TSCA: EPA Releases Proposed Chemical Risk Evaluation Process under New TSCA
On January 19, 2017, the U.S. Environmental Protection Agency (EPA) proposed a process for conducting risk evaluations to determine whether a chemical substance presents an unreasonable risk of injury to health or the environment, including an unreasonable risk to a potentially exposed or susceptible subpopulation, under the conditions of use. 82 Fed, Reg. 7562. The process would not consider costs or other non-risk factors. Risk evaluation is the second step, after prioritization, in a new process of existing chemical substance review and management established under recent amendments to the Toxic Substances Control Act (TSCA). The proposed rule identifies the steps of a risk evaluation process including scope, hazard assessment, exposure assessment, risk characterization, and finally a risk determination. EPA proposes that this process be used for the first ten chemical substances to be evaluated from the 2014 update of the TSCA Work Plan for Chemical Assessments, chemical substances designated as High-Priority Substances during the prioritization process, and those chemical substances for which EPA has initiated a risk evaluation in response to manufacturer requests. The proposed rule also includes the required “form and criteria” applicable to such manufacturer requests. Comments are due March 20, 2017. More information regarding EPA’s proposed rule concerning the first step in the process of reviewing existing chemicals, prioritization, is available in our January 18, 2017, memorandum “EPA Proposes Procedures to Prioritize Chemicals for Risk Evaluation under TSCA.”
Components of a Risk Evaluation
New TSCA Section 6(b)(4)(D) identifies the minimum components that EPA must include in all chemical substance risk evaluations. For each risk evaluation, EPA must publish a document outlining the scope of the risk evaluation that will be conducted. The document must include the hazards, exposures, conditions of use, and the potentially exposed or susceptible subpopulations that EPA expects to consider. Under new TSCA, the scope of the risk evaluation must be published no later than six months after the initiation of the risk evaluation. New TSCA Section 6(b)(4)(F) requires each risk evaluation to: (1) integrate and assess available information on hazards and exposures for the conditions of use of the chemical substance, including information that is relevant to specific risks of injury to health or the environment and information on potentially exposed or susceptible subpopulations; (2) describe whether aggregate or sentinel exposures to a chemical substance under the conditions of use were considered, and the basis for that consideration; (3) not consider costs or other nonrisk factors; (4) take into account, where relevant, the likely duration, intensity, frequency, and number of exposures under the conditions of use of the chemical substance; and (5) describe the weight of the scientific evidence for the identified hazard and exposure.
According to EPA, the risk evaluation process under new TSCA is ultimately how it will determine whether a chemical substance presents an unreasonable risk of injury to health or the environment. The overall objective of the proposed rule is to codify the process by which EPA evaluates risk from chemical substances for purposes of new TSCA Section 6. In the proposed rule, EPA discusses those components of TSCA risk evaluation and key factors that EPA deems are necessary to consider in each risk evaluation to ensure that the public has a full understanding of how risk evaluations will be conducted. EPA states that it “is not proposing to establish highly detailed provisions that will address every eventuality or possible consideration that might arise.” Due to the rapid advancement of the science of risk evaluation and the science and technology that inform risk evaluation, according to EPA, the proposed rule “seeks to balance the need for the risk evaluation procedures to be transparent, without unduly restricting the specific science that will be used to conduct the evaluations, allowing the Agency flexibility to adapt and keep current with changing science as it conducts TSCA evaluations into the future.”
EPA states that it recognizes that other federal agencies may be able to provide important use, exposure, and hazard information that is likely to be relevant to its risk evaluations of chemical substances. EPA is committed to interagency engagement and dialogue. As such, according to the proposed rule, EPA has reached out to other agencies, inviting them to join in an “open and collaborative” dialogue for chemicals management and risk evaluations under TSCA.
To coordinate with other agencies on TSCA implementation generally, EPA intends to continue to use — and expand where appropriate — existing interagency groups, such as the OMNE (U.S. Occupational Safety and Health Administration (OSHA)-Mining Safety and Health Administration (MSHA)-National Institute for Occupational Safety and Health (NIOSH)- National Institute of Environmental Health Sciences (NIEHS)-EPA) Committee and the National Science and Technology Council’s (NSTC) Committee on Environment, Natural Resources, and Sustainability’s new Toxicity Assessment Committee. EPA is also committed to interagency engagement at the working level on individual chemical evaluations.
