable and consistent with the intended application, the
tolerance for potential decision errors, requirements for precision, and the use of secondary data collected
for purposes other than the planned assessment. Similar to single-chemical risk assessments, additional
types of scientific expertise and planning methods may be required to ensure the CRA analysis is feasible
and the results are relevant to the risk management decision (NRC, 2009; Sexton, 2015).
Because numerous datasets of varying quality are possible for any risk assessment, a CRA could be
subject to conditions in which data analysis should be prioritized or triaged to maintain the integrity of the
assessment, as is proposed in the phasing discussion (see Section 2.5). The scope of the CRA, including
the required level of rigor of the assessment (e.g., screening-level versus more refined analysis), could
also influence data quality needs.
2.7. Project and Risk Management Considerations
As recommended elsewhere in these Guidelines, consideration of risk management options is an
important iterative element of CRA. Project management and resource needs should receive particular
attention during the planning and problem formulation phases of a CRA.
Risk Management Issues. NRC's 2009 Science and Decisions: Advancing Risk Assessment emphasized
the importance of considering risk management options in the design of risk assessments, including CRA
(NRC, 2009). Care always should be taken to prevent biasing or manipulating the risk assessment to serve
a preconceived risk management strategy and, at the same time, ensure consideration of risk management
possibilities in the design of risk assessments (U.S. EPA, 2014b).
Risk managers need the ability to compare different management alternatives, using constraints they identify
and describe, to better understand the trade-offs associated with selecting one alternative over another and to
understand uncertainties. Some risk management issues that might be included in planning a CRA are:
• Relevant regulatory considerations (e.g., program-specific guidance influencing CRA scope); and
• Technical feasibility of alternative interventions that might reduce stressor levels to different
extents, introduce new stressors, or be completed more quickly than other intervention(s).
With CRA, there may be numerous risks to remediate, resolve, or otherwise manage. If the CRA team is
asked to evaluate articulated risk management options, the CRA conceptual model is expected to base the
comparison of risks among these options on the stressors identified in the problem definition and the
characteristics of the target population, among any other pertinent factors. The goal of this CRA is to
assess and compare risks for each of the risk management options that might be implemented to reduce
the cumulative human health risk(s) associated with a scenario.
Consistent with long-standing Agency practices and recommendations from NRC, the EPA ensures that
risk assessments are not directed or unduly influenced by risk managers (NRC, 2009). There are two
purposes for considering possible risk management input in the analysis design:
1. To identify a subset of stressor(s) or health outcome(s) on which the CRA would potentially
focus in a systematic and transparent manner. Other stressors or health outcomes that overlap
with the problem definition, but are not the target of remedy, could be eliminated from the CRA
if they do not contribute to the cumulative risk. Because CRA may involve cutting across
multiple media, statutes, and governmental authorities, any design that limits analysis of stressors
and adverse health outcomes to individual programs should be carefully considered.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
21
-------
2. To adjust the conceptual model. Adjustments to, or addition/elimination of stressors and health
outcomes initially identified may be necessary based on new information or a determination of
insignificance. Similarly, additional risk management options may be identified for evaluation (or
a previously evaluated option may need to be eliminated due to no longer being viable).
Decisions about narrowing the evaluation of stressors or outcomes or changing the conceptual model
should be part of the CRA communication plan to ensure stakeholders have the opportunity to understand
and provide input. Finally, any stressors and vulnerability factors that were screened out because they
were unaffected by the intervention(s) could be reintroduced into the CRA analysis to evaluate whether
and how they affect the health outcomes evaluated in the CRA.
Project Management Issues. Because the nature of a CRA can be complex, the use of project management
approaches and tools is advisable. Development of a project management plan in the planning phase can
ensure the CRA achieves all project goals within the given constraints, including defined risk
management objectives, schedule, quality, and budget. Including monitoring and control steps in the
project management plan, as feasible, might help ensure that resources and other possible limiting factors
are addressed in a timely manner. If risk or decision managers are not members of the CRA team, they
should routinely be consulted to ensure the CRA remains responsive to the risk management question or
problem that initiated it.
2.8. Peer Review
Peer review of the CRA may be considered as a final step but can be conducted at any time in the process
that it is determined to be helpful. Key peer review opportunities exist at the completion of the conceptual
model or analysis plan, upon completion of the risk analysis, or upon full completion of the CRA. Peer
review should be performed consistent with the EPA's peer-review policy and Peer Review Handbook
(U.S. EPA, 2015a). Consideration of the following elements of the CRA may be appropriate:
• The purpose and scope, which outlines objectives, constraints, boundaries, and scale of the CRA
and provides the foundation upon which the CRA proceeds.
• The conceptual model, which illustrates the environmental health issue(s) motivating the CRA
and presents the rationale for selecting the assessment elements from sources to effects and risk
metrics, serving as a foundation for the CRA analysis plan.
• The CRA analysis plan, which provided the technical work plan for the risk analysis steps and
specified data/inputs, methods/models, and outputs/risk metrics.
• Data from the risk analysis and the risk characterization.
An independent peer review of the identification of constraints, boundaries, and other challenges to
conducting the CRA may help ensure quality robustness for the assessment: consideration of relevant
stressor-stressor and stressor-receptor relationships, identification of potential exposure-response
modifiers, and recognition of strengths and weaknesses of proposed analytical methods. Acknowledging
that CRA has limitations is appropriate, whether related to data quantity, quality, or availability;
analytical methods; or resources. Peer review offers an opportunity to independently validate the CRA
design to inform the risk management decision and identify any potential weaknesses that might require
the attention of risk managers. Strategies for choice of peer-review mechanism should consider
opportunities to engage affected communities, especially for reviews of the conceptual model and risk
characterization.
Independent reviews can be formal for elements supporting major CRAs or relatively informal for CRAs
of limited scope. Reviewers should be independent from the members of the CRA team. Independent
review benefits the CRA process by identifying potential challenges, constraints, disagreements, and
uncertainties that might arise.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
22
-------
CHAPTER 3. PROBLEM FORMULATION
CRA problem formulation is a technical process that identifies the major factors to be considered in the
CRA. Problem formulation develops an operational CRA analytical structure by producing a conceptual
model and analysis plan. The conceptual model illustrates the associations among stressors, exposure
pathways, receptors (exposed lifestages and populations) and assessment endpoints that will be addressed
in the CRA. The analysis plan, building on the conceptual model, describes the approach for conducting
the risk assessment: design, methods, key inputs, and intended outputs. Together, the conceptual model
and analysis plan, informed by stakeholder discussion as appropriate, reflect the purpose for conducting
the assessment and the considerations identified in the initial scoping (e.g., through a scoping summary
statement).
3.1. Examining Risk Management Options Based on the Initiating Factor
Dialogue and consultation between risk assessors and managers whose decisions will be informed by the
CRA is important to ensure the CRA is tailored to support risk-based decision-making. Input from risk
managers continues throughout planning and problem formulation to help the CRA team identify the
information and assessment methods that will support decision-making.
The following questions (adapted from U.S. EPA (2014b) and NRC (2009)) can be used in consultation
with risk managers to determine an assessment's fit forpurpose determination:
• Does the assessment provide adequate information to inform the choice(s) among risk
management options?
• Will the risk assessment need to be modified or expanded to inform evaluation of risk
management options?
• What level of uncertainty and variability analysis is appropriate and acceptable in the assessment?
Consulting with risk managers during the problem formulation stage of a CRA affords an opportunity to
focus the assessment on the question(s) that the decision-maker needs to have answered. The consultation
might include obtaining insight on how to narrow the focus to fewer or more readily addressed stressors.
The CRA team may consider risk management options during the problem formulation step so the focus
is on stressors that might be most responsive to risk management strategies (Solomon et al., 2016).
Consulting with risk managers also helps ensure the level of complexity and uncertainty is sufficient to
discriminate among risk management options but avoids expending resources when analyses do not
significantly contribute to informing decision-making. Continued dialogue with the risk managers may
lead to the identification of countervailing risks associated with the risk management option.
3.2. Conceptual Model
Conceptual models are pictorial or written depictions of the pathways connecting stressors to health
outcomes in a population of interest (Linder & Sexton, 2011; Menzie et al., 2007). The pathways can
represent known, predicted, or assumed relationships. Conceptual models consist of two principal
components: (1) a set of risk hypotheses that describe predicted relationships among stressor, exposure,
and health endpoints or responses, along with the rationale for their selection; and (2) a diagram that
illustrates the relationships presented in the risk hypotheses (U.S. EPA, 2014b). Conceptual models are
used to inform technical work products (i.e., the analysis plan), and may incorporate the following
considerations:
• The rationale for selecting the sources, stressors, exposure pathways, receptors, exposed
populations, assessment endpoints, and risk metrics.30
• The basis for the conceptual model development.
30 For example, cases of disease or disease incidence, hazard quotient, magnitude of effect, and margin of exposure.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
23
-------
• The scientific and resource (including temporal) implications of additional data gathering.
Development of the conceptual model may rely on a variety of data sources to hypothesize known or
suspected stressors, cumulative combinations among them, and resulting adverse health outcomes. Data
sources may include published studies, toxicological data, and stakeholder input. For purposes of risk
assessment, conceptual models are usually developed by risk assessors, with consideration of input from
stakeholders, such as a relevant government, Tribal authority, community representative, or other experts
whenever appropriate. Conceptual model development is iterative, and models are updated when new
data become available that alter the original model. The inputs and outputs for a conceptual model may be
represented using directed acyclic graphs (DAGs). DAGs are visual representations of assumptions about
the underlying causal structure that generate observable data. DAGs can be used to identify confounders
and modifiers for a given exposure-response relationship. This information can be used to determine
whether the available data are appropriate and sufficient to obtain risk estimates for the population(s) of
interest (Brewer et al., 2017).
The complexity of the conceptual model depends on the complexity of the problem and the scope of the
risk assessment. The complexity might be related to the number of stressors, exposure pathways, or
assessment endpoints; the nature of effects; and the characteristics of the exposed populations or
lifestages. Conceptual models can be particularly useful when addressing multiple, diverse activities
potentially contributing to cumulative risk to receptors of concern (Suter, 1999). A CRA conceptual
model uses available evidence and hypotheses to "draw a picture" of the environmental health issue
generating concern and motivating a CRA.31 The visual representation of the conceptual model is a
diagram that could include the following types of elements: sources, stressors, exposure and outcome
pathways, receptors, endpoints, and risk metrics (Figure 3).
The sources of stressors relevant to a CRA might include point/nonpoint sources. Natural sources and
other background sources (e.g., upwind or upstream) might also be appropriate to consider in combination
with a stressor of concern. Examples of stressors of concern include chemicals, physical stressors,
biological stressors, and psychosocial factors.
Routes of exposure relevant to humans and other animal species include ingestion, dermal uptake,
inhalation, and vector transfers (e.g., mosquitoes). Exposure to some nonchemical stressors might not
involve a traditional environmental pathway. They may also influence responses to biological, physical,
and chemical stressors. Environmental pathways of exposure, particularly along food chains, might
include the processes of bioaccumulation and biomagnification.
31 See U.S. EPA (2006a) Exhibit 3-1, p. 63; Exhibit 4-7, p. 134.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation -
2025
24
-------
Activities that emit stressors
• Point sources (e.g., industrial pollutant discharge and pollutant clean-up sites)
• Non-point sources (e.g., automobiles or consumer use releases)
• Natural sources (e.g., wildfires)
Chemical or nonchemical stressors or agents that can cause an effect or influence the exposure-
response relationship for the primary stressor(s) being evaluated.
• E.g., chemicals; physical stressors; radiation; microbiological or biological agents; nutritional factors;
economic factors; physiological response factors (e.g., allostatic load); habitat loss or fragmentation;
land use alteration (e.g., hydrologic modification, natural disasters)
Pathway - the way a receptor comes into contact with a stressor
• Pathways can be environmental media (e.g., surface water, groundwater, indoor and outdoor air,
soil) or products (e.g., consumer and personal use products, pharmaceuticals)
• Exposure Route - the way a stressor enters a receptor (ingestion, inhalation, dermal absorption,
vector transfer)
Individuals, populations, or lifestages
• E.g., infants, communities with EJ concerns, vulnerable populations
Human health and disease endpoints
• E.g., increased cancer risk, elevated hazard indices, morbidity, mortality
Figure 3. Elements of a Conceptual Model32'33
Receptor definitions range from target organs to individuals and populations and may focus on certain
subpopulations or lifestages. Assessment endpoints could be human or ecological effects or both. For
example, developmental deficits or loss of species diversity could be based on findings from animal
studies, ecotoxicological studies, clinical studies, or sociological or epidemiological data. Risk metrics are
important because they provide a means to quantify risk, such as disease incidence, hazard quotient,
magnitude of effect, and margin of exposure.
The additional factors described below might be important to consider in conceptual model development.
Exposure context and characteristics. Frequency, duration, intensity, and overlap of exposure intervals for
a stressor or mixtures of stressors are important in considering the time frame of the stressor-response
relationships. Exposures could be acute, subchronic, chronic, delayed, intermittent, or a combination for
multiple stressors. Notably, the same level of exposure can have different effects depending on timing,
due in part to differing windows of susceptibility for specific health outcomes. This context can be
especially important for early-life exposures and when considering cumulative exposure (e.g.,
consumption of contaminated food products associated with Tribal lifeways or other subsistence-reliant
32 Figure 3 illustrates the conventional stressor-to-receptor model. As discussed in Section 1.1., CRA can also be conceptualized
as a receptor-(adverse outcome)-to-stressor problem, as depicted in Appendix Figure C-3.
33 Hie phrase "communities with environmental justice concerns" is used by the EPA's Office of Environmental Justice and
External Civil Rights (OE.TCER) fhttps://www.epa.gov/environnientaliustice/learn-about-environniental-iustice'l.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
25
-------
populations) for specified time periods.
Availability of information from different evidence streams. Toxicology, epidemiology, field studies,
Indigenous Knowledge, and other related evidence might be available for consideration in conceptual
model development. A literature review should be performed to gather this evidence and, together with a
priori knowledge, input from stakeholders (including Tribes), expert opinion, and other assumptions,
which can contribute to development of the conceptual model. An initial evaluation of evidence relating
exposure to outcome will facilitate a description and characterization of the strength of that evidence. This
evaluation contributes to an initial WoE for the relationships of interest, which will be useful in deciding
which relationships to include in the conceptual model.
Timing of exposure when combining stressors. Careful consideration of relationships important to
characterizing cumulative risks might necessitate inclusion of temporal relationships between stressors,
including exposures occurring during early-life stages or multigenerational exposures. Potential exposure-
response modifiers, biases, and confounders are additional factors that may be important to consider in a
conceptual model when examining the relationship between a stressor and a receptor (Christensen et al.,
2015). Detailed causal models could be used to identify biases and confounders and subsequently might
inform revision of pathways in the conceptual model. Confirming that the conceptual model includes
assessment endpoints related to risk management objectives and options is also important.
Conceptual models identify and display relationships between various stressors (or factors) and endpoints
for use in developing the risk analysis plan. They may also include factors and endpoints that will not be
analyzed in the risk assessment but that are important in the overall decision-making process. For
example, although a risk assessment for a particular stressor might focus on exposure pathways or media
relevant to the regulatory decision being faced (e.g., ingestion of drinking water), the conceptual model
may also describe the contribution from other pathways (e.g., inhalation of an airborne chemical), thus
ensuring appropriate characterization of and context for the assessment results (U.S. EPA, 2014b). In
addition to the hazard and exposure analysis, the risk characterization may incorporate factors
qualitatively that are important in the overall decision-making process due to vulnerability or risk
management considerations. Input from stakeholders can also provide valuable information for
development of the model. With incorporation of multiple stressors and data streams, there is likely to be
variability in data quality resulting in increased or unforeseen uncertainties. It is important to identify
these uncertainties and variabilities in the conceptual model so that they are appropriately considered
throughout the CRA process and addressed in risk characterization. Completion of the conceptual model
and analysis plan can provide an opportunity for scientific review and stakeholder involvement (see U.S.
EPA, 2014b, Section 2.1.4).
Consulting higher-level conceptual frameworks, or theoretical frameworks as illustrated in the generic
model shown in Figure 4, may help to contextualize causal pathways for complicated relationships (Gee
& Payne-Sturges, 2004; Linder & Sexton, 2011; Morello-Frosch & Shenassa, 2006; Schulz et al., 2005;
Sexton & Linder, 2010). Figure 4 highlights higher-level relationships between generalized factors for
both community- and individual-level vulnerabilities that may be relevant to a risk management problem.
Other examples of conceptual models specific to human health are provided in Section 2.2.2 of the EPA's
CRA Framework (U.S. EPA, 2003b) and the Risk Assessment Guidance for Superfund (RAGS). RAGS
Part A, Exhibit 4-1 provides the elements of a conceptual evaluation model (U.S. EPA, 1989). RAGS Part
D, Table 1 provides an approach for planning and scoping to create a conceptual model (U.S. EPA,
2001b). Appendix A of the Soil Screening Level Guidance provides methods, summary sheets, and other
information that might be useful in constructing a conceptual model (U.S. EPA, 1996a).
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
26
-------
Race,'ethnicity
Residential segregation
Residential location
Neighborhood
resources
Community
stressors
Structural
factors
Environmental
hazards and
pollutants
Community
stress
Exposure
Community-level
vulnerability
Individual stressors
Individual coping
Appraisal process
Internal dose
Individual stress
Biologically
effective dose
Individual-level
vulnerability
Health elleet
(disparities)
Figure 4. Example of a Generalized Conceptual Model Evaluating Cumulative Risk and the Exposure-
Disease Paradigm
Reprinted with permission from Gee and Payne-Sturges (2004).
A more detailed example of a response-based conceptual model (cardiovascular disease, CVD) is
provided in APPENDIX C. Developing a conceptual model can be an iterative process. As elements are
added or considered for addition, the CRA team may gain new understanding of relationships between
model elements, necessitating a revision of the conceptual model. The final conceptual model will be the
culmination of the iterative process and should represent features of both a scientific hypothesis and a
work plan. It should directly inform development of an analysis plan by describing key inputs, methods,
and analytical outputs of the assessment, and include risk metric(s). As described in the Agency's
Framework for Cumulative Risk Assessment, the conceptual model should capture the rationale for ri sk
management decisions and should be used as a risk communication tool (U.S. EPA, 2003b). In some
cases (e.g., in a major assessment of pollutants with controversial or novel methods or scoping issues),
conceptual models might be submitted for review, leading to possible additional iterations.
Table 2 summarizes a stepwise strategy for incorporating factors in planning and problem formulation
with different alternatives (stressor-, receptor-, vulnerability-based approach) that depend on differences
in the initiating factor. Step 1 is relevant particularly to conceptual model development, whereas
subsequent steps 2 and 3 might also serve to inform model development through evaluation of stressors.
Selection of the approach should align with the context and scope of the risk assessment (e.g., a
regulatory standard, a community-based assessment).
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
27
-------
Table 2. Outline for a Stepwise Approach to Cumulative Risk Assessment Planning and
Problem Formulation
Step
Stressor-Based Approach
Effects-Based Approach
Vulnerability-Based
Approach
1. Develop a conceptual
model describing
relationships between
stressors and adverse
health effects.
Identify common
receptors and endpoints.
Identify receptors and
endpoints that stressors affect,
both individually and in
combination.
Specify and characterize
adverse health outcomes of
concern, and assign possible
stressors based on dose-
response and exposure data.
Identify the vulnerable
populations), community, or
lifestage prior to developing the
model, along with adverse
health effects of concern.
2. Screen stressors.
Determine which stressors
need to be included in the
assessment and which might
act in combination.
Identify a manageable number
of priority stressors that can
be linked to effects and can
characterize the problem
adequately.
Identify stressors affecting the
population/community/lifestage
to establish priorities that
should be included in the
analysis.
3. Evaluate the effect of
stressors on critical
endpoints.
Evaluate (1) individual effects
of stressors and (2) combined
effects of combinations of
stressors, including relevant
psychosocial stressors.
Evaluate individual effects of
individual stressors to
determine whether one or a
few stressors are predominant
in any adverse health
outcomes.
Evaluate individual and
combined effects of important
stressors to determine whether
one or a few stressors are
predominant in any adverse
health outcomes .
Table is based on concepts adapted from Sexton (2015).
3.2.1. Consideration of Stressors
CRAs consider human health or ecological effects following exposures to multiple stressors. During
problem formulation and consistent with the question(s) the CRA needs to answer, an initial approach to
evaluating potential stressors is developed. If the initiating factors stem from adverse outcomes noted in
receptors, stressors might not be explicitly identified initially. Such circumstances require an initial
inquiry of the receptor and adverse outcome(s), followed by an investigation of potential stressors based
on available source and exposure data. The problem formulation reflects this inquiry and can address
human health risk, ecological risk, or a combination thereof. A preliminary phasing of stressor evaluation,
as discussed in Section 2.5 and Text Box 4, should be considered when numerous stressors are identified
for evaluation.