To ensure that such collaboration can occur in a timely manner when needed, EPA states that it intends to initiate interagency consultation through the existing mechanisms early in the process, and document these measures in the scope document. EPA is concerned that imposing a single, pre-determined consultation step might lead to an overly bureaucratic process that could limit or complicate ongoing collaboration efforts, however, and so is not proposing to codify any particular process.
Scope of Evaluations
TSCA requires risk evaluations to determine whether a chemical substance presents an unreasonable risk of injury to health or the environment under the conditions of use, with conditions of use being defined as “the circumstances, as determined by the EPA, under which a chemical substance is intended, known, or reasonably foreseen to be manufactured, processed, distributed in commerce, used, or disposed of.”
Although some of the commenters during the public meeting suggested that EPA could evaluate a specific use of a chemical substance, the proposed rule does not adopt such an interpretation. EPA states that it recognizes that under certain circumstances it may be necessary to expedite an evaluation for a particular condition of use to move more rapidly to risk management under new TSCA Section 6(a), including a situation in which a single use presented an unreasonable risk of injury for the population as a whole or for a susceptible subpopulation (e.g., one use results in risks that EPA would determine unreasonable regardless of the risk posed by other uses). In a case where EPA would find it necessary to pursue a risk evaluation in phases, however, EPA states that it will still complete the full risk evaluation on all identified conditions of use within the statutory three-year deadline. Therefore, relying on this discretion, EPA proposes to recognize explicitly its authority to complete risk evaluations in phases, and to manage unreasonable risks as they are identified through those phases under new TSCA section 6(a) in the regulation.
New TSCA defines a number of key terms necessary for interpretation of the new law. The definitions apply to the proposed rule. EPA included some additional definitions in the proposed rule and also proposes to clarify certain statutory definitions.
New TSCA requires EPA to evaluate risk to “potentially exposed or susceptible subpopulation[s],” and although the law defines this term, EPA proposes to expand the definition for TSCA purposes. New TSCA states that “the term ‘potentially exposed or susceptible subpopulation’ means a group of individuals within the general population identified by the EPA who, due to either greater susceptibility or greater exposure, may be at greater risk than the general population of adverse health effects from exposure to a chemical substance or mixture, such as infants, children, pregnant women, workers, or the elderly.” EPA proposes to incorporate the phrase “including but not limited to” before the specific subpopulations identified in the statutory definition, to clarify further that EPA may identify additional subpopulations, where warranted. EPA also proposes to include specific authorization for it to consider both intrinsic (e.g., life stage, reproductive status, age, gender, and genetic traits) and acquired (e.g., pre-existing disease, geography, socioeconomic, cultural, and workplace) factors when identifying this population.
Revising statutory definitions is tricky unless strongly justified. Further, we are concerned by a proposal using unclear terms. We suggest that the better approach is to make such points as part of a policy or guidance document that would allow for development of a clearer understanding over time and as experience is gained. We believe that this approach would allow EPA to explain more clearly the possible additional subpopulations that are under consideration and to provide a thorough explanation of intrinsic and acquired factors.
New TSCA Section 26(k) states that in carrying out risk evaluations, EPA shall consider information that is “reasonably available,” but does not further define this phrase. EPA proposes a definition for “reasonably available” to mean existing information that EPA possesses, or can reasonably obtain and synthesize for use in risk evaluations, considering the deadlines for completing the evaluation. Generally speaking, according to the proposed rule, EPA does not consider information that has not yet been generated as reasonably available, because it will typically not be feasible for EPA to require significant chemical testing and receive and assess those test results during the three to three and a half year window allotted for risk evaluations. Accordingly, EPA intends to ensure generally that sufficient information to complete a risk evaluation exists and is available prior to initiating the evaluation (indeed, as discussed in our January 18, 2017, memorandum, prior to initiating prioritization). EPA also generally intends to use its authority under new TSCA to require the development of new information, as necessary, prior to prioritization.