Stressors typically considered in CRA can be chemical, biological, physical, or psychosocial. Described
in more detail in Section 3.2.3 and Section 3.2.4, these stressors often are grouped on the basis of
mechanism, temporal occurrence, health endpoint, or some other commonality. Stressors and their
source(s) are identified in the context of potentially affected populations or ecosystems, potential stressor
effects, and protective factors that might alter responses to stressors. A unique aspect of CRA is that it
includes the consideration that receptors can be exposed to multiple stressors via multiple exposure
pathways from various environmental media. For example, a local subsistence diet, as is common among
Tribal populations, may result in bioaccumulation of contaminants for that population. Importantly,
exposures might not be associated with traditional environmental media (e.g., air, soil, or water) in some
cases but could be associated with nonchemical stressors (e.g., psychosocial, socioeconomic, temperature,
or habitat). Increasingly, physical factors associated with climate change may be considered important
dose-response modifiers or stressors.34,35
34 See How Climate Change Affects Human Health (https://www.epa.gov/climateimpacts/climate-change-and-human-
health#how).
35 See EPA's Impacts of Climate Change web page (https://www.epa. gov/climatechange-science/impacts-climate-change'l.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
28
-------
In cases involving multiple chemical stressors, potential interactions among chemicals might need to be
considered.36 When available, information from various in vivo and in vitro test systems can be used to
elucidate and characterize the interactions and can provide important insights at different levels of
biological organization. For evaluating chemical mixtures, the EPA typically assumes the chemicals are
toxicologically similar if the chemicals have a common mode of action or if they affect a common target
organ or elicit a common effect. Such mixtures are typically evaluated using methods that assume dose
addition. For exposures to chemicals that elicit a common outcome but are not toxicologically similar, the
EPA has employed methods that are not based on dose addition, such as response addition as well as
other methods that use elicitation of a common outcome as discussed in EPA documents (e.g., U.S. EPA,
1986b, 2000d, 2023a).
Epidemiological studies are a potential source of data for informing CRAs. Epidemiological studies might
encompass both chemical and nonchemical stressors and might evaluate the ways in which possible
relationships among stressors could affect human health and ecological outcomes. Because stressors can
vary together in the environment, covariance measures might be used to examine
interactions/interdependencies among them, such as how changes in levels of one stressor are associated
with changes in levels of another. Other disciplines and areas of research, such as microbiology, public
health, sociology, health psychology, health geography, exposomics, in situ toxicity testing, or in silico
toxicology (including virtual and complex tissue models that resemble an organ in vivo), also might
contribute to identification of stressors of concern for CRAs.
3.2.2. Receptors of Potential Interest
Identifying receptors of potential interest could be targeted based on the initiating factor (e.g., geographic,
receptor, or stressor based). As described in the EPA's CRA resource document (U.S. EPA, 2007a), the
initiating factor for a CRA can influence whether the study boundary or the population is defined first.
For example, the initial population of concern could be a Tribe or a community in a larger city or county,
including any identified vulnerable population groups. The initial description of the study area and
population of concern may be considered preliminary and may change during the course of the risk
assessment.
Factors potentially important to consider in characterizing receptors for a CRA are described in Text
Box 6. Selection of these factors to be included in the CRA will depend on the purpose and scope of the
assessment.
Exposure-Response Modifiers. Numerous
conditions or factors, termed exposure-
response modifiers, can contribute to
altered levels of exposure or altered risk of
a health effect occurring at a given level of
exposure to an environmental pollutant
(Text Box 5). Examples of human health
exposure-response modifiers include
genetics, sex, preexisting disease, altered
physiological functions, psychosocial
stress, and lifestages (Sexton & Linder,
2011). Chemical and nonchemical stressors
that are not the primary stressor(s) in a
CRA may also operate as exposure-
response modifiers. Behavioral variability,
30 Interaction through additivity of dose is of primary interest; however, the potential for other interactions should be considered,
such as synergism, antagonism, and masking (see U.S. EPA, 2000d).
Text Box 5.
Exposure-Response Modifiers and Stressors
In this document, an exposure-response modifier is characterized
as a condition or state (e.g., sex, lifestage, socioeconomic status),
whereas a stressor is characterized as a physical, chemical,
biological, or psychosocial agent (see Glossary), referred to in
this document as nonchemical stressors. The terms can be
somewhat interchangeable, and in some cases, exposure-response
modifiers and nonchemical stressors (e.g., low socioeconomic
status) might appear to overlap. Hie nature and circumstances of
the CRA, however, will inform determination of the appropriate
characterization as either an exposure-response modifier or a
nonchemical stressor. It is possible for a factor to be both an
exposure-response modifier and a nonchemical or chemical
stressor.
Vinikoor-Imler et al. (2014).
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
29
-------
such as occupation or hand-to-mouth activity, can modify exposure. An example of an environmental
exposure-response modifier is pH, which can affect metal bioavailability. Exposure-response modifiers
can alter an exposure or a response negatively or positively. Some factors can also insulate or reduce the
deleterious effects of stressors or reduce exposure and function as protective agents. In these Guidelines,
however, exposure-response modifiers are discussed largely in terms of vulnerability factors (see further
discussion in APPENDIX D).
Exposure-response modifiers associated with a community or a Tribe are the conditions or circumstances
that shape where people are born, live, work, grow, learn, play, and age. These community-level
conditions or circumstances are also known as the social determinants or social causes of health (CSDH,
2008; Solar & Irwin, 2010). Examples include housing, education, food access or security, income,
transportation, physical and social environmental conditions, and health care access (CSDH, 2008; U.S.
HHS, 2020).
Text Box 6.
Factors to Consider in Characterizing Receptors
• Relevant level of organization: Receptors can be organized at the organism, community, population, ecosystem, or other
appropriate level.
• Level of exposure: Differential exposures to primary stressors of concern can be due to geographic area, length of
exposure (including multigenerational), lifestage, sex, racial or ethnic group, economic status, housing type, proximity to
exposure sources, or dietary and lifestyle/cultural factors, etc.
• Exposure-response modifiers:
o Factors that might increase or influence vulnerability or resilience: Greater vulnerability response to the primary
stressors of concern could develop as a result of the presence of social or individual factors that either influence
exposure or response or both. Examples include low socioeconomic status, food insecurity, poor housing quality,
lack of access to health care, preexisting disease, lifestage, occupation (e.g., farmworkers) and sex.
o Factors that might increase resilience in the population: Examples include access to health care, positive social
climate, and food security.
o Landscape or geographic concerns: Watersheds, aquifers, airsheds, regional ecosystems, and recreational lands might
influence exposure and outcomes in the receptor group.
At the individual level, exposure-response modifiers include individual behaviors and practices and
individual characteristics such as nutritional status, genetics, epigenetics, preexisting disease, and
lifestage that could affect exposure and response to stressors. These individual attributes are also referred
to as individual determinants of health. Social determinants can influence health through direct and
indirect impacts on individual determinants and through multiple pathways and vice versa (deFur et al.,
2007). Vulnerability is a differentiating factor for how individuals, communities, populations, or
organisms experience adverse effects related to exposure to environmental stressors (deFur et al., 2007).
Therefore, if a CRA is initiated to address the needs of one or more vulnerable populations or lifestages, it
should identify pertinent exposure-response modifiers to incorporate vulnerability into the conceptual
model and identify whether and how to include this information in analysis and decision-making stages.
3.2.3. Adverse Effect and Exposure Stressor Groups
Multiple approaches are available for grouping effects or exposures for evaluation. Some grouping
approaches are based on statutory requirements (e.g., FQPA), whereas others have been developed for
particular risk assessment efforts. In selecting a risk assessment method for chemical mixtures, an
important initial step is evaluation of data quality (U.S. EPA, 2000d). The EPA's Supplementary
Guidance for Conducting Health Risk Assessments of Chemical Mixtures emphasizes consideration of
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
30
-------
whether data suggest the stressors are acting by similar toxicological processes and whether they can be
grouped by emissions source, chemical structure, or biological activity (U.S. EPA, 2000d).
Several approaches or criteria have been described for grouping chemical stressors on the basis of
toxicology (Moretto et al., 2017), including common adverse health outcomes or similar modes or
mechanisms of action among a group of chemicals. Numerous other considerations also can be used to
identify a preliminary list of stressor groupings (see below). Subsequent determination of stressor
grouping for the purpose of assessing risk of particular health or environmental outcomes could be based
on methods for additivity of hazard from mixtures and determination of response modification, if any.
Considerations for grouping stressor or effects include:
• Environmental medium (e.g., air, soil, water)
• Duration (e.g., acute, chronic, subchronic)
• Source of emissions/releases
• Co-occurrence in the same geographic location
• Functional use
• Chemical structure
• Timing of exposure (e.g., early lifestage and stressor exposure or overlapping exposure to the
chemicals in the mixture exhibiting a common adverse outcome, an overlap in the duration of the
health effect as a consequence of one chemical exposure and subsequent exposures by a second
chemical, or other types of overlap based on the toxicokinetics or toxicodynamics of the
chemicals)
• Mode of action (e.g., dose addition might be applied to chemical groups determined to have
similar modes of action)
• Outcome (e.g., morbidity and mortality, neurotoxicity, cancer, respiratory irritation, development,
reproduction)
• Receptor vulnerabilities, human or ecological, including level of biotic or abiotic organization
and evidence for vulnerability in the target population
3.2.4. Integration of Data for Examining Stressor-Response Relationship(s)
The scope of CRA scientific inquiry is articulated in the problem formulation and is focused on the
stressors that are of greatest importance for the assessment's intended purpose. Therefore, a strategy is
needed to develop the evidence base of key stressor-response relationships. Strategies for assembling data
and information can be more or less formalized but need to describe what evidence is included (or
excluded), rationales for inclusion, and the extent to which evidence supports possible answers to a
scientific question (EFSA Scientific Committee, 2017). This process has been commonly termed a "WoE
strategy," and this phrase is used in these Guidelines to refer to this process.
For CRA, the WoE strategy should identify possible stressor-response relationships and exposure-
response modifiers for consideration in the conceptual model. The results can be expressed qualitatively
or quantitatively, depending on the information available and the needs of the CRA.
This strategy begins with identifying available sources of evidence (e.g., through a literature search or,
when few published data are available, a survey of community-identified stressors). Evidence might be
available from a variety of disciplines, including toxicology, epidemiology, clinical research, sociology,
ecology, or spatial analyses. Different disciplines use different mathematical and statistical techniques,
with varying degrees of applicability to the population(s) of interest. Consequently, the WoE and data
quality assessment should be tailored to arrive at the best possible explanation for the scientific questions
formalized in the CRA problem formulation and conceptual model. This point is especially important
when there are nonquantifiable agents or factors that are considered by the CRA team or stakeholders to
be consequential to the CRA. The WoE documentation becomes the record identifying and justifying the
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
31
-------
decision to include or exclude factors in the analysis plan or, alternatively, the decision to instead consider
and discuss factors in the risk characterization. The World Health Organization (WHO) defines WoE as
"a process in which all of the evidence considered relevant for a risk assessment is evaluated and
weighted" (WHO/IPCS, 2009b). The EPA's Integrated Risk Information System (IRIS) program
characterizes evidence synthesis and integration within a larger framework of systematic review (U.S.
EPA, 2022d).37 The OR J) Staff Handbook for Developing IRIS Assessments (IRIS Handbook) notes that
the term "evidence integration" is analogous to "weight of evidence" used in some other assessment
processes. Other methods and approaches for weighing evidence of various types, in various contexts, and
for various purposes are available (Linder et al., 2010; Linkov & Satterstrom, 2006; Weed, 2005), as are
other frameworks for integrating and evaluating evidence from multiple disciplines (Adami et al., 2011;
European Food Safety Authority [EFSA] Scientific Committee, 2017; Levy, 2008; Stahl & Cimorelli,
2020). Moretto et al. (2017) described use of evidence tables to present the strength of evidence or
uncertainty associated with grouping for additivity analysis. The EPA's Quality System (U.S. EPA,
2020d) is available for data quality assessment. The EPA's CADDIS (Causal Analysis/Diagnosis
Decision Information System), its accompanying text, Ecological Causal Assessment (Norton et al.,
2014), and the Agency's Weight of Evidence for Ecological Assessment (U .S. EPA, 2016c) provide
techniques and resources for weighing evidence for ecological receptors. WoEs are demonstrated in the
existing EPA CRAs (U.S. EPA, 2006b, 2006e, 2006f, 2007b, 2009) and in other guidance documents
(U.S. EPA, 1994, 1998a, 2005).
The evidence for a stressor-response relationship need not be conclusive for it to be documented in
problem formulation. Uncertainty might call for including the possible link between stressor and receptor
in the conceptual model or analysis plan. There, it could direct additional studies toward the uncertain
relationship or, in the absence of sufficient data to form a conclusion, flag the uncertain relationship for
consideration by the risk manager later in the risk characterization phase. The data available to support
the pathways in the conceptual model also might be highly variable. For example, some relationships
might be flagged in the WoE because they are important to stakeholders, but no empirical data have yet
been collected to support stakeholder concerns. The CRA team should expect stakeholders to have
diverse opinions on possible causality or dose-response modifiers.
When available and appropriate, the incorporation of biomonitoring data may provide useful measures of
internal doses reflective of exposures via multiple pathways and multiple different chemicals operating by
the same exposure-to-effect pathway. Biomonitoring data may also provide information regarding
potential exposure disparities among diverse demographics, such as race or income, and may help
establish baseline exposure conditions in a population. Capturing all identified dose/exposure-response
ideas for consideration and possible evaluation is appropriate and, particularly in community-based
CRAs, important for acknowledging stakeholders' concerns.
Community observations can provide significant and practical insights into the problem and assumptions
made. They can also provide evidential weight for practical and relevant decision-making. Community
observations might include an understanding of the relative contribution of stressors to effects (even if
only qualitatively). For example, noise as a physical stressor is generally understood to be worse the
louder it is, especially if baseline or ambient noise is close to the threshold for ear damage (U.S. EPA,
2020a). No empirical data specific to a particular anthropogenic activity are needed for analysts to agree
on this understanding, and inclusion of the relationship in a WoE could be entirely appropriate.
Response- or receptor-initiated CRAs might identify speculative links between outcomes and possible
stressors. Because the information available to develop the initial conceptual model is preliminary, it may
37 In the IRIS program, evidence integration is defined as the "Integration of animal and human evidence synthesis judgments to
draw an overall conclusion(s) that incorporates inferences drawn on the basis of information on the human relevance of the
animal evidence, cross-stream coherence across the human and animal evidence, susceptibility, and biological
plausibility /MOA
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
32
-------
not be sufficient to characterize stressor-response relationships. Subsequent analysis and iterations of the
conceptual model should reflect ongoing assessment of stressor-response and modifier relationships on
the basis of additional information. Such information, however, might or might not be relevant to the
effects, endpoint, or receptors of concern identified in the conceptual model. The CRA team's judgment
should determine whether inclusion of a stressor-response or modifier relationship in the conceptual
model is warranted. Even if the potential relationship is not ultimately evaluated in the analysis phase of
the assessment, capturing the community stakeholders' concerns may be important to reflect issues
important to the stakeholders and their participation in the CRA process and may also be useful in the
characterization phase for consideration by the risk manager or decision-maker.
Evidence evaluation to assess the strength of causal relationships is an essential consideration in CRAs
because of their scope and complexity. Because sources of evidence might be available from multiple
disciplines, the evaluation of the data required for a CRA should consider WoE both within and across
evidence streams—e.g., epidemiology, toxicology, and mechanistic studies. To consider exposure-
response modifiers, such as psychological or socioeconomic stressors, a multidisciplinary team with
expertise in many areas is required to provide a clear "delimitation of the analysis" so the results
meaningfully inform the analysis plan (Solomon et al., 2016).
Study quality, key elements of which can vary by type of study, should be considered when evaluating the
evidence. Data completeness also might require consideration, such as the breadth of toxicity endpoints or
dose-response data availability for individual components in a chemical mixture. For epidemiological
studies, the WoE evaluation might include evaluation of group characteristics, ability to control for
potential confounders and other sources of bias or temporal relationships, comparability of exposure to
multiple stressors, and disease in the population of interest. Such evaluation is done both for individual
studies and across the body of evidence from multiple studies and may be guided by considerations such
as those outlined in the IRIS Handbook (U.S. EPA, 2022d) or the Preamble to the Integrated Science
Assessments (U.S. EPA, 2015b). Both the IRIS Handbook and the Preamble provide frameworks for
synthesizing evidence within a body of evidence (such as all epidemiology studies) and integration of
evidence across evidence streams (such as epidemiology, toxicology, and mechanistic studies).
As in other risk assessments, important considerations for CRAs include the extent to which the scientific
and technical approaches are reasonable and consistent with the intended application, the tolerance for
potential decision errors, precision requirements, and the use of secondary data collected for purposes
other than the planned assessment. The required level of rigor of the assessment (e.g., screening-level
versus a more robust analysis) can influence data quality needs. The WoE will support or be used to
revise a risk hypothesis and will provide the evidence base necessary to develop plans for the analysis
phase. Evidence evaluation and data quality assessment requirements should be considered throughout the
CRA to ensure the product is fit for purpose.
3.3. Analysis Plan
Analysis plan development is the final stage of planning and problem formulation and generates the
technical work plan for subsequent analytical steps. The analysis plan clearly specifies which analyses
will be done and with which methods (e.g., models) and inputs, who will conduct these activities, and on
what schedule. It addresses conceptual model elements and includes the CRA's purpose, scope, methods,
DQOs, data needed for collection or analysis models using these data, and a description of how data
gaps/uncertainty will be addressed. The conceptual model may be packaged in the same document as the
analysis plan. Important in developing the analysis plan is to coordinate with the appropriate program(s)
to ensure any existing guidance, guidelines, policies, and procedures are followed.
The analysis plan, tailored to the specific CRA, describes how the CRA proposes to address the scientific
questions highlighted in the conceptual model. No "one size fits all" model is proposed. The analysis plan
needs to be responsive to the conceptual model elements. It needs to specify a receptor, outcomes/metrics
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
33
-------
evaluated for the receptor, stressors/stressor groups and stressor sources for the outcome, the exposure
data for the stressors, and dose-response methods for the outcome and its associated stressors. A
description of the evidence related to the elements and pathways included in the conceptual model can
inform development of the analysis plan, an initial appraisal of uncertainties, and identification of any
additional data gathering activities needed before risk characterization. Systematic review or other WoE
methods may be an appropriate approach for some or all of these elements as a way to identify, organize,
and synthesize information. The IRIS Handbook outlines one approach to incorporating systematic
review methods in human health risk assessment (U.S. EPA, 2022d).
The elements described below should be considered for inclusion in the analysis plan:38
CRA Purpose and Scope. A description of the CRA question that risk managers need to answer (and
potential risk management options) and an outline of the general objectives, constraints, and boundaries
of the assessment (identified in Section 2.4 as a "scoping summary statement").
Data Compilation/Collection. A description of the initial WoE39 as it pertains to the conceptual model
pathways includes an appraisal of uncertainties and any data collection needs to ""operational ize" the
conceptual model for chemical and nonchemical stressors (e.g., soil sampling, meteorological data, air
monitoring). Some data, particularly for nonchemical stressors, might be qualitative or only relatively
scalable (Meadows, 2008). Depending on the methods selected for the analysis phase of the CRA, such
data might or might not be accommodated in the CRA (although it may be included in the risk
characterization as an uncertainty). The compatibility of data and methods should be identified as early as
possible in the process.
Preliminary or Previous Risk Estimates. The utility of developing preliminary risk estimates that might
better inform decisions regarding the scope of the assessment, the level of effort warranted, or risk
management options could be considered and is described in Section 2.5. An example is screening out
stressors or stressor groups that contribute little or nothing to risk. While "cutoff' values can be identified
(e.g., < 10 " for cancer, <0.1 hazard quotient [HQ]4"), interactions with other factors, such as preexisting
disease, need to be considered. When preliminary risk estimates result in identifying no risks of concern,
even when all stressors, proposed stressor groupings, and chemical mixtures are considered, further
evaluation might not be warranted.41 Such an outcome suggests resources then might be spent in other,
more critical CRAs (U.S. EPA, 2016b). Additionally, an existing, previously performed assessment could
be available that might provide useful information, including salient details of its analysis approach.