New TSCA requires EPA, as a part of the risk evaluation, to document whether it has considered aggregate or sentinel exposure, and the basis for that decision. New TSCA does not define these terms, so EPA proposes a definition for aggregate exposure that is consistent with its current policies and practices. “Aggregate exposure” means the combined exposures to an individual from a single chemical substance across multiple routes and across multiple pathways. “Sentinel” means the exposure(s) of greatest significance, which may be the maximum exposure to an individual, population (or subpopulation), or the environment to the chemical substance of interest (or any combination thereof). EPA notes that although sentinel exposure “is not a novel way of characterizing exposure, this is a new term for EPA.” Given this admission by EPA, it would be helpful to more clearly understand how this parameter is proposed to be determined and used. In this regard, we note that this definition does not align with the practice in established programs where sentinel exposure(s) is a key component. The better known application is in the area of consumer products where a sentinel exposure represents the highest plausible exposure profile for a group of products based on use scenario, physicochemical properties, and bioavailability.
According to EPA, other terms defined in the proposed rule are designed to provide clarity regarding the science that will be used to conduct an evaluation. “Pathways” of exposure refer to the mode through which one is exposed to a chemical substance, including but not limited to: food, water, soil, and air. “Routes” of exposure refer to the particular manner in which a chemical substance may contact the body, including absorption via ingestion, inhalation, or dermally. New TSCA requires EPA to consider “the extent to which the variability and uncertainty . . . are evaluated and characterized.” EPA is proposing to adopt definitions for both “variability” and “uncertainty” from existing EPA guidance. “Uncertainty” means the imperfect knowledge or lack of precise knowledge either for specific values of interest or in the description of a system. “Variability” means the inherent natural variation, diversity, and heterogeneity across time and/or space or among individuals within a population.
Timing of Risk Evaluations
As indicated, new TSCA requires EPA to complete risk evaluations within three years, with the possibility of a six month extension beyond the three year timeframe. The proposed rule adopts these timeframes without modification or elaboration. EPA acknowledges this is a relatively short timeframe, and is proposing to adopt other procedures that will allow it to meet these deadlines.
Chemical Substances for Risk Evaluation
Chemical substances that will undergo risk evaluation can be put into three groups: (1) the first ten chemical substances EPA is required to identify within the first 180 calendar days of enacting the amendments to TSCA; (2) the chemical substances determined as High-Priority Substances through the prioritization process proposed in a separate rulemaking; and (3) requested chemical substances submitted by manufacturers that meet the criteria for EPA to conduct a risk evaluation.
Process for Manufacturer Requested Risk Evaluations
New TSCA allows a manufacturer or group of manufacturers to submit requests for EPA-conducted risk evaluations for chemical substances that they manufacture. EPA proposes to require that manufacturers demonstrate in their request that there is sufficient, reasonably available information for it to conduct a risk evaluation on the chemical substance under the conditions of use. EPA must complete any manufacturer-requested risk evaluation that it determines meets the criteria within the statutory three years. Unlike chemical substances that have come through the prioritization process, manufacturer-requested chemical substances have not undergone initial risk screening, and EPA states that it will not assign such chemicals a high- or low-priority designation. The purpose of the proposed required components of the request is to allow EPA to determine whether sufficient information is “reasonably available” for EPA to complete a risk evaluation of the requested chemical under the conditions of use, as that term is defined under new TSCA Section 3.
EPA proposes to require a manufacturer to submit a list (e.g., citations) of the reasonably available information on hazard and exposure for all the conditions of use. EPA states that it is not requesting that manufacturers submit copies of the cited information. Manufacturers must include a commitment to provide EPA any referenced data if they are not publicly available, and must certify that the information submitted is accurate and complete. EPA will not accept a manufacturer request where any of the relevant data is not in the possession of the requestor, but is with another entity. We applaud EPA’s approach that as proposed does not require potentially duplicative and burdensome reporting of copies of studies in the initial request.
Consistent with new TSCA Section 6(b)(4)(E)(iii), EPA will prioritize requests where there is evidence that restrictions imposed by one or more states have the potential to have a significant impact on interstate commerce or health or the environment, and is therefore proposing to allow (but not require) manufacturers to include any evidence to support such a finding. Following this required initial prioritization, EPA proposes to prioritize further chemical substances for risk evaluation based on initial estimates of exposure(s) and/or hazard(s) under one or more conditions of use or any other factor that EPA determines may be relevant. In general, according to the proposed rule, EPA plans to prioritize those chemical substances where there is evidence of relatively high risk over those with less evidence of risk.
Instructions for submitting confidential business information (CBI) are also included in the proposed rule. EPA states that it believes that new TSCA Section 14(c)(3) is best read as requiring upfront substantiation of non-exempt CBI claims. In addition, EPA believes the obligation to review all non-exempt chemical identification claims and 25 percent of all other non-exempt claims will be best effectuated by requiring substantiation at the time of submission.
Chemical substances that EPA has prioritized through the prioritization process are subject to two separate public comment periods prior to the completion of the prioritization process. EPA expects that these comment periods will ensure that it has the necessary information to evaluate the chemical substances, including information on all conditions of use. Consequently, to ensure that chemical substances subject to manufacturer requests undergo risk evaluation only if the available information is comparable to what EPA will identify or generate through the measures identified in the proposed prioritization framework rule, EPA proposes opportunities to collect additional information from the public.
Upon receipt of the request, EPA proposes to verify that the request is facially valid, i.e., that information has been submitted that is consistent with the regulatory requirements. EPA proposes that within 30 business days of a receiving a facially valid request, EPA will submit for publication an announcement of the receipt of the request in the Federal Register, open a docket for the request, and provide no less than a 30-calendar day comment period to allow the public to identify and/or submit any reasonably available information regarding hazard, exposure, potentially exposed population(s) and subpopulation(s), and conditions of use that may help inform a risk evaluation, including identifying information gaps. EPA notes that the requesting manufacturer may also submit any additional material during this time.
Within nine months after the end of the comment period, EPA will review the request along with any additional information received during the comment period to determine whether the request meets the regulatory criteria and will notify the manufacturer(s) accordingly. This time will allow EPA to develop the equivalent of a conceptual model to describe actual or predicted relationships between the chemical substance and the receptors, either human or environmental, with consideration of potential hazards throughout the life cycle of the chemical substance — from manufacturing, processing, distribution in commerce, storage, use, or disposal. If EPA determines that the request is compliant (i.e., that it has the required information necessary for conducting a risk evaluation), EPA will begin the risk evaluation process consistent with new TSCA Section 6(b)(4)(E)(i). If the request is found insufficient, EPA will identify the information that would be necessary to conduct the risk evaluation in its notification to the manufacturer. The manufacturer will have 60 calendar days from receipt of EPA’s determination to submit the additional information. EPA states that it will consider the request withdrawn if the manufacturer(s) fails to submit the additional information identified. The process for conducting the risk evaluation will otherwise be identical to the process for those chemicals identified as High-Priority Substances through the Prioritization Process, which is addressed in a separate proposed rule and discussed in our January 18, 2017, memorandum.
Risk Evaluation General Provisions
EPA has a number of existing guidance documents that inform its risk assessment approaches and these have been used by EPA to characterize the nature and magnitude of health risks to humans and ecological receptors from chemicals. Over the years, EPA has worked with the scientific community and other stakeholders to develop a variety of guidance, guidelines, methods and models for use in conducting different kinds of assessments. A compendium of existing EPA guidance related to risk assessments is maintained at https://www.epa.gov/risk/risk-assessment-guidelines. A compendium of guidance, databases and models used for assessing pesticide risks is available at https://www.epa.gov/pesticide-science-and-assessing-pesticide-risks, and information about available predictive models and tools for assessing chemicals under TSCA can be found at https://www.epa.gov/tsca-screening-tools. EPA notes that each of these websites identifies and links to a number of written guidance documents, tools, and models. Rather than starting anew, EPA states that it intends to take advantage of its existing guidance, tools, and models that are relevant and available for use in conducting a risk evaluation under this program.
EPA notes that since new TSCA requires the development of additional “policies, procedures, and guidance the Administrator determines are necessary” to carry out the process in new TSCA, EPA may also develop additional guidance(s) for risk evaluation in the future.
Categories of Chemical Substances
TSCA provides EPA with authority to take action on categories of chemical substances — groups of chemical substances that are, for example, similar in molecular structure, in physical, chemical, or biological properties, in use, or in mode of entrance into the human body or into the environment. EPA notes that although the proposed rule most often references “chemical substances,” EPA also proposes to include a clear statement in the regulation that nothing in the proposed rule shall be construed as a limitation on EPA’s authority to take action with respect to categories of chemical substances, and that, where appropriate, EPA can prioritize and evaluate categories of chemical substances.