Exposure Assessment. A description of the exposure assessment approach and metrics, including the
population that is the subject of the assessment; whether methods or guidance (e.g., from a particular EPA
program office) will be used; relevant key definitions; temporal considerations (e.g., only chronic
exposure will be evaluated); use of lifestage-specific exposure factors (see U.S. EPA, 2011);42 monitoring
approaches; modeling approaches; key assumptions (e.g., evaluated receptors are "exposed"' to modeled
38 The RISK21 project (https://risk21.org/about-risk21/) of the Health and Environmental Sciences Institute, itself a project
of the International Life Sciences Institute ('https://ilsi.org/'). is a notable resource for identifying questions to develop the analysis
plan (Solomon et al., 2016). The CRA team should reference EPA's Guidance for Quality' Assurance Project Plans (O^iPPs)
(https://www.epa.gov/aualitv/guidance-aualitv-assurance-proiect-plans-epa-qag-5') when developing a QAPP.
39 The WoE and other elements of the analysis plan that are developed during problem formulation and scoping may be revised at
later stages of the risk assessment process as new information becomes available.
40 Usually, an HQ <1 for a single chemical indicates that adverse effects are not likely to occur and is a negligible hazard;
however, an HQ <0.1 is used in screening exposure to multiple chemicals.
41 Hie CRA team should contextualize cumulative risk assessments within the full scope of stressors as holistically as possible,
especially when vulnerable communities constitute part or all of the population of interest. Hie process should also consider the
uncertainties associated with data limitations (e.g., when stressors may be left out due to a lack of data).
42 Hie Exposure Factors Handbook is being updated by chapter (see U.S. EPA (2011) updated chapters, available at
https://www.epa.gov/expobox/about-exposure-factors-handbook').
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
34
-------
or monitored concentrations for certain exposure durations); exposure factors by activity; land use—
current and future; cultural and subsistence practices; and associated variabilities and uncertainties. The
exposure metrics should relate to those identified for effects/outcomes.
Effects/Outcomes Assessment. A description of adverse effect/outcome metrics to be used for stressor
exposures in evaluated receptors (e.g., cancer potency estimates for quantitative dose-response CRA) and
qualitative CRA risk descriptions (e.g., low/medium/high) also might be described. Potential sources of
dose-response data include chemical assessments from the EPA's IRIS database, physical stressor
assessments (e.g., noise) developed by the U.S. National Institute for Occupational Safety and Health, and
the peer-reviewed literature.
Data Quality and Relevance. A discussion of DQOs and needs (U.S. EPA, 2020d) relevant for the subject
CRA should be included. Such descriptions might include robustness of the experimental design for an
epidemiological study, statistical strength, spatial and temporal scales, target receptor specificity, and
other issues specific to the CRA. The primary goal of the Quality Program43 is to ensure that
environmental data are of sufficient quantity and quality to support their intended use. In addition, the
relevance and appropriateness of the data (which includes quality and uncertainty analysis) relative to the
specific CRA question need to be determined. The discussion should address the evaluation of data
quality for existing data, including the DQOs.
Identification and Selection of CRA Techniques/Methods to Apply to Integrated Stressor Groups. A
description of risk assessment methods and procedures that might be used to conduct CRA (e.g., methods
described in U.S. EPA Mixtures Guidance and related documents (U.S. EPA, 1986b, 2000d, 2023a);
Pesticides Cumulative Risk Assessment (U.S. EPA, 2002a); Risk Assessment Guidance for Siiperfiind
(U.S. EPA, 2001b, 2001c); Office of Air and Radiation-Technology Transfer Network (OAR-TTN) Fate,
Exposure, and Risk Analysis44).
Characterization of Baseline Conditions. A description of the characterization of the baseline ("before")
population or ecosystem in a CRA can allow evaluation of "before/after' scenarios. For example,
measures included in ecosystem health indicators in study area streams prior to construction of a new
roadway could be compared with predictions of changes both during and after construction. This element
could take the form of a comparison among risk scenarios or options to inform risk management decisions
when appropriate.
Risk Characterization/Risk Description Plan. The plan for characterizing risk, with attention to the
context and purpose for the assessment. For example, why the CRA is being conducted, what question(s)
the CRA is designed to answer, what it includes and excludes, and, if available, preliminary risk estimates
or comparative risk estimates. A discussion should be included on how the CRA will approach
uncertainty analysis (e.g., qualitatively, quantitatively) and whether sensitivity analysis is possible or
planned for any individual or combined input variables. Important limitations and assumptions of the
planned analysis should also be presented.
Cost Estimates. Anticipated or estimated costs for conducting the CRA, including staffing and expertise
needed and a plan for how costs will be paid.
Schedule. An estimated schedule for conducting the CRA may be provided.
43 The EPA's Quality Program manages the collection, production, evaluation, and use of environmental information. Hie
primary goal of the Quality Program is to ensure that the Agency's environmental information is of sufficient quantity and
quality to support the intended use ("https://www.epa.gov/aualitv/about-epas-qualitv-program'l.
44 See the EPA's Air: Fate, Exposure, and Risk Analysis web page ("https://www.epa.gov/fera).
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
35
-------
3.4. Uncertainties and Variability
CRA requires consideration of uncertainty and variability in terms of both source identification and
analysis approach. CRAs address multiple stressors that could be dissimilar and rely on a variety of
sources of quantitative or qualitative data. These attributes can complicate the evaluation of variability
and uncertainty in CRA.
Potential sources of variability and uncertainty in all CRA phases (exposure assessment, effects
assessment, risk characterization) to be considered, especially if further study of particular aspects of
variability or uncertainty, is required to support decision-making, including:
• Scenario uncertainty resulting from errors, typically of omission, that stem from incorrect or
incomplete specification of the cumulative risk scenario to be evaluated (e.g., all relevant
stressors have not been or cannot be identified)
• Consistency of the overall database for estimating risks associated with important adverse
outcomes
• Variability in exposure characterization (e.g., different durations of exposure, differences in
exposures based on lifestage, activity patterns, dietary preferences)
• Dose metric(s) used for dose-response modeling, route-to-route extrapolation, or extrapolation to
humans; relevant issues include the strength of evidence associating a dose metric with critical
effects, strength of evidence for human relevance of the dose metric (if based on an animal
study), and whether extrapolation to humans relies on chemical-specific evidence or default
allometric relationships
• Human toxicokinetic and toxicodynamic variability
• Model uncertainty attributable to the use of empirical and mechanistic models
• Statistical uncertainty, as characterized by the model-estimated confidence interval, which is
generally due to variability associated with the particular data set
• Input or parameter uncertainty from any errors in characterizing the empirical values used as
inputs to any model used (e.g., engineering, physical, chemical, biological, or behavioral
variables)
• Uncertainty as to whether relationships in the conceptual model are causal and due to lack of or
gaps in data; gaps in data should be considered alongside information from available evidence in
making a determination about the extent of uncertainty in the conceptual model
Early consideration should be given to the approach, methods, and metrics that will be used to evaluate
variability and uncertainty. These approaches might be quantitative or qualitative, and they address data
evaluation, procedures, measures, methods, and models used in the CRA. The extent to which variability
and uncertainty can be evaluated and characterized in the CRA also should be considered. Depending on
the availability of suitable information and the targeted needs of individual CRAs, qualitative discussion
and syntheses of uncertainty might be enhanced by quantitative analyses, including sensitivity analyses
for decisions (e.g., selection of study populations, dose metrics, or model parameters). Modeling
uncertainty using ranges or probability distributions could also be useful when data are adequate.
In developing the approach to address variability and uncertainty for the CRA, the purpose and context
for the assessment, as well as the timeline and resources, should be considered. The variability and
uncertainty analysis should be appropriate for the decision context (IOM, 2013).
Resources for characterizing uncertainty and variability include the following:
• Guiding Principles for Monte Carlo Analysis (U.S. EPA, 1997d)
• Science Policy Council Handbook: Risk Characterization (U.S. EPA, 2000b)
• Process for Conducting Probabilistic Risk Assessment. Part A of Vol. Ill of Risk Assessment
Guidance for Superfund (U.S. EPA, 2001c)
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
36
-------
• An Examination of Risk Assessment Principles and Practices (U.S. EPA, 2004a)
• A Framework for Assessing Health Risk of Environmental Exposures to Children (U.S. EPA,
2006c)
• Guidelines for Human Exposure Assessment (U.S. EPA, 2019)
• ExpoKids: Children's Aggregate Exposure Visualization Tool45
• Risk Assessment Forum White Paper: Probabilistic Risk Assessment Methods and Case Studies
(U.S. EPA, 2014c)
3.5. Next Steps in Cumulative Risk Assessment
These Guidelines provide a generalized approach to the planning and problem formulation of a CRA.
Following completion of the risk analysis, there are two additional steps that may be tailored for
completion of the CRA:
• Risk characterization
• Risk communication
Risk characterization integrates and interprets the results of the analysis phase. It describes the qualitative
or quantitative risk assessment results; lists key assumptions, limitations, variability, and uncertainties
associated with those results; and discusses the application of the results for risk management decisions
(U.S. EPA, 2003b). It is also appropriate for the CRA team to consider whether information in the risk
characterization may be usefully shared with other governmental agencies that are positioned to address
identified risks or exposure-response modifiers. There are many different assessment processes conducted
by federal agencies for chemicals found in the environment, workplace, consumer products, hazardous
waste sites, food, or cosmetics. Most serve to provide public health guidance or recommendations, and
opportunities exist for collaboration and coordination through risk assessment (Shaffer, 2021).
Risk characterization information may be channeled into risk communication. Although communication
begins early in the CRA planning process, the risk communication step requires that decision-makers
have a thorough understanding of the risk characterization; however, risk characterization is not
synonymous with risk communication (U.S. EPA, 2014b). Because the risk characterization may be
technical and complex, communication tools with summarized levels of detail may be developed for risk
communication. These communication tools are often a set of documents tailored to meet the needs of
different stakeholders and audiences (e.g., Siiperfiind Community Involvement Handbook (U.S. EPA,
2020e)). In addition to meeting the requirements of the risk management decision, risk communication
materials and tools may also be tailored to provide information to the public—including individuals,
groups, and other institutions—about not only the cumulative risks identified, but also how people can
mitigate risks through behavioral changes, protective actions, and joint problem solving.
45 See the EPA's ExpoKids: Children's Aggregate Exposure Visualization Tool ("https://www.epa.gov/expobox/expokids-
childrens-aggregate-exposure-visualization-tool'l.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
37
-------
REFERENCES
Adami, H. O., Berry, S. C., Breckenridge, C. B.. Smith, L. L., Swenberg, J. A., Trichopoulos, D„ Weiss, N. S., &
Pastoor, T. P. (2011). Toxicology and epidemiology: Improving the science with a framework for
combining toxicological and epidemiological evidence to establish causal inference. Toxicological
Sciences, 122(2), 223-234. https://doi.org/10.1093/toxsci/kfrl 13
ATSDR (Agency for Toxic Substances and Disease Registry). (2004). Guidance manual for the assessment ofjoint
toxic action of chemical mixtures. U.S. Department of Health and Human Services.
https://stacks.cdc. gov/view/cdc/693 5
ATSDR (Agency for Toxic Substances and Disease Registry). (2011). Principles of community engagement: Second
edition (NIH Publication No. 11-7782). U.S. Department of Health and Human Services.
https://www.atsdr.cdc.gov/communitv-engagement/php/about/index.html
Barrett, E. S., & Padula, A. M. (2019). Joint impact of synthetic chemical and non-chemical stressors on children's
health. Current Environmental Health Reports, (5(4), 225-235. https://doi.org/10.1007/s40572-019-0Q252-6
Barros, N„ Tulve, N. S., Bailey, K„ & Heggem, D. T. (2018a). Outdoor air emissions, land use, and land cover
around schools on tribal lands. International Journal of Environmental Research and Public Health, 16( 1),
Article 36. https://doi.org/10.3390/iierphl6010Q36
Barros, N„ Tulve, N. S., Heggem, D. T„ & Bailey, K. (2018b). Review of built and natural enviromnent stressors
impacting American Indian/Alaska-Native children. Reviews on Environmental Health, 33(4), 349-381.
https://doi.org/10.1515/reveh-2018-0Q34
Brewer, L. E„ Wright, J. M„ Rice, G., Neas, L„ & Teuschler, L. (2017). Causal inference in cumulative risk
assessment: The roles of directed acyclic graphs. Environment International, 102, 30-41.
https://doi.Org/10.1016/i.envint.2016.12.005
Burke, T. A., Cascio, W. E., Costa, D. L., Deener, K„ Fontaine, T. D., Fulk, F. A., Jackson, L. E„ Munns, W. R., Jr.,
Onne-Zavaleta, J., Slimak, M. W., & Zartarian, V. G. (2017). Rethinking enviromnental protection:
Meeting the challenges of a changing world. Environmental Health Perspectives, 125(3), A43-A49.
https://doi.org/10.1289/ehpl465
Callahan, M. A., & Sexton, K. (2007). If cumulative risk assessment is the answer, what is the question?
Environmental Health Perspectives, 115(5), 799-806. https://doi.org/10.1289/ehp.9330
CEQ (Council on Environmental Quality). (2021). Indigenous traditional ecological knowledge and federal decision
making. Executive Office of the President, Office of Science and Technology Policy.
https://www.wliitehouse.gOv/wp-content/uploads/2021/l 1/111521-OSTP-CEO-ITEK-Memo.pdf
CEQ (Council on Environmental Quality). (2022a). Guidance for federal departments and agencies on indigenous
knowledge. Executive Office of the President, Office of Science and Technology Policy.
https://www.wliitehouse.gov/wp-content/uploads/2Q22/12/OSTP-CEO-IK-Guidance.pdf
CEQ (Council on Environmental Quality). (2022b). Implementation of guidance for federal departments and
agencies on indigenous knowledge. Executive Office of the President, Office of Science and Technology
Policy. https://www.wliitehouse.gov/wp-content/uploads/2022/12/IK-Guidance-Implementation-Memo.pdf
Christensen K., Cliristensen, C. H„ Wright, J. M„ Galizia, A., Glenn, B. S., Scott, C. S., Mall, J. K„ Bateson, T. F„
Murphy, P. A., & Cooper, G. S. (2015). The use of epidemiology in risk assessment: Challenges and
opportunities. Human and Ecological Risk Assessment, 21(6), 1644-1663.
https://doi.org/10.1080/10807039.2014.967Q39
CSDH (Commission on Social Determinants of Health). (2008). Closing the gap in a generation: Health equity
through action on the social determinants of health. Final report of the Commission on Social
Determinants of Health. World Health Organization.
http://apps.who.int/iris/bitstream/10665/43943/1/9789241563703 eng.pdf
deFur, P. L„ Evans, G. W., Cohen Hubal, E. A., Kyle, A. D., Morello-Frosch, R. A., & Williams, D. R. (2007).
Vulnerability as a function of individual and group resources in cumulative risk assessment. Environmental
Health Perspectives, 115(5), 817-824. https://doi.org/10.1289/ehp.9332
EFSA Scientific Committee. (2017). Scientific opinion: Guidance on the use of the weight of evidence approach in
scientific assessments. EFSA Journal, 15(8), Article e04971. https://doi.Org/10.2903/i.efsa.2017.4971
Embry, M. R„ Bacliman, A. N, Bell, D. R., Boobis, A. R., Cohen, S. M„ Dellarco, M., Dewhurst, I. C., Doerrer, N.
G., Hines, R. N„ Moretto, A., Pastoor, T. P., Phillips, R. D., Rowlands, J. C., Tanir, J. Y., Wolf, D. C., &
Doe, J. E. (2014). Risk assessment in the 21st century: Roadmap and matrix. Critical Reviews in
Toxicology, -/-/(Suppl. 3), 6-16. https://doi.org/10.3l69/10408444.2014.931924
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025 38
-------
EOP (Executive Office of the President). (1994). Executive Order 12898: Federal actions to address environmental
justice in minority populations and low-income populations. Federal Register. 59(32). 7629-7633.
https://www.epa.gov/laws-regulations/summarv-executive-order-12898-federal-actions-address-
enviromnental-iustice
EOP (Executive Office of the President). (1997). Executive Order 13045: Protection of children from enviromnental
health risks and safety risks. Federal Register, 62(78), 19885-19888. https://www.epa.gov/laws-
regulations/summarv-executive-order-13045-protection-children-environmental-health-risks-and
EOP (Executive Office of the President). (2021). Executive Order 13985: Advancing racial equity and support for
underserved communities through the federal government. Federal Register, 8<5(14), 7009-7013.
https://www.federalregister.gOv/documents/2021/01/25/2021-01753/advancing-racial-eauitv-and-support-
for-underserved-communities-through-the-federal-government
EOP (Executive Office of the President). (2023a). Executive Order 14091: Further advancing racial equity and
support for underserved communities through the federal government. Federal Register. 88(35), 10825-
10833. https://www.federalregister.gov/documents/2023/02/22/2023-03779/further-advancing-racial-
eauitv-and-support-for-underserved-communities-through-the-federal
EOP (Executive Office of the President). (2023b). Executive Order 14096: Revitalizing our nation's commitment to
environmental justice for all. Federal Register, 88(80), 25251-25261.
https://www.federalregister.gOv/documents/2023/04/26/2023-08955/revitalizing-our-nations-commitment-
to-enviromnental-iustice-for-all
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(A), Article 389. https://doi.org/10.3390/iierphl404Q389
Gallagher, S. S., Rice, G. E., Scarano, L. J., Teuschler, L. K„ Bollweg, G., & Martin, L. (2015). Cumulative risk
assessment lessons learned: A review of case studies and issue papers. Chemosphere, 120, 697-705.
https://doi.Org/10.1016/i.chemosphere.2014.10.030
Gee, G. C., & Payne-Sturges, D. C. (2004). Enviromnental health disparities: A framework integrating psychosocial
and enviromnental concepts. Environmental Health Perspectives, 112(11), 1645-1653.
https://doi.org/10.1289/ehp.7074
Ginsberg, G. L., Dietert, R. R., & Sonawane, B. R. (2014). Susceptibility based upon chemical interaction with
disease processes: Potential implications for risk assessment. Current Environmental Health Reports 7(4),
314-324. https://doi.org/10.1007/s40572-014-003Q-z
Hibbert, K., & Tulve, N. S. (2019). State-of-the-science review of non-chemical stressors found in a child's social
environment. International Journal of Environmental Research and Public Health, 16(22), Article 4417.
https://doi.org/10.3390/iierphl6224417
IOM (Institute of Medicine). (2013). Environmental decisions in the face of uncertainty. The National Academies
Press, https://doi.org/10.17226/12568
Juster, R. P., McEwen, B. S., & Lupien, S. J. (2010). Allostatic load biomarkers of chronic stress and impact on
health and cognition. Neuroscience andBiobehavioral Reviews, 35(1), 2-16.
https://doi.Org/10.1016/i.neubiorev.2009.10.002
Kashuba, R., Menzie, C., & Martin, L. (2021). Risk of cardiovascular disease is driven by different combinations of
environmental, medical and behavioral factors: Building a conceptual model for cumulative risk
assessment. Human and Ecological Risk Assessment, 27(1), 1902-1925.
https://doi.org/10.1080/10807039.2021.1925Q83
Kasperson, J. X., Kasperson, R. E„ & Turner, B. L„ 2nd (Eds.). (1995). Regions at risk: Comparisons of threatened
environments. United Nations University Press.
Kasperson, R. E„ Dow, K., Archer, E. R. M„ Caceres, E„ Downing, T. E., Elmqvist, T., Eriksen, S., Folke, C., Han
G., Iyengar, K„ Vogel, C., Wilson, K. A., & Ziervogel, G. (2005). Vulnerable peoples and places. In R. M.
Hassan, R. J. Scholes, & N. Ash (Eds.), Ecosystems and human well-being: Current state and trends:
Findings of the Condition and Trends Working Group (pp. 146-162). Island Press.
Levy, J. I. (2008). Is epidemiology the key to cumulative risk assessment? Risk Analysis. 28(6). 1507-1513.
https://doi.org/10.111 l/i."l539-6924.2008.01121.x
Lichtveld, K., Thomas, K„ & Tulve, N. S. (2018). Chemical and non-chemical stressors affecting childhood obesity:
A systematic scoping review. Journal of Exposure Science and Environmental Epidemiology. 28(1), 1-12.
https://doi.org/10.1038/ies.2Q17.18
Linder, S. H., Delclos, G., & Sexton, K. (2010). Making causal claims about environmentally induced adverse
effects. Human and Ecological Risk Assessment. 16(1). 35-52. https://doi.org/10.1080/108070309Q3459288
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025 39
-------
Linder, S. H., & Sexton, K. (2011). Conceptual models for cumulative risk assessment. American Journal of Public
Health, 70i(Suppl. 1), S74-S81. https://doi.org/10.2105/aiph.2011.30Q318
Linkov, I., & Satterstrom, F. K. (2006). Weight of evidence: What is the state of the science? Risk Analysis, 26( 3),
573-575. https://doi.org/10.1111/i. 1539-6924.2006.00756.X
Lokke, H. (2010). Novel methods for integrated risk assessment of cumulative stressors — Results from the
NoMiracle project. Science of The Total Environment, -/08(18), 3719-3724.
https://doi.Org/10.1016/i.scitotenv.2010.05.009
Meadows, D. H. (2008). Thinking in systems: A primer. Chelsea Green Publishing.
Meek, M. E„ Boobis, A. R., Crofton, K. M., Heinemeyer, G., Raaij, M. V., & Vickers, C. (2011). Risk assessment
of combined exposure to multiple chemicals: A WHO/IPCS framework. Regulatory Toxicology and
Pharmacology, 60(2 Suppl.), S1-S14. https://doi.org/10.1016/i.vrtph.2011.03.010
Menzie, C. A., MacDonell, M. M., & Mumtaz, M. (2007). A phased approach for assessing combined effects from
multiple stressors. Environmental Health Perspectives, 115(5), 807-816. https://doi.org/10.1289/ehp.9331
Morello-Frosch, R., & Shenassa, E. D. (2006). The environmental "riskscape" and social inequality: Implications for
explaining maternal and child health disparities. Environmental Health Perspectives, 114(8), 1150-1153.
https://doi.org/10.1289/ehp.8930
Morello-Frosch, R., Zuk, M„ Jerrett, M., Shamasunder, B„ & Kyle, A. D. (2011). Understanding the cumulative
impacts of inequalities in enviromnental health: Implications for policy. Health Affairs, 30(5), 879-887.
https://doi.org/10.1377/1i1t1iafF.2011.0153
Moretto, A., Bacliman A., Boobis, A., Solomon, K. R„ Pastoor, T. P., Wilks, M. F., & Embry, M. R. (2017). A
framework for cumulative risk assessment in the 21st century. Critical Reviews in Toxicology, 47(2), 85-
97. https://doi.org/10.1080/10408444.2Q16.1211618
NASEM (National Academies of Sciences, Engineering, and Medicine). (2017). Using 21st century science to
improve risk-related evaluations. The National Academies Press, https://doi.org/10.17226/24635
NASEM (National Academies of Sciences, Engineering, and Medicine). (2023). Building confidence in new
evidence streams for human health risk assessment: Lessons learned from laboratory mammalian toxicity
tests. The National Academies Press, https://doi.org/10.17226/26906
National EPA-TSC (Tribal Science Council). (2006). Paper on tribal issues related to tribal traditional lifewavs,
risk assessment, and health & well being: Documenting what we've heard (EPA/620/R-06/004). U.S.
Enviromnental Protection Agency, https://nepis.epa.gov/Exe/ZvPURL.cgi?Dockev=P1006LIF.txt
NEJAC (National Enviromnental Justice Advisory Council). (2004). Ensuring risk reduction in communities with
multiple stressors: Environmental justice and cumulative risks/impacts. U.S. Enviromnental Protection
Agency, https://www.epa.gov/sites/production/files/2015-04/documents/ensuringriskreducationneiac.pdf
Nilsen, F. M., Frank, J., & Tulve, N. S. (2020a). A systematic review and meta-analysis investigating the
relationship between exposures to chemical and non-chemical stressors during prenatal development and
childhood externalizing behaviors. International Journal of Environmental Research and Public Health,
17(1). Article 2361. https://doi.org/10.3390/iierphl7072361
Nilsen, F. M„ Ruiz, J. D. C., & Tulve, N. S. (2020b). A meta-analysis of stressors from the total enviromnent
associated with children's general cognitive ability. International Journal of Environmental Research and
Public Health, 77(15). Article 5451. https://doi.org/10.3390/iierphl7155451
Nilsen, F. M„ & Tulve, N. S. (2020). A systematic review and meta-analysis examining the interrelationships
between chemical and non-chemical stressors and inherent characteristics in children with ADHD.
Environmental Research, 180, Article 108884. https://doi.Org/10.1016/i.envres.2019.108884
Norton S. B„ Cormier, S. M., & Suter, G. W., 2nd (Eds.). (2014). Ecological causal assessment. CRC Press.
https://doi.org/10.1201/bl76Q3.
NRC (National Research Council). (1980). Principles of toxicological interactions associated with multiple
chemical exposures. The National Academies Press.
https://nap.nationalacadeniies.org/catalog/20435/principles-of-toxicological-interactions-associated-with-
multiple-chemical-exposures
NRC (National Research Council). (1983). Risk assessment in the federal government: Managing the process. The
National Academies Press, https://doi.org/10.17226/366
NRC (National Research Council). (1993). Pesticides in the diets of infants and children. The National Academies
Press, https://doi.org/10.17226/2126
NRC (National Research Council). (1996). Understanding risk: Informing decisions in a democratic society. The
National Academies Press, https://doi.org/10.17226/5138
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025 40
-------
NRC (National Research Council). (2006). Fluoride in drinking water: A scientific re\>iew of EPA's standards. The
National Academies Press, https://doi.org/10.17226/11571
NRC (National Research Council). (2007). Toxicity testing in the 21st century: A vision and a strategy. The National
Academies Press, https://doi.org/10.17226/11970
NRC (National Research Council). (2008a). Phthalates and cumulative risk assessment: The tasks ahead. The
National Academies Press, https://doi.org/10.17226/12528
NRC (National Research Council). (2008b). Public participation in environmental assessment and decision making.
The National Academies Press, https://doi.org/10.17226/12434
NRC (National Research Council). (2009). Science and decisions: Advancing risk assessment. The National
Academies Press, https://doi.org/10.17226/12209
NRC (National Research Council). (201 la). Improving health in the United States: The role of health impact
assessment. The National Academies Press, https://doi.org/10.17226/13229
NRC (National Research Council). (2011b). Sustainabilitv and the U.S. EPA. The National Academies Press.
https://doi.org/10.17226/13152
NRC (National Research Council). (2012). Exposure science in the 21st century: A vision and a strategy. The
National Academies Press, https://doi.org/10.17226/13507
Payne-Sturges, D. C., Korfmacher, K. S., Cory-Slechta, D. A., Jimenez, M., Symanski, E.. Carr Shmool, J. L..
Dotson-Newman, O., Clougherty, J. E.. French, R., Levy, J. I., Laumbach, R„ Rodgers, K„ Bongiovanni,
R., & Scammell, M. K. (2015). Engaging communities in research on cumulative risk and social stress-
environment interactions: Lessons learned from EPA's STAR program. Environmental Justice. 8(6), 203-
212. https://doi.org/10.1089/env.2015.0Q25
Payne-Sturges, D. C., Scammell, M. K., Levy, J. I., Cory-Slechta, D. A., Symanski, E„ Carr Shmool, J. L.,
Laumbach, R., Linder, S., & Clougherty, J. E. (2018). Methods for evaluating the combined effects of
chemical and nonchemical exposures for cumulative enviromnental health risk assessment. International
Journal of Environmental Research and Public Health, 75(12), Article 2797.
https://doi.org/10.3390/iierphl5122797
PCCRARM (Presidential/Congressional Commission on Risk Assessment and Risk Management). (1997).
Framework for environmental health risk management: Final report. Volumes 1 and 2 (EPA/950/R-
97/011). U.S. Enviromnental Protection Agency.
https://nepis.epa.gov/Exe/ZvPURL.cgi?Dockev=9101KlC3.txt
Rodriquez, E. J., Kim, E. N„ Sumner, A. E„ Napoles, A. M„ & Perez-Stable, E. J. (2019). Allostatic load:
Importance, markers, and score determination in minority and disparity populations. Journal of Urban
Health, 96(Suppl. 1), 3-11. https://doi.org/10.1007/sll524-019-0Q345-5
Ruiz, J. D. C., Quackenboss, J. J., & Tulve, N. S. (2016). Contributions of a child's built, natural, and social
environments to their general cognitive ability: A systematic scoping review. PLoS One, 11(2). Article
e0147741. https://doi.org/10.1371/iournal.pone.0147741
Ryan, P. B., Burke, T. A., Cohen Hubal, E. A., Cura, J. J., & McKone, T. E. (2007). Using biomarkers to inform
cumulative risk assessment. Environmental Health Perspectives, 115(5), 833-840.
https://doi.org/10.1289/ehp.9334
Schafer, R. B„ Jackson, M., Juvigny-Khenafou, N„ Osakpolor, S. E., Posthuma, L., Schneeweiss, A., Spaak, J., &
Vinebrooke, R. (2023). Chemical mixtures and multiple stressors: Same but different? Environmental
Toxicology and Chemistry, 42(9), 1915-1936. https://doi.org/10.10Q2/etc.5629
Schulz, A. J., Kannan, S., Dvonch, J. T., Israel, B. A., Allen, A., 3rd, James, S. A., House, J. S., & Lepkowski, J.
(2005). Social and physical enviromnents and disparities in risk for cardiovascular disease: The Healthy
Enviromnents Partnership conceptual model. Environmental Health Perspectives, 113(12), 1817-1825.
https://doi.org/10.1289/ehp.7913
Segal, D„ Lin, Y. S., Ginsberg, G., & Sonawane, B. (2015). A conceptual framework for evaluating the interaction
of a chemical and nonchemical stressor in human health risk assessments: A case study for lead and
psychosocial stress. Human and Ecological Risk Assessment, 21(1), 1840-1868.
https://doi.Org/10.1080/10807039.2014.992852
Sexton, K. (2015). Cumulative health risk assessment: Finding new ideas and escaping from the old ones. Human
and Ecological Risk Assessment, 21(A), 934-951. https://doi.org/10.1080/10807039.2Q14.946346
Sexton, K., & Hattis, D. (2007). Assessing cumulative health risks from exposure to enviromnental mixtures—three
fundamental questions. Environmental Health Perspectives, 115(5), 825-832.
https://doi.org/10.1289/ehp.9333
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025 41
-------
Sexton, K„ & Linder, S. H. (2010). The role of cumulative risk assessment in decisions about enviromnental justice.
International Journal of Environmental Research and Public Health. 7(11), 4037-4049.
https://doi.org/10.3390/iierph7114Q37
Sexton, K., & Linder, S. H. (2011). Cumulative risk assessment for combined health effects from chemical and
nonchemical stressors. American Journal of Public Health. /0/(Suppl. 1), S81-S88.
https://doi.org/10.2105/aiph.2011.300118
Shaffer, R. M. (2021). Enviromnental health risk assessment in the federal government: A visual overview and a
renewed call for coordination. Environmental Science and Technology, 55(16), 10923-10927.
https://doi.org/10.1021/acs.est. Ic01955
Simmons, J. E„ Teuschler, L. K., Gennings, C., Speth, T. F„ Richardson, S. D„ Miltner, R. J., Narotsky, M. G.,
Schenck, K. D., Hunter, E. S., 3rd, Hertzberg, R. C., & Rice, G. (2004). Component-based and whole-
mixture techniques for addressing the toxicity of drinking-water disinfection by-product mixtures. Journal
of Toxicology and Environmental Health, Part A, (57(8-10), 741-754.
https://doi.org/10.1080/1528739049Q428215
Solar, O., & Irwin, A. (2010). A conceptual framework for action on the social determinants of health: Social
determinants of health discussion paper 2 (policy and practice). World Health Organization.
http://apps.who.int/iris/bitstream/10665/44489/l/9789241500852 eng.pdf
Solomon, K. R., Wilks, M. F., Bachman, A., Boobis, A., Moretto, A., Pastoor, T. P., Phillips, R., & Embry, M. R.
(2016). Problem formulation for risk assessment of combined exposures to chemicals and other stressors in
humans. Critical Reviews in Toxicology. 46( 10), 835-844. https://doi.org/10.1080/10408444.2Q16.1211617
Stahl, C. H., & Cimorelli, A. J. (2020). Environmental public policy making exposed: A guide for decision makers
and interested citizens. Springer. https://doi.org/10.1007/978-3-030-3213Q-7
Suter, G. W., 2nd. (1999). Developing conceptual models for complex ecological risk assessments. Human and
Ecological Risk Assessment. 5(2), 375-396. https://doi.org/10.1080/10807Q39991289491
Suter, G. W., 2nd. (2006). Ecological risk assessment (2nd ed.). CRC Press. https://doi.org/10.1201/9781420Q12569
Teuschler, L. K„ Rice, G. E., Wilkes, C. R., Lipscomb, J. C., & Power, F. W. (2004). A feasibility study of
cumulative risk assessment methods for drinking water disinfection by-product mixtures. Journal of
Toxicology and Environmental Health, Part A. (57(8-10), 755-777.
https://doi.org/10.1080/1528739049Q428224
Tulve, N. S., Ruiz, J. D. C., Lichtveld, K„ Darney, S. P., & Quackenboss, J. J. (2016). Development of a conceptual
framework depicting a child's total (built, natural, social) enviromnent in order to optimize health and well-
being. Journal of Environment and Health Sciences. 2(2). 1-8.
U.S. EPA (U.S. Environmental Protection Agency). (1986a). Guidelines for mutagenicity risk assessment
(EPA/630/R-98/003). Risk Assessment Forum.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=23160
U.S. EPA (U.S. Environmental Protection Agency). (1986b). Guidelines for the health risk assessment of chemical
mixtures (EPA/630/R-98/002). Risk Assessment Forum.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=22567
U.S. EPA (U.S. Environmental Protection Agency). (1989). Risk assessment guidance for Superfund: Volume I:
Human health evaluation manual (part A) (EPA/540/1-89/002). Office of Emergency and Remedial
Response, https://www.epa.gov/risk/risk-assessment-guidance-superfund-rags-part
U.S. EPA (U.S. Environmental Protection Agency). (1991a). Guidelines for developmental toxicity risk assessment
(EPA/600/FR-91/001). Risk Assessment Forum.
https://cfpub.epa.gov/ncea/risk/recordisplav.cfm?deid=23162
U.S. EPA (U.S. Environmental Protection Agency). (1991b). Risk assessment guidance for Superfund: Volume I -
Human health evaluation manual (part B, development of risk-based preliminary remediation goals)
(EPA/540/R-92/003). Office of Emergency and Remedial Response, https://www.epa.gov/risk/risk-
assessment-guidance-superfund-rags-part-b
U.S. EPA (U.S. Environmental Protection Agency). (1994). Methods for derivation of inhalation reference
concentrations and application of inhalation dosimetry (EPA/600/8-90/066F). Office of Research and
Development, http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=71993
U.S. EPA (U.S. Environmental Protection Agency). (1995a). Environmental justice strategy: Executive Order
12898. Administration and Resources Management, Office of Enviromnental Justice.
https://nepis.epa.gov/Exe/ZvPURL.cgi?Dockev=40000AEB.txt
U.S. EPA (U.S. Environmental Protection Agency). (1995b). Policy on evaluating health risks to children.
https://www.epa.gov/cliildren/epas-policv-cliildrens-health
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025 42
-------
U.S. EPA (U.S. Environmental Protection Agency). (1996a). Attachment A: Conceptual site model summary. In
Soil screening guidance: User's guide (2nd ed.). (EPA/540/R-96/018). Office of Solid Waste and
Emergency Response. https://nepis.epa.gov/Exe/ZvPURL.cgi?Dockev=100027WI.txt
U.S. EPA (U.S. Environmental Protection Agency). (1996b). Guidelines for reproductive toxicity risk assessment
(EPA/630/R-96/009). Risk Assessment Forum, https://www.epa.gov/risk/guidelines-reproductive-toxicitv-
risk-assessment
U.S. EPA (U.S. Environmental Protection Agency). (1997a). Cumulative risk assessment guidance-Phase I
planning and scoping (Memorandum, July 3). https://www.epa.gov/sites/production/files/2015-
01/documents/cumulrisk.pdf
U.S. EPA (U.S. Environmental Protection Agency). (1997b). Ecological risk assessment guidance for Superfund:
Process for designing and conducting ecological risk assessments (EPA 540-R-97-006). Solid Waste and
Emergency Response, https://semspub.epa.gov/work/HO/157941 .pdf
U.S. EPA (U.S. Environmental Protection Agency). (1997c). Guidance on cumulative risk assessment. Part 1.
Planning and scoping. Science Policy Council, https://www.epa.gov/risk/guidance-cumulative-risk-
assessment-part-1 -planning-and-scoping
U.S. EPA (U.S. Environmental Protection Agency). (1997d). Guiding principles for Monte Carlo analysis
(EPA/630/R-97/001). Risk Assessment Forum, https://www.epa.gov/risk/guiding-principles-monte-carlo-
analvsis
U.S. EPA (U.S. Environmental Protection Agency). (1998a). Guidelines for ecological risk assessment (EPA/630/R-
95/002F). Risk Assessment Forum, https://www.epa.gov/risk/guidelines-ecological-risk-assessment
U.S. EPA (U.S. Environmental Protection Agency). (1998b). Guidelines for neurotoxicity risk assessment
(EPA/630/R-95/001F). Risk Assessment Forum, https://www.epa.gov/risk/guidelines-neurotoxicitv-risk-
assessment
U.S. EPA (U.S. Environmental Protection Agency). (1998c). Methodology for assessing health risks associated with
multiple pathways of exposure to combustor emissions (EPA/600/R-98/137). Office of Research and
Development, National Center for Enviromnental Assessment.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=55525
U.S. EPA (U.S. Environmental Protection Agency). (1999). Guidance for identifying pesticide chemicals and other
substances that have a common mechanism of toxicity, https://www.epa.gov/pesticide-science-and-
assessing-pesticide-risks/guidance-identifving-pesticide-chemicals-and-other
U.S. EPA (U.S. Environmental Protection Agency). (2000a). Conducting a risk assessment of mixtures of
disinfection by-products (DBPs) for drinking water treatment systems (EPA/600/R-03/040). Office of
Research and Development, National Center for Enviromnental Assessment.
https://cfpub.epa.gov/ncea/risk/recordisplav.cfm?deid=57054&CFID=80080479&CFTOKEN=68320311
U.S. EPA (U.S. Environmental Protection Agency). (2000b). Risk characterization handbook (EPA 100-B-00-002).
Office of Research and Development, Office of Science Policy, https://www.epa.gov/risk/risk-
characterization-handbook
U.S. EPA (U.S. Environmental Protection Agency). (2000c). Stressor identification guidance document
(EPA/822/B-00/025). Office of Water, Office of Research and Development.
https://cfpub.epa.gov/ncea/risk/recordisplav.cfm?deid=20685
U.S. EPA (U.S. Environmental Protection Agency). (2000d). Supplementary guidance for conducting health risk
assessment of chemical mixtures (EPA/630/R-00/002). Risk Assessment Forum.
https ://cfpub .epa. gov/ncea/risk/recordisplav .cfm?deid=20533
U.S. EPA (U.S. Environmental Protection Agency). (2001a). General principles for performing aggregate exposure
and risk assessments. Office of Pesticide Programs, https://www.epa.gov/pesticide-science-and-assessing-
pesticide-risks/general-principles-performing-aggregate-exposure
U.S. EPA (U.S. Environmental Protection Agency). (2001b). Risk assessment guidance for Superfund: Volume I:
Human health evaluation manual (part D, standardizing planning, reporting, and review of superfund risk
assessments) (Publication 9285.7-47). Office of Emergency and Remedial Response.
https://www.epa.gov/risk/risk-assessment-guidance-superfund-rags-part-d
U.S. EPA (U.S. Environmental Protection Agency). (2001c). Risk assessment guidance for Superfund: Volume III -
Part A, process for conducting probabilistic risk assessment (EPA 540-R-02-002; OSWER 9285.7-45).