Information and Information Sources
The timeframe for completing risk evaluation is compressed. For those chemical substances chosen by EPA to undergo the risk evaluation process, EPA expects to initiate the process only when EPA has determined that most of the information necessary to complete the evaluation is reasonably available, which in most cases means the information already exists. EPA states that as appropriate, however, it will exercise its TSCA information collection, testing, and subpoena authorities, including those under new TSCA Sections 4, 8, and 11(c) to develop the information needed for a risk evaluation. EPA also reminds readers of TSCA Section 8(e) that requires any person who manufacturers, processes, or distributes in commerce a chemical or mixture and who obtains information that supports the conclusion that this substance presents a substantial risk of injury to health or the environment, shall immediately inform EPA.
To conduct a risk evaluation, EPA will rely on a combination of information, models, screening methods, and accepted science policies, including defaults, reasonable estimates, and uncertainty factors, in addition to considering information generated from evolving science and technology. EPA states that it expects to obtain scientific advice from the Science Advisory Committee on Chemicals, which EPA is required to develop and convene under new TSCA Section 26(o). In compliance with the statute, EPA will work to reduce and replace, to the extent practicable, the use of vertebrate animals in testing chemical substances as outlined in new TSCA Section 4(h).
Risk Evaluation Steps
The first step of a risk evaluation is the development of the scope. In compliance with new TSCA, the scope will identify the conditions of use, hazards, exposures, and any potentially exposed or susceptible subpopulations that EPA expects to consider. EPA also proposes to include additional information in the scoping document, including any models, screening methods, and any accepted science policies expected to be used during the risk evaluation. EPA further proposes to include a conceptual model that will describe the actual or predicted relationships between the chemical substance and the receptors, either human or environmental, with consideration of potential hazards throughout the life cycle of the chemical substance — from manufacturing, processing, distribution in commerce, storage, use, to release, or disposal. Also included will be an analysis plan that will identify the approaches and methods EPA plans to use to assess exposure, effects, and risk, including associated uncertainty and variability, as well as a strategy for approaching science policy decisions (e.g., defaults or uncertainty factors). It is not clear, however, whether this step includes the elements of problem formulation as postulated by the National Academy of Sciences National Research Council (NAS/NRC) and as used by EPA in the Work Plan process.
The announced availability of the final scope will be published in the Federal Register within six months of the initiation of the risk evaluation. Although not required under new TSCA, EPA proposes to provide a draft scope for a 45-calendar day public comment period during this six month period. EPA states that it “welcomes all public participation, but specifically encourages commenters to provide information they believe might be missing or may further inform the risk evaluation.” That said, according to the proposed rule, EPA expects to use the comment periods during the prioritization process to reduce the likelihood of significant comments on the draft scope. Consequently, the proposed rule makes clear that all comments that could be raised on information and approaches presented in the scope must be presented during this comment period. According to EPA, any issues related to scope not raised in comments at this time cannot form the basis for an objection or challenge in a future administrative or judicial proceeding. EPA states: “This is a well-established principle of administrative law and practice, see, e.g., Nuclear Energy Institute v. EPA, 373 F.3d 1251, 1290-1291 (D.C. Cir. 2004), and the need for such a provision is reinforced by the statutory deadlines under which EPA must operate for completing TSCA risk evaluations.” EPA notes that it is not proposing to preclude parties from raising newly discovered information, or from raising issues that could not have been fairly raised during this comment period. Rather, “EPA seeks merely to prevent parties from delaying the risk evaluation by withholding information or by providing it piecemeal.”
In compliance with new TSCA Section 6(b)(4)(F), EPA proposes that a hazard assessment be conducted on each chemical substance or category. A hazard assessment identifies the types of adverse health or environmental effects that can be caused by exposure to some agent in question, and to characterize the quality and weight of evidence (WOE) supporting this identification. Hazard Identification is the process of determining whether exposure to a stressor can cause an increase in the incidence of specific adverse health or environmental effects (e.g., cancer or developmental toxicity).