Office of Emergency and Remedial Response, https://www.epa.gov/risk/risk-assessment-guidance-
superfund-rags-volume-iii-part
U.S. EPA (U.S. Environmental Protection Agency). (2001d). Stakeholder involvement & public participation at the
U.S. EPA: Lessons learned, barriers, & innovative approaches (EPA-100-R-00-040). Office of Policy,
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025 43
-------
Economics, and Innovation, https://www.epa.gov/sites/default/files/2015-09/documents/stakeholder-
involvement-public-participation-at-epa.pdf
U.S. EPA (U.S. Environmental Protection Agency). (2002a). Guidance on cumulative risk assessment of pesticide
chemicals that have a common mechanism of toxicity. Office of Pesticide Programs.
https://www.epa.gov/pesticide-science-and-assessing-pesticide-risks/guidance-cumulative-risk-assessment-
pesticide
U.S. EPA (U.S. Environmental Protection Agency). (2002b). Lessons learned on planning and scoping for
environmental risk assessments (EPA/600/R-02/802). Office of Science Policy.
https://nepis.epa.gov/Exe/ZvPURL.cgi?Dockev=P1008PP7.txt
U.S. EPA (U.S. Environmental Protection Agency). (2003a). Developing relative potency factors for pesticide
mixtures: Biostatistical analyses of joint dose-response (EPA/600/R-03/052). Office of Research and
Development, National Center for Enviromnental Assessment.
https://cfpub.epa.gov/ncea/risk/recordisplav.cfm?deid=66273&CFID=80059878&CFTOKEN=47378702
U.S. EPA (U.S. Environmental Protection Agency). (2003b). Framework for cumulative risk assessment
(EPA/630/P-02/001F). Risk Assessment Forum.
https://nepis.epa.gov/Exe/ZvPURL.cgi?Dockev=30004TJH.txt
U.S. EPA (U.S. Environmental Protection Agency). (2003c). Public involvement policy of the U.S. Environmental
Protection Agency (EPA 233-B-03-002). Office of Policy, Economics, and Innovation.
https://nepis.epa.gov/Exe/ZvPURL.cgi?Dockev=40000P9G.txt
U. S. EPA (U.S. Environmental Protection Agency). (2004a). An examination of EPA risk assessment principles and
practices: Staffpaper prepared for the U.S. Environmental Protection Agency by members of the Risk
Assessment Task Force (EPA/100/B-04/001). Office of the Science Advisor.
https://nepis.epa.gov/Exe/ZvPURL.cgi?Dockev=100045MJ.txt
U.S. EPA (U.S. Environmental Protection Agency). (2004b). Risk assessment guidance for Superfund: Volume I:
Human health evaluation manual (part E, supplemental guidance for dermal risk assessment). Office of
Superfund Remediation and Technology Innovation, https://www.epa.gov/risk/risk-assessment-guidance-
superfund-rags-part-e
U.S. EPA (U.S. Environmental Protection Agency). (2005). Guidelines for carcinogen risk assessment (EPA/630/P-
03/00IF). Risk Assessment Forum, https://www.epa.gov/risk/guidelines-carcinogen-risk-assessment
U.S. EPA (U.S. Environmental Protection Agency). (2006a). ^Air Toxics Risk Assessment reference library: Volume
3: Community-scale assessment (EPA-452/K-06-001C). Office of Air Quality Planning and Standards.
https://www.epa.gov/sites/production/files/2013-08/documents/volume 3 communitvassess.pdf
U.S. EPA (U.S. Environmental Protection Agency). (2006b). Cumulative risk from chloroacetanilide pesticides.
Office of Pesticide Programs.
https://web.arcliive.Org/web/20150417192548/littp://epa.gov/pesticides/cumulative/cliloro cumulative risk
.pdf
U.S. EPA (U.S. Environmental Protection Agency). (2006c). A framework for assessing health risk of
environmental exposures to children (EPA/600/R-05/093F). Office of Research and Development, National
Center for Enviromnental Assessment, http://cfpub.epa. gov/ncea/cfm/recordisplav.cfm?deid= 158363
U.S. EPA (U.S. Environmental Protection Agency). (2006d). Guidance on systematic planning using the Data
Quality Objectives Process: EPA OA/G-4 (EPA/240/B-06/001). Office of Enviromnental Information.
https://www.epa.gov/qualitv/guidance-svstematic-planning-using-data-qualitv-obiectives-process-epa-qag-
4
U.S. EPA (U.S. Environmental Protection Agency). (2006e). Organophosphorus cumulative risk assessment - 2006
update. Office of Pesticide Programs.
https://vosemite.epa.gov/oann/ali/ali web docket.nsf/Filing+Additional+Attacliments/55EB8690337D456
185258A0C00440F0F/$File/RX%2018%20Revised%20Organophosphorous%20Pesticide%20Cumulative
%20Risk%20 Assessment.pdf
U.S. EPA (U.S. Environmental Protection Agency). (2006f). Triazine cumulative risk assessment. Office of
Pesticide Programs, https://arcliive.epa.gov/pesticides/reregistration/web/pdf/triazine cumulative risk.pdf
U.S. EPA (U.S. Environmental Protection Agency). (2007a). Concepts, methods, and data sources for cumulative
health risk assessment of multiple chemicals, exposures and effects: A resource document (EPA/600/R-
06/013F). Office of Research and Development, National Center for Enviromnental Assessment.
littp://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=190187
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025 44
-------
U.S. EPA (U.S. Environmental Protection Agency). (2007b). RevisedN-methvl carbamate cumulative risk
assessment. Office of Pesticide Programs.
https://arcliive.epa.gov/pesticides/reregistration/web/pdf/mnc revised cra.pdf
U.S. EPA (U.S. Environmental Protection Agency). (2009). Integrated Science Assessment for particulate matter
(EPA/600/R-08/139F). Office of Research and Development, National Center for Environmental
Assessment. https://cfpub.epa.gov/ncea/risk/recordisplav.cfm?deid=216546
U.S. EPA (U.S. Environmental Protection Agency). (2010). EPA's action development process: Interim guidance on
considering environmental justice during the development of an action.
https://www.epa.gov/enviromnentaliustice/interim-guidance-considering-enviromnental-iustice-during-
development-action
U.S. EPA (U.S. Environmental Protection Agency). (2011). Exposure factors handbook: 2011 edition (EPA/600/R-
09/052F). Office of Research and Development, National Center for Enviromnental Assessment.
https://cfpub.epa.gov/ncea/risk/recordisplav.cfm?deid=236252
U.S. EPA (U.S. Environmental Protection Agency). (2013). America's children and the environment: Third edition
(EPA 240-R-13-001). https://www.epa.gov/sites/production/files/2015-06/documents/ace3 2013.pdf
U.S. EPA (U.S. Environmental Protection Agency). (2014a). Cumulative risk webinar series: What we learned
(EPA/600/R-14/212). https://www.epa.gov/sites/production/files/2015-10/documents/cra-webinar-
suimnarv.pdf
U.S. EPA (U.S. Environmental Protection Agency). (2014b). Framework for human health risk assessment to
inform decision making (EPA/100/R-14/001). Office of the Science Advisor, Risk Assessment Forum.
https://nepis.epa.gov/Exe/ZvPURL.cgi?Dockev=P100IEKK.txt
U.S. EPA (U.S. Environmental Protection Agency). (2014c). Risk assessment forum white paper: Probabilistic risk
assessment methods and case studies (EPA/100/R-14/004). Office of the Science Advisor, Risk
Assessment Forum, https://www.epa.gov/scientific-leadersliip/risk-assessment-forum-wliite-paper-
probabilistic-risk-assessment-methods-and
U.S. EPA (U.S. Environmental Protection Agency). (2015a). Peer review handbook: 4th edition (EPA/100/B-
15/001). Science and Technology Policy Council, https://www.epa.gov/scientific-leadersliip/peer-review-
handbook-4th-edition-2015-0
U.S. EPA (U.S. Environmental Protection Agency). (2015b). Preamble to the Integrated Science Assessments
(EPA/600/R-15/067). Office of Research and Development, National Center for Enviromnental
Assessment. https://cfpub.epa.gov/ncea/isa/recordisplav.cfm?deid=310244
U.S. EPA (U.S. Environmental Protection Agency). (2016a). EJ 2020 Action Agenda: The U.S. EPA's
environmental justice strategic plan for 2016 - 2020. https://www.epa.gov/environmentaliustice/ei-2020-
action-agenda-epas-environmental-iustice-strategy-0
U.S. EPA (U.S. Environmental Protection Agency). (2016b). Pesticide cumulative risk assessment: Framework for
screening analysis. Office of Pesticide Programs, https://www.epa.gov/pesticide-science-and-assessing-
pesticide-risks/pesticide-cumulative-risk-assessment-framework
U.S. EPA (U.S. Environmental Protection Agency). (2016c). Weight of evidence for ecological assessment
(EPA/100/R-16/001). Office of the Science Advisor, Risk Assessment Forum.
https ://nepis. epa. gov/Exe/ZvPURL ,c gi?Dockev=P 100 SFXR.txt
U.S. EPA (U.S. Environmental Protection Agency). (2019). Guidelines for human exposure assessment
(EPA/100/B-19/001). Risk Assessment Forum, https://www.epa.gov/risk/guidelines-human-exposure-
assessment
U.S. EPA (U.S. Environmental Protection Agency). (2020a). Clean Air Act title IV-Noise pollution. Retrieved May
20, 2020 from https://www.epa.gov/clean-air-act-overview/clean-air-act-title-iv-noise-pollution
U.S. EPA (U.S. Environmental Protection Agency). (2020b). Cumulative Risk Assessment 2012 webinar series.
Retrieved May 20, 2020 from https://arcliive.epa.gov/ncer/multimedia/videos/web/litml/cumulative-
risk/webinar/2012/index.html
U.S. EPA (U.S. Environmental Protection Agency). (2020c). EPA'sReport on the Environment (ROE). Retrieved
May 20, 2020 from https://www.epa.gov/report-enviromnent
U.S. EPA (U.S. Environmental Protection Agency). (2020d). How EPA manages the quality of its environmental
data. Retrieved May 20, 2020 from https://! 9ianuarv2021 snapshot.epa. gov/qualitv .html
U.S. EPA (U.S. Environmental Protection Agency). (2020e). Superfund community involvement handbook (OLEM
9230.0-51). Office of Superfund Remediation and Technology Innovation.
https ://semspub. epa. gov/work/HO/100002505 .pdf
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025 45
-------
U.S. EPA (U.S. Environmental Protection Agency). (2021a). 2021 Policy on children's health: October 5, 2021.
Office of the Administrator, https://www.epa.gov/sites/default/files/2014-
05/documents/1995 childrens health policy statement.pdf
U.S. EPA (U.S. Environmental Protection Agency). (2021b). List of alternative test methods and strategies (or new
approach methodologies [NAMs]). Second update: February 4th, 2021. Office of Chemical Substances and
Pollution Prevention, https://www.epa.gov/sites/default/files/2021-
02/documents/nams list second update 2-4-21 final.pdf
U.S. EPA (U.S. Environmental Protection Agency). (2022a). ^Adverse Outcome Pathway Database (AOP-DB).
Retrieved 2022 from https://www.epa.gov/healtliresearcli/adverse-outcome-pathwav-database-aop-db
U.S. EPA (U.S. Environmental Protection Agency). (2022b). Chemical safety for sustainabilitv: Strategic research
action plan: Fiscal years 2023-2026 (EPA/600/R-22/238). Office of Research and Development.
https://www.epa.gov/svsteni/files/documents/2022-10/CSS%20FY23-26%20StRAP EPA-
ORD Qctober%202022 508.pdf
U.S. EPA (U.S. Environmental Protection Agency). (2022c). Cumulative impacts research: Recommendations for
EPA's Office of Research and Development (EPA/600/R-22/014a).
https://www.epa.gov/svstem/files/documents/2022-
09/Cumulative%201mpacts%20Research%20Final%20Report FINAL-EPA%20600-R-22-014a.pdf
U.S. EPA (U.S. Environmental Protection Agency). (2022d). ORD staff handbook for developing IRIS assessments
(EPA/600/R-22/268). Office of Research and Development, Center for Public Health and Enviromnental
Assessment, https://iris.epa.gov/Document/&deid=356370
U.S. EPA (U.S. Environmental Protection Agency). (2023a). Advances in dose addition for chemical mixtures: A
white paper (EPA/100/R-23/001). Risk Assessment Forum.
https://assessments.epa.gov/risk/document/&deid%3D359745
U.S. EPA (U.S. Environmental Protection Agency). (2023b). EPA legal tools to advance environmental justice:
Cumulative impacts addendum (Publication No.: 360R22002). Office of General Counsel.
https://www.epa.gov/svstem/files/documents/2022-12/bh508-
Cumulative%20Impacts%20Addendum%20Final%202022-l l-28.pdf
U.S. EPA (U.S. Environmental Protection Agency). (2023c). EPA policy on consultation with Indian tribes.
https://www.epa.gov/svsteni/files/documents/2023-12/epa-policv-on-consultation-with-indian-tribes-
2023.pdf
U.S. EPA (U.S. Environmental Protection Agency). (2024). Interim framework for advancing consideration of
cumulative impacts, https://www.epa.gov/svstem/files/documents/2024-ll/epa-interim-cumulative-
impacts-framework-november-2024.pdf
U.S. HHS (U.S. Department of Health and Human Services). (2020). Healthy People 2020: Social determinants of
health. Retrieved April 14, 2020 from
https://web.arcliive.org/web/202004141Q5159/littps://www.healthypeople. gov/2020/topics-
obiectives/topic/social-determinants-of-health
Vandenberg, L. N„ Rayasam, S. D. G., Axelrad, D. A., Bennett, D. H., Brown, P., Carignan, C. C., Chartres, N„
Diamond, M. L., Joglekar, R., Shamasunder, B„ Slirader-Frechette, K., Subra, W. A., Zarker, K., &
Woodruff, T. J. (2023). Addressing systemic problems with exposure assessments to protect the public's
health. Environmental Health, 2/(Suppl. 1), Article 121. https://doi.org/10.1186/sl2940-022-0Q917-0
Varshavsky, J. R., Rayasam, S. D. G., Sass, J. B„ Axelrad, D. A., Cranor, C. F., Hattis, D„ Hauser, R„ Koman, P.
D„ Marquez, E. C., Morello-Frosch, R„ Oksas, C., Patton, S., Robinson, J. F„ Sathyanarayana, S., Shepard,
P. M„ & Woodruff, T. J. (2023). Current practice and recommendations for advancing how human
variability and susceptibility are considered in chemical risk assessment. Environmental Health, 2/(Suppl.
1), Article 133. https://doi.org/10.1186/sl2940-022-00940-l
Vinikoor-Imler, L. C., Owens, E. O., Nichols, J. L., Ross, M„ Brown, J. S., & Sacks, J. D. (2014). Evaluating
potential response-modifying factors for associations between ozone and health outcomes: A weight-of-
evidence approach. Environmental Health Perspectives, 722(11), 1166-1176.
https://doi.org/10.1289/ehp.1307541
Weed, D. L. (2005). Weight of evidence: A review of concept and methods. Risk Analysis, 25(6), 1545-1557.
https://doi.org/10.1111/i. 1539-6924.2005.00699.X
Weiss, B., & Bellinger, D. C. (2006). Social ecology of children's vulnerability to environmental pollutants.
Environmental Health Perspectives, 114(10), 1479-1485. https://doi.org/10.1289/ehp.9101
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025 46
-------
WHO/IPCS (World Health Organization/International Programme on Chemical Safety). (2009a). Assessment of
combined exposures to multiple chemicals: Report of a WHO/IPCS international workshop on
aggregate/cumulative risk assessment. World Health Organization, https://iris.who.int/handle/10665/44113
WHO/IPCS (World Health Organization/International Programme on Chemical Safety). (2009b). Principles and
methods for the risk assessment of chemicals in food (Enviromnental Health Criteria 240). World Health
Organization.
https://apps.who.int/iris/bitstream/handle/10665/44065/WHQ EHC 240 eng.pdf?sequence=152
WHO/IPCS (World Health Organization/International Programme on Chemical Safety). (2009c). Principles for
modelling dose-response for the risk assessment of chemicals (Enviromnental Health Criteria 239). World
Health Organization. http://apps.who.int/iris/bitstream/10665/43940/l/9789241572392 eng.pdf
Williams, A. J., Lambert, J. C., Thayer, K„ & Dome, J. L. C. M. (2021). Sourcing data on chemical properties and
hazard data from the US-EPA CompTox Chemicals Dashboard: A practical guide for human risk
assessment. Environment International, 154, Article 106566. https://doi.org/10.1016/i.envint.2021.106566
Zartarian, V. G., & Schultz, B. D. (2010). The EPA's human exposure research program for assessing cumulative
risk in communities. Journal of Exposure Science and Environmental Epidemiology, 20(4), 351-358.
https://doi.org/10.1038/ies.2009.2Q
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025 47
-------
APPENDIX A. EVOLUTION OF CUMULATIVE RISK ASSESSMENT AT THE EPA
A.l. Background and History
The EPA Administrator's 1997 memo on Cumulative Risk Assessment Guidance - Phase I Planning and
Scoping first established cumulative risk assessment (CRA) policy at the EPA. Referencing the 1997
CRA Guidance, the Administrator directed "each office to take into account cumulative risk issues in
scoping and planning major risk assessments and to consider a broader scope that integrates multiple
sources, effects, pathways, stressors and populations for cumulative risk analyses in all cases for which
relevant data are available" (U.S. EPA, 1997a). The 1997 CRA Guidance states:
EPA's risk assessment emphasis has shifted increasingly to a more broadly based
approach characterized by greater consideration of multiple endpoints, sources,
pathways, and routes of exposure; community-based decision making; flexibility in
achieving goals; case-specific responses; a focus on all of the environmental media;
and significantly, holistic reduction of risk. This more complex assessment involves
cumulative risk assessment (U.S. EPA, 1997c).
A major impetus for the 1997 CRA Guidance was Executive Order 12898, Federal Actions to Address
Environmental Justice in Minority Populations and Low-Income Populations (EOP, 1994). Provisions of
the order included development of Agency implementation strategies and a requirement in the Research,
Data Collection, and Analysis section, that "Environmental human health analysis, whenever practicable
and appropriate, shall identify multiple and cumulative exposures." Subsequently, in the EPA's
Environmental Justice Strategy: Executive Order 12898, the Agency specified that as a component of
sound science for contributing to environmental justice, "EPA will evaluate the current state of
knowledge in exposure and cumulative risk fields, and then identify data gaps and research needs" (U.S.
EPA, 1995a).
The EPA's Science Policy Council and Risk Assessment Forum led the work on cumulative risk
identified in the EPA's Environmental Justice Strategy, integrating it with the Agency's ongoing
assessment of risk from chemical mixtures research. A milestone in this effort was the 2003 Framework
for Cumulative Risk Assessment, which helped advance CRA development by clarifying basic concepts
and objectives. The two main purposes of the 2003 Framework are (1) to offer a simple, flexible structure
for designing, conducting, and evaluating CRA at EPA; and (2) to construct basic principles around
which later guidelines might be organized (U.S. EPA, 2003b). An important contribution of the 2003
Framework was to define the terms "cumulative risk" and "cumulative risk assessment" in a way that is
generally accepted (Sexton, 2015).
The 2003 Framework specifies three discrete but overlapping assessment phases: (1) planning, scoping,
and problem formulation (assessment design); (2) analysis; and (3) risk characterization. The assessment
phases specified in the 2003 CRA Framework were also adopted by the Framework for Human Health
Risk Assessment (U.S. EPA, 2014b). These Guidelines draw on and are consistent with both earlier
publications. Planning and scoping define both the process for conducting the risk assessment and its
general scope, whereas problem formulation identifies major factors considered in a specific assessment
to inform its technical approach (U.S. EPA, 2014b).
The 2003 Framework states that the analysis phase produces risk results associated with multiple
chemical or nonchemical stressors to which study population(s) might be exposed (U.S. EPA, 2003b).
The risk characterization phase puts analysis findings and risk estimates into perspective in terms of
significance, reliability, and overall confidence. Risk characterization also evaluates whether the
assessment met the goals and objectives it initially set forth.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
A-l
-------
Distinguishing between a chemical- or stressor-focused assessment approach and a "receptor-focused"
assessment approach was central to the 2003 Framework. When the effects on the receptor are the
initiating factor for the assessment, the importance of the organism's (or population's) interactions with
stressors or agents (which could include factors that affect or mitigate the exposure or response to
stressors) is explicitly recognized. For example, as Weiss and Bellinger (2006) point out in the context of
developmental neurotoxicity, ".. .toxicity is not simply an inherent property of the toxicant but derives
from an assortment of jointly acting variables bound implacably into the individual...In a CRA context,
this could mean that "receptor" properties such as stressor vulnerability, background exposure, and
exposure history can be modifying factors. Evaluating the importance of such factors to the system being
assessed is a major challenge in CRA problem formulation.
The EPA has maintained an active CRA research and study program since the publication of the 2003
Framework that informed development of these Guidelines (U.S. EPA, 2003b). The 2007 CRA resource
document, Concepts, Methods and Data Sources for Cumulative Health Risk Assessment of Multiple
Chemicals, Exposures and Effects, is an important reference for CRA planning and problem formulation.
It describes and explains strategies such as the consideration of initiating factors for a CRA to organize
population characterization, data collection, and assessment planning and how to implement chemical
grouping to facilitate toxicity analysis (U.S. EPA, 2007a). Grouping stressors to simplify analysis is
recommended as a critical component of the CRA planning process to design a tractable study and to
target resources properly.