The hazard assessment may include, but may not be limited to, evaluation of the potential toxicity of the chemical substance with respect to cancer; mutation; reproductive, developmental, respiratory, immune, metabolic, and cardiovascular impacts; and neurological impairments. The assessment will evaluate effects at life stage(s) most appropriate for a receptor target. The hazard assessment will consider the dose or concentration and resulting effect or response. Potential information sources that may support the health assessment include but are not limited to human epidemiological studies; in vivo and/or in vitro laboratory studies; mechanistic or kinetic studies in a variety of test systems, including but not limited to toxicokinetics and toxicodynamics; computational toxicology; data from structure-activity relationships; high-throughput assays; genomic response assays; and ecological field data. Specifically, according to the proposed rule, for human health hazards the assessment will consider all potentially exposed or susceptible subpopulation(s) identified in the scope and use appropriate combination, if available, of population-based epidemiological studies, information related to geographic location of susceptible subpopulations, models representing health effects to the population, and any other relevant, scientifically valid information, or methodology. In an environmental hazard assessment, the relationship between the chemical substance and the occurrence of an ecological response will be evaluated using field or laboratory data, modeling strategies, and species extrapolations.
Where possible, a hazard assessment will also include a dose-response assessment. A dose-response relationship describes how the likelihood and severity of adverse health effects (the responses) are related to the amount and condition of exposure to an agent (the dose provided). The same principles generally apply for studies where the exposure is to a concentration of the agent (e.g., airborne concentrations applied in inhalation exposure studies or water or other media concentrations for ecological exposure studies), and the resulting information is referred to as the concentration-response.
Pursuant to new TSCA Section 6(b)(4)(F), EPA, where relevant, will take into account the likely duration, intensity, frequency, and number of exposures under the conditions of use in an exposure assessment. An exposure assessment includes some discussion of the size, nature, and types of individuals or populations exposed to the agent, as well as discussion of the uncertainties in this information. EPA states that exposure can be measured directly, but more commonly is estimated indirectly through consideration of measured concentrations in the environment, consideration of models of chemical transport and fate in the environment, and estimates of human intake or environmental exposure over time.
Using reasonably available information, exposures will be estimated (usually quantitatively) for the identified conditions of use. For human health exposure, the assessment would consider all potentially exposed or susceptible subpopulation(s) identified in the scope and use any combination, as available, of population-based epidemiological studies, information related to geographic location of susceptible subpopulations, models representing exposures to the population, measurements in human tissues or relevant environmental or exposure media, and any other relevant, scientifically valid information or methodology. In an environmental exposure assessment, the interaction of the chemical substance with any ecological characteristics identified in the scope will be characterized and evaluated.
New TSCA requires that a risk evaluation “integrate and assess available information on hazards and exposures.” According to EPA, a risk characterization conveys the risk assessor’s judgment as to the nature and presence or absence of risks, along with information about how the risk was assessed, where assumptions and uncertainties still exist, and where policy choices will need to be made. Risk characterization takes place for both human health risk assessments and ecological risk assessments.
In practice, each component of the risk assessment (e.g., hazard assessment, dose-response assessment, or exposure assessment) has an individual characterization written to carry forward the key findings, assumptions, limitations, and uncertainties. EPA states that the set of these individual characterizations provide the basis of information on which to write an integrative risk characterization analysis. The final, overall risk characterization thus consists of the individual component characterizations plus an integrative analysis.
Each risk evaluation will quantitatively and/or qualitatively estimate and characterize risk for the identified populations and ecological characteristics under the conditions of use. The risk characterization will also describe whether aggregate or sentinel exposures were considered and provide the evidence and information to support the consideration.
In the risk characterization, EPA will further carry out the obligations under new TSCA Section 26(h); for example, by assessing uncertainty and variability in each step of the risk evaluation, discussing considerations of data quality such as the reliability, relevance, and whether the methods used were reasonable and consistent, explaining any assumptions used, and discussing information generated from independent peer review. EPA states that it may also exercise its discretion to include a discussion of any alternative interpretation of results generated from the risk evaluation. For environmental evaluations specifically, EPA plans to include a discussion of the nature and magnitude of the effects, the spatial and temporal patterns of the effects, implications at the individual, species, and community level, and the likelihood of recovery subsequent to exposure to the chemical substance.
For each risk evaluation conducted on chemicals identified pursuant to new TSCA Section 6(b)(4)(A), EPA will conduct peer reviews using the guidance provided in executive branch peer review directives included in the Office of Management and Budget Final Information Quality Bulletin for Peer Review (Office of Management and Budget (OMB) Bulletin) and the guidance set forth in EPA’s 2015 Peer Review Handbook (EPA Peer Review Handbook), or its updates.
EPA states that the goal of the peer review process is to obtain independent review from experts who have not contributed to its development. According to EPA’s peer review policy, peer review of all scientific and technical information that is intended to inform or support EPA decisions is encouraged and expected. Both the EPA Peer Review Handbook and the OMB Bulletin provide standards for when and how to conduct peer review on science documents. According to EPA, the documents do not contemplate that peer review is necessary for every document or risk assessment, but is expected to occur for those documents that have either:
- Influential scientific information: scientific information that EPA reasonably can determine will have or does have a clear and substantial impact on important public policies or private sector decisions; or
- Highly influential scientific assessment: a subset of influential scientific information that could have a potential impact of more than $500 million in any year on either the public or private sector, or is novel, controversial, or precedent-setting, or has significant interagency interest.
The EPA Peer Review Handbook, first released in 1998 and last updated in 2015, has also been instrumental in providing guidance on the methods for conducting peer review at EPA for the past two decades. According to the Handbook, the peer review approach can consist of internal or external reviewers and can range from a letter review, an ad hoc expert panel review, review of a journal manuscript by a referred scientific journal, review by an established Federal Advisory Committee (FAC), review by an EPA-appointed special board or commission, or review by the NAS. EPA states that given that this guidance reflects long-standing and well-accepted EPA practices on peer review, and given the public’s familiarity with it, EPA proposes to continue to rely on that established guidance, rather than attempt to modify it or create some new methodology in this rulemaking. As discussed earlier in this proposal, EPA will identify aspects of the analysis on which peer review will be conducted, and the planned methodologies, as part of the draft scoping document that will undergo public comment for each chemical substance that undergoes risk evaluation. These may include novel models or analyses that warrant an in-depth peer review. In addition to any targeted peer review of a specific aspects of the analysis, the entire risk assessment will also undergo peer review, “as it is important for peer reviewers to consider how the various underlying analyses fit together to produce an integrated risk characterization which will form the basis of an unreasonable risk determination.”
EPA states that the peer review will address aspects of the science underlying the assessment, including, but not limited to hazard assessment, assessment of dose-response, exposure assessment, and risk characterization. EPA notes that it will not seek review of any determination as to whether the risks are “unreasonable,” which is an EPA policy judgement. The purpose of peer review is for independent review of the science underlying the risk assessment, not to evaluate EPA’s policy judgments. This is an important clarification that frequently gets lost in the process. EPA needs to bring this aspect out more prominently when presenting the results of a risk evaluation to highlight that the peer review process does not address policy conclusions/outcomes — only the scientific components.
TSCA expressly reserves to EPA the final determination of whether risk posed by a chemical substance is “unreasonable.” According to the proposed rule, “EPA nevertheless will include its unreasonable risk judgment as part of the risk evaluation that is subject to public review and comment.”
Unreasonable Risk Determination
The final step of a risk evaluation is for EPA to determine whether the chemical substance presents an unreasonable risk of injury to health or the environment. EPA may find that the substance does not present an unreasonable risk of injury to health or the environment under the conditions of use. This will be issued by order, published in the Federal Register, and considered to be a final EPA action. Alternatively, EPA may determine that the substance does present an unreasonable risk under one or more conditions of use, in which case EPA must, pursuant to new TSCA Section 6(a), impose requirements to the extent necessary so that the substance no longer presents such risk.
EPA will announce in the Federal Register the availability of and solicit public comment on the draft risk evaluation, including the unreasonable risk determination. EPA states that all comments that could be raised on components of the draft risk evaluation must be presented during this comment period. Under the proposed rule, any issues not raised during this time will be considered to have been waived, and may not form the basis for an objection or challenge in any subsequent administrative or judicial proceedings.