Building on the method for grouping stressors, the EPA's Pesticide Cumulative Risk Assessment:
Framework for Screening Analysis (U.S. EPA, 2016b) provides guidance on how to screen groups of
pesticides for cumulative evaluation during Registration Reviews.46 A screening analysis is used to
determine whether the available toxicological data for a group of chemicals suggest that a common mode
of action47 can be established. If so, a screening-level toxicological and exposure analysis is used to
provide an initial determination of whether further assessment is necessary. The Pesticide CRA
Framework implements a tiering strategy to identify risks effectively and to use resources efficiently.
Tiering a CRA, discussed further in Section 2.5, is an important planning method for targeting priority
stressors and matching appropriate assessment efforts to the risk management decision.
The EPA supported a series of grants exploring methods to address nonchemical stressors in CRA
(Payne-Sturges et al., 2015) and commissioned a series of issue papers that investigated a range of CRA
themes, including combined health effects from multiple stressors, incorporating vulnerability into CRA,
assessing environmental mixtures, and using biomarkers to inform CRA (Callahan & Sexton, 2007; deFur
et al., 2007; Menzie et al., 2007; Ryan et al., 2007; Sexton & Hattis, 2007). Those sources provided
important background to support the development of these Guidelines.
Gallagher et al. (2015) examined CRAs in case studies and issue papers (including those cited above)
written by or commissioned for the EPA between 2000 and 2007 and concluded the following:
• A tiered approach can be valuable to CRA because it focuses resources on the most important
factors in a CRA.
• The spatial scale of a CRA influences data needs, model choice, and the utility of the information
for comparing risk management options.
• Consideration of population vulnerabilities is necessary.
• Early and regular communication with stakeholders is important.
• The use of iterative processes (e.g., continued reexamination of assessment goals, verifying that
40 The Pesticide CRA Framework is an extension of two prior documents: Guidance for Identifying Pesticide Chemicals and
Other Substances That have a Common Mechanism of Toxicity' (U.S. EPA, 1999) and Guidance on Cumulative Risk Assessment
of Pesticide Chemicals That Have a Common Mechanism of Toxicity' (U.S. EPA, 2002a).
47 A common mode of action is a CRA requirement of the Food Quality Protection Act but not for other EPA programs.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
A-2
-------
the necessary data and methods are available) is beneficial to the CRA process.
The EPA presented a CRA series of 15 public webinars between August 2012 and December 2013,
featuring EPA grant awardees and other scientists studying CRA methods. A list of topics and recordings
of the webinars is available in the EPA's archive (U.S. EPA, 2020b). The summary of conclusions from
the webinar series (U.S. EPA, 2014a) provides useful perspective when planning a CRA. These
conclusions include:
• Complexity of Vulnerability and Nonchemical Stressors. Extensive studies show associations
between disadvantaged communities and suboptimal health. Examination of how to include social
stressors in CRA has led to epidemiological studies receiving greater emphasis in CRA as an
approach to assessing the relative contribution of different stressors and potential interactions
between chemical and social stressors.
• Epidemiological Methods. Effect Modifiers, and Dose-Response Curves. Evidence continues to
emerge that social conditions can amplify the effect of environmental agents on health and can
contribute to health disparities. These social conditions can be incorporated quantitatively in CRA
as effect modifiers in the dose-response assessment if data exist to support the relationship.
• Statistical Models. CRA at community levels is challenged by the limited availability of relevant
exposure and health data at the appropriate geographic scale. Statistical models can be useful
tools for evaluating cumulative exposures and risks. An example is a regression model based on
National Health and Nutrition Examination Survey (NHANES) biomonitoring data that could
predict the effect of exposures on common health endpoints in communities.
• Mapping and Screening for Cumulative Burden (Indices). Geographic information system
mapping of multiple pollution sources overlaid with demographics information that incorporates
indicators of population vulnerability and nonchemical stressors can help identify vulnerable
populations that also have high environmental hazard burdens. Visualization of areas of overlap
can be useful in planning a CRA. Mapping approaches for cumulative burden have been used as a
proxy for cumulative impacts, identifying overburdened populations and helping to prioritize
areas for further analysis.
• Legal/Decision Frameworks. Risk assessors have expressed concern that CRA requires
considering every possible stressor in the assessment. Concerns about the limitations of statutory
authority for the use of CRA in decision-making also exist. Statutes may impose a general
provision on the Agency to "protect human health and the environment," which could support the
use of cumulative risk in a decision. For this, the EPA should be able to demonstrate that (1) its
CRA approach and its use of the results are scientifically reasonable (appropriate use of data and
assumptions) and (2) the Agency considered all factors the statute requires.
• Differing Meanings of CRA/Use of Terms (Impacts. Risks. Levels. Effects'). Participants in the
discussion observed many and divergent understandings of "cumulative." The insight gained is
that a cumulative assessment need not necessarily provide a bright line for risk or a single
number. "Cumulative" can be described in multiple ways—as impacts, levels, risks, and effects,
both quantitatively and qualitatively, and on different spatial and temporal scales—and all are
useful for informing different decisions.
The National Research Council (NRC) helped shape risk assessment policy at the EPA, beginning with
the seminal 1983 publication, Risk Assessment in the Federal Government: Managing the Process (NRC,
1983), widely known as the "Red Book" in the risk assessment community, that led to formalization of
the EPA's use of risk assessment to underpin risk management supporting regulations. Before the Red
Book, NRC took on the challenge of characterizing Principles of Toxicological Interactions Associated
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
A-3
-------
with Multiple Chemical Exposures in 1980 for the U.S. Coast Guard, where there was concern for marine
inspectors exposed to multiple different chemicals during their workdays. This NRC report laid out the
basis for CRA with an examination of "the mechanisms of toxicological interactions in terms of the
toxicokinetic and toxicodynamic factors that are involved" (NRC, 1980). The study of "sites and
mechanisms at which and through which toxicological interactions can occur" (NRC, 1980) dominated
NRC's inquiries into risk assessment through its publication of Toxicity Testing in the 21st Century (NRC,
2007). In 2008, NRC examined a specific class of chemicals in Phthalates and Cumulative Risk
Assessment: The Tasks Ahead (NRC, 2008a). NRC examined the utility of a CRA for phthalates and
made recommendations for how a CRA could be undertaken. In 2009, with publication of Science and
Decisions, NRC enhanced the visibility of public health perspectives, following the observation that:
.. .there is increasing concern among stakeholder groups... that such a narrow
(individual chemical) focus does not accurately capture the risks associated with
exposure, given simultaneous exposure to multiple chemical and nonchemical
stressors and other factors that could influence vulnerability.
NRC further inquired about the limitations of mechanistic toxicology and concluded that "Without
additional modifications, risk assessment might become irrelevant in many decision contexts, and its
application might exacerbate the credibility and communication gaps between risk assessors and
stakeholders" (NRC, 2009). NRC then noted key differences between the CRA paradigm and traditional
human health risk assessments:
• CRA is not necessarily quantitative.
• CRA extends beyond chemicals to include psychosocial, physical, and other factors.
In Sustainability and the U.S. EPA, NRC acknowledged that the limitations of risk assessment to
adequately approach complex issues such as characterizing sustainability:
... has led to approaches to widen the risk paradigm, to include the context in which
the analysis is performed, the early consideration of a broad range of decision options,
and the cumulative threats of multiple social, environmental, and economic stressors
to health and the environment (NRC, 201 lb).
NRC has asserted that standard risk-based regulatory approaches have limitations. Limitations include
difficulty in dealing with complex problems that are not readily addressed by analyses that seek to
"simplify the multidimensionality of the risk or make sense of the uncertainty" (NRC, 1996) or that
require a volume of information and analyses that far outstrip the resources available to provide them
(NRC, 2006). NRC has stated that CRA provides the EPA with a "broader and more comprehensive
understanding of the complex interactions between chemicals, humans, and the environment" (NRC,
2012). NRC has asserted throughout numerous publications addressing risk management and
environmental decision-making that CRA should be implemented in a way that is responsive to
community needs and extends beyond chemical toxicology to incorporate psychosocial, physical, and
other factors. NRC's observation is consistent with the assertions of environmental justice advocates who
encourage the EPA to recognize the need for more comprehensive assessments in communities burdened
by elevated levels of multiple environmental stressors from multiple sources, where populations could be
at risk because they are highly exposed to numerous environmental as well as psychosocial stressors.
These populations might be more vulnerable to such stressors due to other socioeconomic conditions
(NEJAC, 2004).48
48 These communities are sometimes called "environmental justice communities." Environmental justice advocates have focused
attention on the need for increased understanding of population vulnerabilities and increased stakeholder involvement in the risk
assessment process.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
A-4
-------
NRC's evolving perspective on risk assessment mirrors the EPA's progress in development of the CRA
Guidelines. A conundrum is created when the need to keep risk assessment relevant to public and
environmental health realities requires a broader inquiry that includes qualitative factors not easily
incorporated into the risk assessment paradigm. This challenge is resolved in part through characterizing
the different types of information needed to meet various purposes served by different types of
assessments. The extent and scope of quantitative analyses in an assessment will vary depending on the
objectives of a management decision. For example, establishing pollutant release limits generally requires
quantitative analysis but may benefit from information that is not easily quantifiable or traditionally part
of the risk management paradigm. The 2003 Framework for CRA recognized the utility of
nonquantitative information; accordingly, these Guidelines provide for the consideration of such
information within the constraints of the risk management decision.
A.2. Examples of CRA at the EPA
EPA programs have pioneered CRA applications. Past use of CRA at the EPA offers examples of when a
CRA can be useful.49 The Office of Water/Office of Ground Water and Drinking Water in conjunction
with the Office of Research and Development undertook the cumulative assessment of drinking water
disinfection by-products (DBPs) that identified potential adverse outcomes (U.S. EPA, 2000a). DBPs are
classes of chemicals that can form during the treatment of drinking water and could present human health
risks for developmental and reproductive effects and cancer. The assessment supported development of
the Stage 1 and Stage 2 rules regulating disinfectant/disinfection by-products using an approach to CRA
in which DBPs with a common response were summed to establish acceptable exposure levels. The
approach developed for conducting such assessments was to construct physiologically based
pharmacokinetic (PBPK) models to estimate internal dose from exposure to two classes of DBPs
(Simmons et al., 2004; Teuschler et al., 2004).
The EPA's Office of Land and Emergency Management and regional risk assessors routinely conduct
site-specific human and ecological risk assessments under the Comprehensive Environmental Response,
Compensation, and Liability Act (CERCLA), also known as ""Superfund." as amended by the Superfund
Amendments and Reauthorization Act (SARA), the Resource Conservation and Recovery Act, and other
statutory authorities. The site-specific risk assessments generally consider elements of cumulative risk
when evaluating exposure, calculating risk, and performing risk characterization. Exposure assessments
typically quantify all possible routes reasonably expected to occur at a site. The Superfund risk
assessments are based on an estimate of the reasonable maximum exposure (RME) expected to occur
under both current and future land-use conditions consistent with the EPA's Risk Assessment Guidance
for Superfund (RAGS), published in several specific guidance documents.5" The RME is defined as the
highest exposure that is reasonably expected to occur at a site. RMEs are estimated for each exposure
pathway. If a population is exposed via more than one pathway, the combination of exposures across
pathways also should represent an RME. The exposure assessment considers all RME individuals who
may be exposed now or in the future, including residents, workers, trespassers, and others depending on
the nature of contamination at the Superfund site and potential exposures. RME assessments for the RME
individual use a combination of upper-bound and central tendency estimates of exposure parameter values
for both an adult and a child. The combination of parameter estimates is designed to be protective of the
RME individuals exposed at the site. For screening and related efforts, the most sensitive receptor
typically is chosen, and cultural and lifestyle-specific exposures are considered when indicated. The dose-
response relationship(s) of chemicals present in mixtures might be evaluated using hazard index or
relative potency methods and cancer risks from exposures to site-related multiple chemicals and multiple
49 EPA Legal Tools to Advance Environmental Justice: Cumulative Impacts Addendum (U.S. EPA, 2023b) is a compilation of
legal authorities available to the EPA for identifying and addressing cumulative impacts and cumulative risks on communities
with environmental justice concerns and other underserved populations.
50 See the EPA's Risk Assessment Guidance for Superfund (RAGS): Part A web page (https://www.epa.gov/risk/risk-assessment-
guidance- superfund-rag s-part).
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
A-5
-------
pathways. OLEM generally does not consider the impact of nonchemical stressors in human health risk
assessment (radiological risks are an exception); ecological risk assessments, however, do occasionally
consider nonchemical stressors (e.g., habitat impairment).
Within the EPA's Office of Chemical Safety and Pollution Prevention, the Office of Pesticide Programs
(OPP) has conducted CRAs for pesticides with a common mechanism of toxicity, as stipulated by the
Food Quality Protection Act of 1996 (FQPA). The first pesticide group for which OPP conducted a CRA
was the organophosphate (OP) pesticides group. Variations in toxicity among OP pesticides were
considered by assigning each pesticide in the group a relative potency factor, relative to an index OP.
Using a similar methodology, OPP also completed CRAs fortriazines, chloroacetanilides, pyrethroids,
and TV-methyl carbamates using established OPP cumulative guidance. These guidance documents and
OPP's experience with conduct of CRAs on pesticide groups were used to develop Pesticide Cumulative
Risk Assessment: Framework for Screening Analysis (U.S. EPA, 2016b), discussed above.
The EPA's Office of Air and Radiation (OAR) uses CRA to support decisions in the regulation of
hazardous air pollutants (HAPs) and in reviews of some National Ambient Air Quality Standards
(NAAQS). For example, in the HAP program, cumulative assessments of cancer risk and noncancer
hazard associated with exposures to multiple HAPs emitted from stationary emission sources inform
regulatory decisions on source category-specific emission standards for HAPs (also called air toxics).
Cumulative cancer risk is estimated as the sum of all individual HAP cancer risk estimates. For
assessment of cumulative noncancer hazard, target organ-specific hazard indices are estimated. OAR's
nonregulatory Annual Air Toxics Update, performed to inform programmatic priorities, does the same for
multiple stationary and mobile sources of HAPs. Additionally, CRAs have been performed to inform
decisions on some of the individual NAAQS. The NAAQS, as single pollutant-based standards for
ambient air, reflect consideration of the cumulative concentrations of various precursor or constituent
chemicals in ambient air, which result from emissions from many sources. In the case of risk assessments
for fine particulate matter, the assessment is of the whole mixture of fine particulate matter and reflects
cumulative health risk associated with all particulate substances in ambient air that fall into the particle
size class of interest. Cumulative ecological risk assessment has also been performed to inform NAAQS
decisions (e.g., in assessing ecological risk associated with the co-occurrence in ambient air of multiple
oxides of sulfur and nitrogen).
A.3. Looking Ahead toward CRA Advances
Advances in CRA methodologies are occurring primarily in two areas: (1) adverse outcome pathways and
common mode of action and (2) the combined effects of chemical and nonchemical stressors (Fox et al.,
2017).
The first area of advancement is in assessing the effects of chemical mixtures that share common modes
of action or that cause common adverse outcomes. In this area, two primary models are used for
predicting effects from exposure to mixtures of chemicals: dose addition and response addition (NRC,
1993, 2008a; U.S. EPA, 1986b, 2000d, 2002a, 2003a, 2023a). The EPA updated and provided detailed
models for examining additivity of chemical mixtures in Advances in Dose Addition for Chemical
Mixtures: A White Paper (U.S. EPA, 2023a). The second area of advancement in CRA methods is in
evaluating the combined effects of chemical and nonchemical stressors (NRC, 2009; U.S. EPA, 1997c;
2003b). Methods for evaluating the combined effects of chemical and nonchemical stressors have been
proposed, and approaches are being developed (Barrett & Padula, 2019; Hibbert & Tulve, 2019; Payne-
Sturges et al., 2018; Varshavsky et al., 2023). A particular challenge is how to incorporate mechanistic
and mode of action information for the combination of chemical and nonchemical stressors that impact
communities and associated uncertainties, as well as the evaluation of differential vulnerability. The EPA
stated it will "Increase understanding of the factors that influence environmental health disparities, and
develop methods and data to assess adverse and cumulative risks" to strengthen the scientific foundation
for considering the factors important to incorporating environmental justice in decision-making (U.S.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
A-6
-------
EPA, 2016a). As scientific innovation advances with new analytical methods, a corresponding growth
occurs in the types and complexity of cumulative assessment problems that can be analyzed. The use of
new approach methods,51 such as high-throughput screening systems and global gene-expression analysis,
in assessing cumulative risk brings substantial new methodological capacity to CRA and broadens its
potential application. Using 21st Century Science to Improve Risk-Related Evaluations (NASEM, 2017)
described how these methods are developing rapidly. The EPA's development and implementation of
such methods are in part described in the Chemical Safety for Sustainability Strategic Research Action
Plan (U.S. EPA, 2022b) and List of Alternative Test Methods and Strategies (or New Approach
Methodologies) (U.S. EPA, 2021b). These Guidelines anticipate the evolution of CRA analysis methods
to include an ever wider range of possible stressors and exposure-response modifiers.
51 In a risk assessment context, new approach methods (NAMs) are defined as any technology, methodology, approach, or
combination thereof that can provide information useful for risk assessment (including hazard assessment, dose-response
assessment, and exposure assessment) without the use of traditional test animals (e.g., rats, mice), including in silico, in chemico,
in vitro, and ex vivo approaches.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025 A-7
-------
APPENDIX B. CUMULATIVE RISK ASSESSMENT FOUNDATIONAL DOCUMENTS
Table B-l provides a list of EPA publications that are foundational to the evolution of the field of
cumulative risk assessment (CRA) planning and problem formulation in chronological order and a brief
description of their relevance. Table B-2 serves the same function but highlights foundational publications
from the National Research Council (NRC) and other non-EPA sources.
Table B-l. EPA Publications Relevant to Cumulative Risk Assessment Planning and Problem Formulation"
EPA Publication
Relevance to CRA Planning and Problem Formulation
Guidelines for the Health Risk
Assessment of Chemical Mixtures
(U.S. EPA, 1986b)
Indicates scoping inputs from internal and external inputs, describes problem (significant
data gaps), terminology, and conceptual approach for addressing gaps to support decision
needs.
Supplementary Guidance for
Conducting Health Risk Assessment
of Chemical Mixtures (U.S. EPA,
2000d)
Builds on previous context, identifying data requirements: mixture of concern, similar
mixture, components.
Conducting a Risk Assessment of
Mixtures of Disinfection By-
Products (DBPs) for Drinking
Water Treatment Systems (U.S.
EPA, 2000a)
Includes scoping and problem formulation by scientific experts, including assessing the
state of toxicity and exposure data, conducting an initial example assessment to guide the
more detailed follow-on, and gaining inputs and insights via a targeted workshop.
Risk Assessment for Supetfund Sites
(R4GS) (U.S. EPA, 1989)
Oriented to CERCLA requirements, focusing on environmental contaminants (notably
chemicals and radionuclides); reflects the National Contingency Plan, which includes
distinct stakeholder involvement requirements; describes project scoping process and
remedial investigation/feasibility study goal (of which baseline health risk assessment is a
part), project planning, conceptual model, and data quality; outlines a tiered approach from
initial screening to more detailed assessment. Superfund Guidance is available at:
httDs://www.eDa.aov/risk/risk-assessment-auidance-suDerfund-raas-part.
Guidelines for Ecological Risk
Assessment (U.S. EPA, 1998a)
First EPA report to articulate problem formulation—what is at risk, what needs to be
protected—as one of three main phases, the other two being analysis (exposure and
toxicity) and risk characterization; emphasizes conceptual model and iterating throughout
the process; outlines stressor identification, including ways to think about nonchemical
stressors.
Guidance on Cumulative Risk
Assessment. Part 1. Planning and
Scoping (U.S. EPA, 1997a, 1997c)
Encourages early and continued planning and scoping; describes process for engaging
stakeholders and others, including experts; urges identifying these at the outset: overall
purpose and risk management objectives, scope, participants and roles, resources,
conceptual model, analysis plan.
Lessons Learned on Planning and
Scoping for Environmental Risk
Assessments (U.S. EPA, 2002b)
Uses examples to reinforce importance of formal planning and dialogue from outset;
conceptual models help outline stressor-effect relationships, program-regulatory activities;
encourages considering same sources.
General Principles for Performing
Aggregate Exposure and Risk
Assessments (U.S. EPA, 2001a)
Oriented to FQPA requirements; identifies scope as overall framework and principles
focusing on exposure to a single chemical by multiple pathways and routes; places
emphasis on quality evaluations of available data from multiple sources.
Guidance on Cumulative Risk
Assessment of Pesticide Chemicals
That Have a Common Mechanism
of Toxicity (U.S. EPA, 2002a)
Oriented to FQPA requirements, pesticides with a common mechanism of toxicity;
emphasizes importance of determining need for/capability to perform a CRA at outset and
identifying CRA objectives; notes that not all CRAs need be of the same depth and scope;
considers both screening-level and refined or detailed assessments.
Frameworkfor Cumulative Risk
Assessment (U.S. EPA, 2003b)
Flexible structure to anchor program-specific guidance; identifies CRA components, terms,
technical and coordination issues, with planning and scoping as a key early step to gain
consensus on CRA goals; includes principles for stakeholder involvement.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
B-l
-------
EPA Publication
Relevance to CRA Planning and Problem Formulation
A Framework for Assessing Health
Risk of Environmental Exposures to
Children (U.S. EPA, 2006c)
Examines the impact of potential exposure to stressors during developmental lifestages and
subsequent lifestages while emphasizing the iterative nature of the analysis phase with a
multidisciplinary team; outlines the framework in which mode of action(s) (MOAs) can be
considered across lifestages; is based on existing approaches adopted in the Framework on
Cumulative Risk Assessment and identifies existing guidance, guidelines, and policy
papers that relate to children's health risk assessment emphasizes the importance of an
iterative approach between hazard, dose response, and exposure analyses; includes
discussion of principles for weight-of-evidence consideration across lifestages for the
hazard characterization database.
Concepts, Methods, and Data
Sources for Cumulative Health Risk
Assessment of Multiple Chemicals,
Exposures and Effects (U.S. EPA,
2007a)
Emphasizes early planning and scoping, decision-focused stakeholder involvement;
problem formulation and conceptual model; introduces initiating factors to determine
whether CRA is appropriate, what key data are needed; addresses data quality.
Framework for Human Health Risk
Assessment to Inform Decision
Making (U.S. EPA, 2014b)
Clarifies the many decisions needed in this step, the risk management focus, and
stakeholder involvement; aggregates exposure issues and some toxicity issues such as
influence of population characteristics.
Pesticide Cumulative Risk
Assessment: Framework for
Screening Analysis (U.S. EPA,
2016b)
Provides guidance on how to screen groups of pesticides for cumulative evaluation using a
two-step approach, beginning with the evaluation of available toxicological information
and, if necessary, followed by a risk-based screening approach; an example of tiering, the
incremental approach to determining which level of effort is required to assess risks cost
effectively.
'Acronyms: CERCLA = Comprehensive Environmental Response, Compensation, and Liability Act, as amended; FQPA = Food Quality
Protection Act of 1996.
Guidelines for the Health Risk Assessment of Chemical Mixtures (U.S. EPA, 1986b, 2000d)
The 1986 mixture risk guidelines are notable for their brevity (25 pages), as the detailed procedures were
intentionally left to the EPA's program offices to address legislation-specific guidance. (Note that the
Superfund guidance summarized in the document's Section 1.4.1.4 was the earliest and most extensive
program guidance that addressed mixtures). The 1986 mixture guidelines included some general concepts
of dose and response addition, and they recommended the dose-additive hazard index (HI) for assessing
the potential for noncancer effects and response addition for estimating cancer risk. These guidelines also
identified cautions about potential impacts of missing information, including toxicological interactions.
The 2000 supplement greatly expands on those concepts and methods, primarily regarding dose-response
assessment, introducing the interaction-based HI as a way to incorporate known toxicological interactions
quantitatively. The most important contribution of both documents to CRA is the encouragement to
account for toxicological interactions among chemicals (including a default five-fold interaction
magnitude when specific data are absent), to use qualitative or subjective analyses when necessary (e.g.,
the weight-of-evidence scheme for interactions, the quality scores for data on exposure, toxicity, and
interactions) and to prefer the use of scenario-specific information (e.g., exposure and toxicity
information on the mixture of concern instead of its component chemicals).
The 2000 supplement extensively discussed concepts and terminology, including several applications of
dose addition based on the available information, including toxicity equivalence factors, relative potency
factors, and HI. These documents led to the adoption of dose addition as the primary default approach for
mixtures of chemicals causing similar effects, promoting Agency-wide consistency of the approach's use
when interaction data are absent. Another program-specific application published that year is the report
on drinking water disinfection by-product (DBP) mixtures (U.S. EPA, 2000a), which applied and
extended the potency factor approach to a cumulative relative potency factor to integrate data on mode of
action and multiroute exposures over physiologically relevant time frames.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
B-2
-------
Risk Assessment Guidance for Superfund (RAGS) (U.S. EPA, 1989)
The EPA's Superfund office developed a set of guidance documents for human and ecological risk
assessments at waste sites on the National Priorities List to address the mandate of the
1980 Comprehensive, Environmental Response, Compensation, and Liability Act (CERCLA), as
amended by the 1986 Superfund legislation. This suite of documents has been adopted for many other
applications, including operational facilities. Tiering is a key concept of RAGS, with analyses ranging
from screening level to detailed, depending on the operable units addressed and needs of a given decision.
The initial guidance focused on human health, and subsequent documents have added concepts and
approaches to the portfolio of available methods (e.g., ecological, dermal exposures, toxicity value
selection, inhalation).
The conceptual model is a central element of the RAGS exposure assessment, with direct and indirect
exposures considered across multiple pathways involving multiple chemicals (including radionuclides).
Environmental fate, such as partitioning and transformation over time, can be very important, depending
on the nature of the contaminants (e.g., volatile compounds versus metals). Per supplemental guidance
(U.S. EPA, 2002a), the concentration term typically reflects the 95%upper confidence limit of the
arithmetic average based on existing measurements, with modeling conducted to predict future
concentrations as indicated by the scenarios being assessed.
The RAGS approach for developing preliminary remediation goals (U.S. EPA, 1991b) considers multiple
chemicals in multiple media and generally uses default assumptions to assess several standard exposure
scenarios (including hypothetical residential use). A subsequent RAGS document (Part E), released in
2004 (U.S. EPA, 2004b), addressed dermal exposures to contaminants in soil, sediment, and water. This
document addresses a relatively limited set of organic compounds (those with dermal absorption fractions
exceeding a threshold level), considers the cumulative amount available over a time profile to support
permeability coefficients, and calls for using gastrointestinal absorption data to adjust for systemic
toxicity when dermal data are unavailable. The original RAGS approach for assessing joint toxicity of
multiple chemicals reflected the method outlined in the 1986 mixtures guidelines, with the HI approach
used to assess the potential for noncancer effects and response addition applied for cancer risk. This
approach was subsequently updated electronically to incorporate the mixtures supplement (U.S. EPA,
2000d). Pathway-specific risks and/or His are summed to produce combined risk and combined HI
estimates for each scenario. If the combined HI exceeds the target level of one, it is commonly segregated
by endpointto address organ- or system-specific differences. Risk assessments conducted for these
cleanup sites (and other applications that follow RAGS) typically use reference values developed by the
EPA in accordance with several toxicity guidelines, notably for mutagenic (U.S. EPA, 1986a),
developmental (U.S. EPA, 1991a), reproductive (U.S. EPA, 1996b), and neurotoxic endpoints (U.S. EPA,
1998b), as well as for cancer (U.S. EPA, 2005).
Other RAGS documents have further expanded the scope of tools for assessing disparate risk, including
the 2001 guidance for probabilistic risk assessment that focuses on Monte Carlo approaches to quantify
variability and uncertainty in risk estimates (although not specific to CRAs) and the supplemental
guidance (U.S. EPA, 2001c) for assessing health risk from inhalation exposures. The latter explicitly
discusses estimating aggregate and cumulative risk for multiple chemicals and exposure routes, noting the
importance of indicating whether the assumption of independent action underlying the default additive
approach is valid (i.e., "no synergistic or antagonistic interactions and all chemicals produce the same
effect, i.e., cancer"). Further, this document introduces an approach for considering different exposure
periods—acute, subchronic, and chronic—indicating that in general (1) hazard estimates from multiple
acute events should be summed only when the exposures occur simultaneously; (2) when a series of acute
events occur, the highest single exposure concentration should be compared with the relevant reference
value; (3) distinct His should be calculated for each exposure period; and (4) if the HI exceeds the
threshold level, the assessor should segregate it by target organ. These guidelines also explicitly call for
placing susceptibilities in context within the risk characterization section; receptor characteristics such as
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
B-3
-------
age, disease, sex, and genetic characteristics are included, and two examples are discussed (children and
workers).
Standard default exposure assumptions are provided on the following web page:
https://www.epa.gov/risk/update-standard-default-exposure-factors.
In addition, Superfund risk assessment guidance is available at: https://www.epa.gov/risk/risk-
assessment-guidance-superfund-rags-part.
Guidelines for Ecological Risk Assessment (U.S. EPA, 1998a)
The 1998 EPA ecological risk guidelines and two related documents—the 1997 ecological risk guidance
for Superfund (U.S. EPA, 1997b) and the stressor identification guidance from 2000 (U.S. EPA, 2000c)
—provide much of the basis for CRA by addressing combinations of dissimilar stressors over varying
time frames and with multiple measurable endpoints for adverse impacts. The guidelines also articulated a
three-phased structure for risk assessment: problem formulation, analysis, and risk characterization
(shown in Figure 1-1 of the 1998 EPA ecological risk guidelines), and they highlight the importance of
involvement by risk managers in the risk assessment process. To accommodate data of widely differing
quality and type, the ecological risk methods allow for both qualitative and quantitative analyses and
characterizations. Problem formulation in particular is emphasized, as is the need for iterating throughout
the assessment process. "Successful completion of problem formulation depends on the quality of three
products: assessment endpoints, conceptual models, and an analysis plan. Because problem formulation is
an interactive, nonlinear process, substantial reevaluation is expected to occur during the development of
all problem formulation products../' (U.S. EPA, 1998b). With details provided for construction of
conceptual models and specific guidance on stressor identification, these three documents address many
of the difficult areas of cumulative risk analysis and provide a valuable foundation upon which CRA
methods can be developed for human health risk as well as for joint impacts on health and ecosystems.
CRA Planning and Scoping (U.S. EPA, 1997c, 2002b)
Two documents, the 1997 Guidance on Cumulative Risk Assessment, Part 1, Planning and Scoping and
the 2002 Lessons Learned on Planning and Scoping for Environmental Risk Assessments, focus on the
initial steps in the CRA process. The first brief guidance is designed to raise awareness and promote
dialogue on CRAs, and it describes a process for engaging assessors, managers, and stakeholders. The
intent is to help assessors and managers plan and document the scope of a given CRA and consider
appropriate participants and their roles (technical, advisory, or stakeholder, as determined by the risk
manager), as well as information sources. It considers multiple stressors or agents, multiple sources,
media, pathways, routes of exposure, and multiple endpoints. Emphasis is placed on community-based
decision-making, flexibility in achieving goals, case-specific responses or approaches, and holistic
reduction of risk. While acknowledging data limitations, this initial guidance focused on integrating
human health and ecological effects from synthetic chemicals, radiation, and biological stressors in the
environment, with a longer-term plan to address further issues.
Additional concerns related to social, economic, behavioral, or psychological stressors are mentioned, but
because relevant data on psychosocial influences were limited, the focus is on the other stressors.
Vulnerable populations are emphasized, notably children and sex-related differences in both vulnerability
and exposure. A key theme is involving stakeholders and others (including economists and other social
scientists) in planning and scoping and throughout the risk assessment and risk management process (e.g.,
to provide input to the decision-maker(s)). The aim is for technical experts and management to work as a
team, informed by stakeholder input, beginning with early planning and scoping. Such a coordinated
process would increase the likelihood of the analysis addressing the risk management needs and the risk
managers understanding the strengths and limitations of the analysis.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
B-4
-------
The integration of environmental risks includes dialogue on the given assessment, definition of the main
CRA term(s), products needed to make a risk decision, tentative budget and schedule, and planned
approach. Implementation tasks include:
• Define purpose and management objectives
• Determine scope, problem statement, participants, and resources for the CRA
• Identify questions to answer, technical approach, conceptual model, and plan
• Outline six risk dimensions: sources (point, nonpoint, natural), stressors, pathways and exposure
routes, population(s) (human, ecological, landscape/geography), endpoints (human, ecological),
time frames (acute, chronic, subchronic, intermittent)
The second report supplements the 1997 Guidance on Cumulative Risk Assessment, Part 1 document,
using case studies to illustrate organizing principles for identifying participants, bounding the problem,
developing a conceptual model, and planning the analysis for a CRA. It aims to reinforce the importance
of formal planning and dialogue prior to conducting complex cumulative assessments and to provide case
study "lessons learned" for those planning an assessment. Case studies included pesticide registration,
water permit conditions for a concentrated animal feeding operation, a citizen petition-based cumulative
risk initiative under the Toxic Substances Control Act, EPA's National Air Toxics Assessment for
national screening of hazardous organics in urban air, and a screening-level surface impoundment study
for cumulative risk from hazardous constituents in wastewater treatment ponds. This report shows that
(1) early stakeholder input can help focus the analysis and improve confidence in the decision-making
process, and (2) planning and scoping help ensure the CRA better informs a risk manager's decisions. The
other findings emphasized communication, notably:
• Involving the decision-maker (e.g., risk manager) throughout the process helps ensure that the
products will meet the decision needs.
• Involving stakeholders at the beginning helps identify public health endpoints to study.
• Providing a clear set of definitions of key terms facilitates consensus.
• Obtaining clear, objective resource commitments and estimated schedules from risk managers
contributes to identifying the scope and detail of an assessment.
• Having conceptual models helps identify relationships between stressors-effects and programs-
regulatory activities.
• Explaining the uncertainties is important to build trust, credibility, and support for the decision-
making process.
Multi-pathway Combustor Emissions (U.S. EPA, 1998c)
This methodology document updates previous reports on indirect exposures to chemicals in incinerator
emissions. It is intended to apply to aggregate (direct and indirect) exposures "resulting from atmospheric
pollutants that are emitted from a stationary combustion source, transferred through the atmosphere, and
deposited downwind to environmental media and biota." Because direct exposure data for all pathways
are rarely available, the document discusses the use of data and models to estimate uptake and transfer of
atmospheric agents through the terrestrial or aquatic food chains. Aggregate exposures could occur
through many routes, including inhalation; ingestion (e.g., of tap water, produce, livestock, fish, and
breast milk); and dermal contact with soil and water contaminated by aerial deposition. Most of this
report presents mathematical models for estimating average daily route-specific exposures. Combinations
across pathways and routes are estimated by summing the absorbed daily doses or by summing the oral
equivalent daily doses. Cross-route conversion is identified as an uncertainty and recommended only for
dermal-to-oral exposures and only under certain conditions.
This document is one of the first EPA methodology reports to incorporate different factors for children
versus adults to estimate daily intake. The final chapter (Risk Assessment) presents several advances
relevant to cumulative risk. One is the recommendation to align the dose-response data with the exposure
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
B-5
-------
duration being addressed, including using mixture dose-response data when appropriate. A second is to
recognize the likelihood of multiple sources, not just incinerators, and multiple chemicals, so that
"cumulative exposure" is defined as the total exposure (single chemical or multiple stressors, including
nonchemical pollutants) resulting from all sources through multiple routes over the interval of interest.
Finally, this report emphasizes the additional uncertainties associated with multi-pathway exposures,
including the possible interactions in exposure (one chemical affects the fate and transport of another) and
uptake (two chemicals compete for the same metabolic pathway or toxicodynamic process). Considerable
attention is paid to various methods of characterizing variability and uncertainty. Because this report is
not an Agency-wide or program-specific guidance per se, it has had limited application beyond combustor
emissions. Nevertheless, its impact on cumulative risk methods has been significant.
Cumulative Risk Assessment of Pesticide Chemicals (U.S. EPA, 2001a, 2002a)
In response to the Food Quality Protection Act (FQPA) of 1996, the Office of Pesticide Programs (OPP)
began conducting CRAs to evaluate potential human health risks arising from combined exposures to
pesticides in classes that act through a common mechanism of toxicity. The primary methods are
presented in OPP's Guidance for Identifying Pesticide Chemicals and Other Substances That Have a
Common Mechanism of Toxicity (U.S. EPA, 1999), General Principles for Performing Aggregate
Exposure and Risk Assessments (U.S. EPA, 2001a), and Guidance on Cumulative Risk Assessment of
Pesticide Chemicals That Have a Common Mechanism of Toxicity (U.S. EPA, 2002a). The major
contribution to cumulative risk methods concerns the extreme detail and care in addressing multi-pathway
exposure. Methods are also presented for combining inputs across multiple data types that have varying
quality and different metrics. For example, using monitoring data along with both probabilistic and
deterministic models, the exposure estimates reflect combined exposure pathways and routes separated by
time, geographic scale, and populations of concern. The methods used for toxicity or dose-response
assessments are not much changed from previous EPA approaches; they involve calculating relative
potency factors and determining risk by dose addition (the total margin of exposure used with pesticides
is conceptually similar to the HI).
These extensive and detailed guidance documents articulate several additional important methodological
changes relevant to CRA. The determination regarding which pesticides constitute a common mechanism
group is an important toxicological advance. The dose-response assessment considers a single uncertainty
factor for the chemical group instead of multiple uncertainty factors for the individual chemicals in that
group, reducing the potential for a poorly studied chemical (with a very large uncertainty factor) to drive
the calculation of the mixture risk. The potential application of a 10-fold safety factor for children (a
potentially vulnerable population) is a consideration specifically identified in the FQPA. These methods
have been applied to several pesticide classes, providing valuable examples of CRAs derived from
complex information resources.
Framework for Cumulative Risk Assessment (U.S. EPA, 2003b)
This flexible framework defines a general structure and components for a CRA, laying out three basic
phases of the process (shown in Figure 1-3 of the framework): (1) planning, scoping, and problem
formulation; (2) analysis; and (3) risk characterization. It also describes several technical and coordination
issues and defines many terms relevant to CRA. In addition, the framework suggests some analytical
methods, techniques for analyzing and interpreting data, and other options and tools for carrying out a full
CRA. The steps outlined in this CRA Framework are consistent with those of the ecological risk
guidelines. While the first phase is expanded to include planning, scoping, and problem formulation, the
description is similar to the problem formulation step in the ecological risk guidelines but includes
additional detail.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
B-6
-------
Concepts, Methods, and Data Sources for Cumulative Health Risk Assessment of Multiple
Chemicals, Exposures and Effects: A Resource Document (U.S. EPA, 2007a)
This report is designed as a resource for the EPA and other organizations to use in identifying elements of
and implementing CRAs. It focuses on two areas: (1) concepts concerning CRA initiating factors, along
with procedures for data collection and organization; and (2) technical approaches for assessing and
characterizing health risks associated with a subset of cumulative risk issues (multiple chemicals,
exposures, effects), with examples pertaining to contaminated sites, drinking water, and ambient air.
Some of the emphasis areas and innovations proposed in this document include:
• Developing a description of CRA initiating factors and procedures for population
characterization, data collection, and organization based on these factors.
• Implementing chemical grouping as a potentially helpful way to scope analyses into manageable
pieces to be assessed as chemical mixtures with co-occurring exposures.
• Indicating approaches and data sources for evaluating the timing of exposures, including
discussions of kinetics and dynamics.
• Integrating internal dose measurements to account for multiple-route exposures.
• Further developing the quantitative method for the interaction-based HI, first introduced in the
EPA's 2000 mixtures guidance document.
• Extending the relative potency factor method to cumulate across exposure routes, an approach
first presented in the report on drinking water disinfection by-product (DBP) mixtures.
• Integrating outputs from multiple-effects modeling (illustrated using a categorical regression
model) with the HI and response-addition models to express risks.
• Providing added detail on the cumulative HI approach used by the Superfund Program, including
discussions of the impacts for risk characterization.
• Presenting a method for cumulative risk characterization that considers factors unique to the
conduct of a CRA, including the recognition of uncertainties in cumulative dose-response and
exposure assessment.
• Generally emphasizing close integration of exposure and dose-response analysis, a
recommendation presented in earlier EPA reports, including the methods document on combustor
emissions.
This report showed that resources exist for performing some types of CRAs. One important section
considers how to decide whether a CRA is needed; a table is included that compares CRA with classic
single-chemical, source-based risk assessment. In addition to an extensive glossary, this report also
includes multiple tables and figures to explain and illustrate various concepts and issues and an appendix
(toolbox) with tables and internet links of informational resources covering key aspects of a CRA,
including environmental and health data, predictive modeling and analysis methods, and risk
communication techniques.
Framework for Human Health Risk Assessment to Inform Decision Making (U.S. EPA, 2014b)
The goal of this report is to disseminate information to EPA staff and managers, external stakeholders,
and the public on the general process for conducting human health risk assessments. As such, it focuses
on the context, utility, and planning for risk assessment rather than on providing details for conducting the
assessment. With this orientation, it identifies CRA as an overarching EPA interest and notes that
"combining" does not necessarily mean the risks being assessed should be added; rather, it indicates that
some analysis should be conducted to determine how those risks from various agents or stressors interact.
This report highlights two main points from the 2003 CRA Framework (see Sections 1.4.1.7 and 1.2.1).
The first is the importance of planning, scoping, and problem formulation in designing a risk assessment
that will address a specific need and purpose, which means that the characteristics of a given assessment
will depend on the specific scientific needs and risk management decisions being addressed, which might
not necessarily require a quantitative assessment of cumulative risks. The second point is the importance
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
B-7
-------
of involving the public, especially stakeholders and the community being assessed. This involvement is
emphasized as a key element of risk assessment and an integral part of CRA and environmental justice
actions. Some specifics about the CRA process are contained in the section on planning and scoping,
mainly summarizing concepts presented in the 2003 CRA Framework. The report also identifies examples
of different CRAs for different situations, with links to several EPA reports related to aggregate (multi-
pathway) risks and cumulative risk (including many reports summarized in this appendix).
Pesticide Cumulative Risk Assessment: Framework for Screening Analysis Purpose (U.S. EPA,
2016b)
This guidance was developed to assist scientists and decision-makers in screening pesticides for potential
common mechanism groups (CMGs) and conducting screening-level CRAs. The document provides
guidance for screening available information to identify groups of pesticides that might have a common
mechanism of toxicity (candidate CMGs). This document follows up on previous guidance documents the
Office of Pesticide Programs developed, including Guidance on Cumulative Risk Assessment of Pesticide
Chemicals That Have a Common Mechanism of Toxicity (U.S. EPA, 2002a), which describes the steps
used in conducting CRA on pesticide chemicals. The 2002 CRA guidance provides methods resulting in a
highly refined CRA but requires an extensive number of resources and large amounts of toxicology and
exposure data and might involve sophisticated modeling. The level of refinement provided by that
approach is not necessary or even feasible, however, for all existing pesticide classes. The 2002 CRA
guidance notes that not all cumulative assessments need to be of the same depth and scope and that
determining the need for a comprehensive risk assessment by considering the exposure profile is
important.
The screening-level assessment described in this document applies more conservative approaches and
health-protective overestimates of toxicity and/or exposure than would a refined CRA conducted using
the 2002 CRA guidance. The screening analysis for CRA described in this guidance begins with an
evaluation of the toxicological knowledgebase available.52 If the toxicological characterization of
potential for a common mechanism suggests a candidate CMG can be established, a screening-level
toxicology and exposure analysis could be conducted to provide an initial screen for multiple pesticide
exposure. This framework methodology was developed as an initial screening tier for pesticide CRAs
and, in this respect, illustrates the tiering strategy recommended in this Guidelines document for CRA
planning and problem formulation.
Table B-2. National Research Council and Other Publications Relevant to Cumulative Risk Assessment
Planning and Problem Formulation
Publication
Relevance to CRA Planning and Problem Formulation
Risk Assessment in the Federal
Government: Managing the
Process (NRC, 1983)
Describes the four basic elements of risk assessment (hazard identification, exposure
assessment, dose-response assessment, risk characterization); suggests a good risk
assessment should include nonchemical factors, such as characteristics of the exposed
population and other variables that might affect response.
Understanding Risk: Informing
Decisions in a Democratic Societ\'
(NRC, 1996)
Emphasizes an analytical-deliberative process with problem formulation as a "paramount
consideration."
Framework for Environmental
Health Risk Management, Risk
Assessment and Risk Management
in Regulatory Decision-Making
(PCCRARM, 1997)
Identifies the need to address multiple-issue situations with multisource, multimedia,
multichemical, and multi-risk characteristics and the need to be sensitive to societal
constraints (e.g., political, social, legal, cultural); advocates involvement of stakeholders
throughout the risk assessment process and recommends development and adoption of a
common metric for comparing and assessing diverse health effects.
52 An example is a screening analysis of a particular group of pesticides that would be derived from experimental toxicology
studies submitted for pesticide registration evaluation, as well as from the scientific literature.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
B-8
-------
Publication
Relevance to CRA Planning and Problem Formulation
Guidance Manual for the
Assessment of Joint Toxic Action oj
Chemical Mixtures (ATSDR,
2004)
Presents detailed qualitative methods for evaluating toxicological interactions, including a
structured weight-of-evidence approach that has spawned similar schemes for evaluating
interactions between chemicals and other stressors; advocates evaluation of uncertainties,
health implications of other medical and toxicological factors and sensitive populations,
community-specific health outcome data, and consideration of community health
concerns.
Ensuring Risk Reduction in
Communities with Multiple
Stressors: Environmental Justice
and Cumulative Risks/Impacts
(NE.TAC, 2004)
Recommends a community-based collaborative problem-solving model for addressing
cumulative risks and impacts and the aligning of response to community and risk
assessment goals; includes case studies and lists of tools for screening and assessment.
Phthalates and Cumulative Risk
Assessment. The Tasks Ahead
(NRC, 2008a)
Recommends inclusion in the CRA of all chemicals with common adverse outcomes
instead ol limiting to chemicals grouped by mechanistic similarity.
Science and Decisions: Advancing
Risk Assessment (NRC, 2009)
Advocates assessment of combined risks posed by aggregate exposure to multiple agents
or stressors, including lifestyle, vulnerability, and background risk factors; recommends
developing simplifying tools and databases for screening assessments, including
stakeholder-run assessments, and aligning the assessment effort with the decision
alternatives.
Risk Assessment of Combined
Exposures to Multiple Chemicals:
A WHO/IPCS Framework
(WHO/IPCS, 2009a)
Presents a tiered approach to CRA, moving from simplest, conservative screening or
priority-setting evaluations to detailed, predictive biomathematical models; urges
matching the effort (thus the tier) with the extent of understanding and precision required
for the risk management decision.
Improving Health in the United
States: The Role of Health Impact
Assessment (NRC, 2011a)
Links health with multiple types of decisions, personal and societal, allowing evaluation
of positive and negative impacts of various regulatory decisions on different communities
with varying characteristics.
Environmental Decisions in the
Face of Uncertainty (IOM, 2013)
Identifies a wide variety of uncertainty approaches and recommends that the effort to
analyze specific uncertainties through probabilistic risk assessment or quantitative
uncertainty analysis be guided by the ability of those analyses to affect the environmental
decision; urges development of new graphic methods for communication of cumulative
risk uncertainties to the public.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025 B-9
-------
APPENDIX C. EXAMPLE OF A RESPONSE-BASED CONCEPTUAL MODEL
This appendix provides a more detailed example of a response-based conceptual model (cardiovascular
disease [CVD]) as shown in Figure C-l. The development of this conceptual model for factors
influencing the risk of CVD is detailed in Kashuba et al. (2021), but a brief description is included here as
an example. Subject matter experts in fields including medicine (specifically cardiology), epidemiology,
biostatistics, toxicology, and human health risk assessment participated in model development. The
conceptual model was constructed following four major steps, with indication of corresponding location
in Figure C-l:
1. (Bottom) Begin with the health endpoint—CVD.
2. (One level up) Review the literature to identify the major proximal biological mechanisms
considered to contribute collectively or individually to CVD.
3. (Across the top) List major stressor categories that are related (directly or indirectly) to the
biological mechanisms identified in step 2.
a. A literature search was performed to identify common, predominant factors and topics
associated with CVD as a starting point for discussion with experts. Search terms used
were purposely broad: "causes of cardiovascular disease" and "factors associated with
cardiovascular disease."
b. The results of the literature search were organized and presented to experts for discussion
and included studies that quantified the relationship between identified risk factors and
CVD. Studies were excluded if no data supported the postulated relationship.
4. The stressors identified in step 3 were grouped and related to each other through expert review of
the literature and discussion. Note that this was an iterative process; original stressor
classifications were often revisited and amended; nodes were added and deleted, condensed, and
expanded; and the existence and directionality of relationships were revised. Stressors were
grouped into the following categories:
a. (Far left) Immutable risk factors (e.g., sex, genetics, age) - In a risk assessment and
management context, information on these immutable factors can serve to broaden
understanding of vulnerability within the population and the need for focused risk
management.
b. (Far right) Potentially modifiable risk factors - These represent opportunities wherein
risk management has the potential to be effective in decreasing risk through prevention
and intervention. Such risk-reducing options are also included in the conceptual model.
i. (Right) Risk-increasing behaviors - Examples include sedentary lifestyle,
smoking, diet, and alcohol.
ii. (Near left) Environmental exposures - Includes exposure with both direct and
indirect influence on CVD risk; consider other stressors associated with CVD,
such as lead and cadmium (which water hardness might influence), arsenic, and
halogenated organic compounds.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
C-l
-------
iii. (Near right) Other medical issues and conditions - Arrows indicate direction of
relationships and may be bidirectional (if CVD and the medical issue affect each
other).
iv. (Top) Nonproximal stressors - Factors that do not directly cause CVD but do
modify the magnitude of direct stressors (e.g., socioeconomic influences).
Lipid
status
Halogenated
organics
Obesity
Psychological
stress
Medication
side effects
Arsenic
Mental
and health
disabilities
Alcohol
Atherosclerosis
(plaque buildup
on arteries)
Hypertension
(blood
pressure) /
Heartbeat
arrhythmia
Inflammation
Tissue injury
IMMUTABLE
FACTORS:
Genetics
(family
history of
heart
disease)
Age/
Lifestage
Gender
Low birth
weight
STRESSOR
CATEGORIES:
Socioeconomic conditions that can influence stressors
Environmental exposure
Other medical issues/conditions
Risk-increasing behaviors
Mercury
RISK-
REDUCING
FACTORS:
Diet
(fish oil,
fruits and
vegetables,
low in
saturated
fat,
dietary
fibers)
Physical
activity
Medical
treatment
High-
density
lipoprotein
(HDL)
Adhering to
public
health
advisories
(air quality,
fish MeHg) |
Water
Cadmium
Dietary risk
factors
Infection
(rubella,
herpes
virus)
Diabetes
Sedentary
lifestyle
Smoking
(nicotine, CO,
PAHs,
nitrosamines)
Figure C-l. Conceptual Model for Factors Influencing the Risk of Cardiovascular Disease
CVD = cardiovascular disease.
Prepared for the EPA by Charlie Menzie and Roxolana Kashuba, Exponent.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025 C-2
-------
APPENDIX D. EXPOSURE-RESPONSE MODIFIERS
It has been well characterized that many conditions or factors can contribute to altered levels of exposure
or altered risk of a health effect occurring at a given level of exposure to an environmental pollutant.
Within the scientific literature, there is a large degree of variability in how these conditions or factors are
referred to, but they generally encompass exposure disparities, social vulnerabilities, and biological
vulnerability (Morello-Frosch et al., 2011). For simplicity, in these Guidelines, conditions or factors that
alter a receptor's exposure or response to an environmental pollutant are referred to as exposure-response
modifiers.
In these Guidelines, exposure-response modifiers are discussed largely in terms of vulnerability factors.
Vulnerability in CRA is a multidimensional concept. Kasperson et al. described vulnerability as a product
of exposure, resistance (ability to withstand impacts), and resilience (ability to recover from exposure to a
stressor) (Kasperson et al., 1995). The National Environmental Justice Advisory Committee similarly
defines vulnerability as "a matrix of physical, chemical, biological, social, and cultural factors which
result in certain communities and subpopulations being more susceptible to environmental toxins, being
more exposed to toxins, or having compromised ability to cope with and/or recover from such exposure"
(NEJAC, 2004). This definition explicitly names categories/types of vulnerability factors that are
important determinants of the environmental health of an individual or a community. It refers to the
intrinsic predisposition of an exposed element (individual, community, population, or ecological entity) to
be affected by external stresses and perturbations and the element's ability to recover from such stresses
(resilience). It is based on variations in disease vulnerability, psychological and social factors, exposures,
and adaptive measures to anticipate and reduce future harm and to recover from an insult (Kasperson et
al., 1995; Kasperson et al., 2005; NRC, 2009).
Vulnerability is a differentiating factor for how individuals, communities, populations, or organisms
experience adverse effects related to exposure to environmental stressors. The presence of a vulnerability
factor or condition in an assessment can increase the health impact of exposure to environmental stressors
(deFur et al., 2007). Although the identification of vulnerable human populations often focuses on
individual conditions or factors that can contribute to increased exposures or health risks from
environmental pollutants, such factors do not occur in isolation. Instead, vulnerabilities likely emerge
from a combination of multiple factors or conditions, including both chemical and nonchemical stressors,
at both the community and individual levels. Vulnerability factors can be differentially experienced across
social groups in any given population. This differential experience underscores environmental justice
concerns and uneven distribution of the health-supporting and health-degrading aspects of social
determinants of health (e.g., access to health care and healthy foods, increased exposure to multiple
pollution sources) among different social groups (e.g., by race and class).
As described above, vulnerability can be driven by exposure to a single community- or individual-level
factor or from the complex interactions between multiple factors at multiple levels and environmental
exposures. Figure D-l models the risk paradigm, with stressors on the left affecting receptors in the
middle, and with outcomes on the right. The pathways through which vulnerability factors may operate as
exposure-response modifiers are denoted with dashed arrows. The bidirectional arrows indicate the
dynamic interactions between environment and receptor and the impact of an outcome on the subsequent
vulnerability of a receptor.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
D-l
-------
Environmental conditions
Receptor characteristics
(biological/social/psychosocial)
V.
Exposure
Figure D-l. Vulnerability Factor Categories, Interactions, and Pathways in Exposure-Response Relationships
Source: deFur et al. (2007).
While the focus of discussion has been on human populations, ecological receptors can also exhibit traits
or behaviors that could affect their ability to respond to stressors or could reduce their resilience.
Examples of vulnerable ecological populations include wetland communities stressed by human
development and an endangered species stressed by land-use alterations. A CRA would note, for
example, that for Tribal and Indigenous Peoples and some rural communities, ecological stress may be
linked to human stress as well as to food security and malnutrition.
Vulnerability factors can influence the exposure-response relationship in one or more ways, irrespective
of whether they primarily operate intrinsically (i.e., via preexisting or genetic conditions, as identified in
Figure D-2) or extrinsically through their effects on external conditions):
• Changing the contact (greater or lesser) with a stressor in the environment.
• Changing internal levels (increase or decrease) of a stressor once contact has occurred such that
for the same amount of external exposure, a vulnerable person has different (higher or lower)
internal levels of the stressor or its toxic form, which could emerge through influences on
toxicokinetic processes (absorption, distribution, metabolism, and excretion).
• Altering how a given internal level of the stressor affects target tissues and organs by influencing
toxicodynamic processes such as DNA repair, which serves to change processes that impose
tolerance and resilience and thus affects the occurrence or the severity of disease.
• Influencing disease pathways, the stressor changes (potentiates or reduces) the adverse outcomes
associated with exposure.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
D-2
-------
Figure D-2. Key Intrinsic Events in Exposure-Response Relationships
Source: Ginsberg et al. (2014).
In CRA, understanding how vulnerability factors might influence the assessment is critical during
problem formulation. Vulnerability factors should be evaluated as "exposure modifiers" or "response
modifiers," recognizing that the pathway to modification could occur via influences on conditions
extrinsic or intrinsic to the individual.
deFur et al. (2007) and Ginsberg et al. (2014) offer helpful frameworks for examining vulnerability in this
regard (see Figure D-l and Figure D-2). The framework of deFur et al. emphasizes the need to clarify the
relationships among the various factors and their roles in affecting vulnerability (deFur et al., 2007) - the
influence of vulnerability factors on exposure-response relationships should be evaluated in terms of the
relationships and interactions among the vulnerability factors, identified stressors, and outcomes of
interest.
The Ginsberg et al. (2014) framework emphasizes delineation of key intrinsic events in the exposure-
response relationship to understand vulnerability. This framework approaches vulnerability through the
lens of the effect of preexisting disease on toxicokinetics and toxicodynamics, although its main
constructs are applicable to other vulnerability factors. Three pathways are proposed:
• Altered chemical processing by disease in ways that materially change the internal dose;
• Weakened host defense mechanisms that impair tolerance; and
• Altered disease pathways that the disease process also affects.
Segal et al. (2015) proposed a framework for integrating the effect of a single nonchemical stressor within
the context of a risk assessment for a single-chemical stressor (Figure D-3). The approach considers the
nonchemical stressor as a modifier of response. The framework offers a method for considering a
vulnerability factor in a one-chemical risk assessment and thus has some utility in thinking about
assessments of cumulative risks for chemical and nonchemical stressor combinations.
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
D-3
-------
Does interaction
occur between the
stressor being
assessed and the
exposure/response
modifier under
consideration?
Does the potential exposure/response
modifier co-occur and exist
disproportionately in certain populations?
Does the potential exposure/response
modifier produce similar adverse health
effects?
Does the potential exposure/response
modifier affect similar signaling or response
pathways?
I
Z
Is there evidence for an interaction in
epidemiological or experimental studies?
Interrelated and sequential
questions
. Options for incorporation
into risk evaluation
Figure D-3. A Framework for Incorporating Nonchemical Stressors into Risk Assessments
Adapted from Segal et al. (2015).
Considering the social determinants of health augments the set of vulnerability factors typical in risk
assessment, such as lifestage and preexisting disease. In traditional risk assessments, social and individual
determinants other than stressors being assessed might be integrated into risk assessments as vulnerability
factors. In CRA scenarios, however, it is plausible that a primary/target stressor could also operate as a
vulnerability factor, modifying how one or more other stressors in the mix lead to adverse health
outcomes. Thus, a vulnerability factor may be considered an "exposure modifier," a "response modifier,"
or both.
Such exposure-response modifiers can be integrated into risk assessment frameworks such as Segal et al.
(2015).
Integrating Vulnerability into Problem Formulation. A systematic approach is necessary for identifying
and examining evidence of vulnerability during problem formulation. This ensures the CRA team is
aware of key vulnerability considerations in the assessment and can develop well-informed hypotheses
about important risk and adverse outcomes for the assessment. The following questions are suggested to
inform the integration of vulnerability in the hypotheses for the assessment and in the development of a
conceptual model for the assessment:
• What factors (external to the set of stressors that are the focus of the CRA) are potential modifiers
of exposure or response for one or more stressors in the CRA? This inquiry can be initiated from
the toxicological and epidemiological data for stressors. For example, if Chemicals A, B, and C
are the focus of a CRA, what factors might modify exposure or response to one or any
combination of A, B, and C?
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025 D-4
-------
o Is there a factor/set of factors (chemical, biological) or condition(s) that can modify the
exposure and/or response of A, B, or C or any combination thereof? For this discussion,
refer to these other factors/conditions as "E" and "F."
o What type of evidence is available to support this finding of exposure or response
modification—toxicological, epidemiological?
o Is the information sufficient to estimate differentiated dose-response assessments for each
chemical, A, B, or C?
• If data suggest exposure or response modification, what is the extent to which these factors (E or
F) co-occur with the stressors (A, B, or C) in the population of interest?
• Given available data, what are plausible ways in which these factor(s) might increase or decrease
exposure or response?
In addition to approaching the identification of vulnerability factors through existing epidemiological and
toxicological data specific to the stressors in the assessment (A, B, and C), it may be helpful or necessary
to identify potential vulnerability factors by researching conditions that are characteristic of or important
to the population(s) of interest. This inquiry should be a routine step during communication with
stakeholders. By identifying community-level characteristics and concerns, the CRA team will be able to
better scrutinize the epidemiological and toxicological literature further for information that might inform
potential exposure-response modification by these factors. Consistent with the U.S. Department of Health
and Human Services' Healthy People 2020 Social Determinants of Health Framework (U.S. HHS, 2020),
the CRA team should consider the following questions:
• To what extent do one or more issues within one or more categories of social factors (e.g., food
insecurity as an economic stability issue) co-occur with stressors A, B, or C in the population?
o Do populations with higher exposures to A, B, or C experience greater food insecurity
(food insecurity becomes a potential E or F)?
o Is exposure to an economic stability issue (e.g., food insecurity) prevalent in the
population that is the focus of the CRA?
• To what extent are one or more economic stability issues prevalent among populations that
experience the health outcomes linked with exposure to stressors A, B, or C?
o Is the economic stability issue prevalent in the population that is the focus of the CRA?
• Do the epidemiological and toxicological data offer information to indicate whether these
economic stability issues might modify exposure or response?
• What are plausible pathways (given available data) through which an economic stability issue
might modify exposure or response?
Guidelines for Cumulative Risk Assessment Planning and Problem Formulation - 2025
D-5
-------
&EPA
United States
Environmental Protection
Agency
PRESORTED STANDARD
POSTAGE & FEES PAID
EPA
PERMIT NO. G-35
Office of Research and Development (8101R)
Washington, DC 20460
Official Business
Penalty for Private Use
$300
Recycled/Recyclable Printed on paper that contains a minimum of
50% postconsumer fiber content processed chlorine free
------- |