Additional Publically Available Information
Pursuant to new TSCA Section 26(j), EPA will make available: (1) all notices, determinations, findings, consent agreements, and orders; (2) any information required to be provided by EPA under new TSCA Section 4; (3) a nontechnical summary of the risk evaluation; (4) a list of the studies with the results of the studies, considered in carrying out each risk evaluation; and (5) the final peer review report, including the response to peer review comments.
Reassessment of Unreasonable Risk Determination
EPA may reassess a final unreasonable risk determination of a chemical substance at any time based on information available to it.
Request for Comments
While EPA is seeking public comment on all aspects of this proposed rule, EPA states that it specifically requests public input on the areas below.
Redefining Scientific Terms
EPA states that it received a number of stakeholder comments regarding its approach to defining a number of important terms within the proposed rule. These terms include “best available science,” “weight-of-the-evidence,” “sufficiency of information,” “unreasonable risk,” and “reasonably available information,” among others. EPA notes that many of the terms used in the proposed rule are not novel concepts and are already in use — the meanings of which are discussed extensively in existing EPA guidance. For example, extensive descriptions for the phrases “best available science,” “weight-of-the-evidence,” and “sufficiency of information” can be found in EPA’s Risk Characterization Handbook, and in other existing EPA guidance. EPA states that it “believes further defining these and other terms in the proposed rule is unnecessary and ultimately problematic. These terms have and will continue to evolve with changing scientific methods and innovation.” According to EPA, codifying specific definitions for these phrases “may inhibit the flexibility of the Agency to quickly adapt and implement changing science.” Be that as it may, we believe that a definition of best available science under TSCA should include reference to the scientific standards under Section 26(h), recognizing that the standards are influenced by the regulatory context under consideration. Operationally, the best available science could be interpreted as being receptive to new/different types of information that provide information relevant to a risk evaluation and that meet the data quality standards in Section 26(h). An example here would be the developing body of work on mode of action, a type of information that EPA has accepted in a few cases.
The definition of WOE provided in the preamble is inclusive and could, with further elaboration, form the basis to formulate descriptive, acceptable terms for the various facets of the program. We note that there is neither a reference to nor an attempt to meet the requirement to develop a WOE for exposure as stipulated under section 6(b)(4)(F)(v). EPA should consider not using “strength” as a qualifier in the above definition and to avoid confusion with the “strength of evidence” approach.
EPA is being asked to codify a process, with steps and components that are critical to an effective WOE. The preamble contains language leading in that direction. For example, there are several scenarios that call for WOE including prioritization, Section 6 risk evaluations, and Section 5 determinations. The information that supports risk-based conclusions for each of these scenarios differs in quantity and quality. We believe there is value in starting from a conceptual model, deriving characteristics for WOEs for each of these varied purposes in new TSCA, and then describing those characteristics to make clear how they differ. For example, the Work Plan Chemicals Method assigns weights to different components — this is an ingredient leading to an approach to WOE for this step. The WOEs for other scenarios (risk evaluation, new chemicals) may work differently and these differences can be described and discussed. A clear statement of EPA’s approach to WOE in conducting risk evaluations would provide a discernable starting point for the development of new guidance for carrying out Sections 4, 5, and 6 as required under Section 26(l)(1). One gap that must be filled by EPA in the procedural rule is development of a WOE for exposure assessment, an aspect that was not discussed in the proposal and for which there is no existing EPA model. The requirement to develop this WOE, as noted above, is made clear by Section 6(b)(4)(F)(v) which states that the risk evaluation shall “describe the weight of the scientific evidence for the identified hazard and exposure.”
EPA has stated that it intends to use existing guidance definitions and will update definitions and guidance as necessary. As discussed above and in the concluding paragraphs in the Commentary below, we believe that this is not sufficient.
EPA states that it welcomes public comments regarding the pros and cons of codifying these or other definitions and/or approaches for these or any other terms, however. EPA also encourages commenters to suggest alternative definitions that EPA should consider for codification in this procedural rule. EPA specifically requests comments on whether it should define “unreasonable risk” in the final rule. If so, acknowledging that new TSCA precludes consideration of costs and other nonrisk factors at this step, EPA asks what factors it should consider in making such a determination.
Margin of Exposure
EPA currently uses a margin-of-exposure (MOE) approach in risk characterization of TSCA risk assessments. EPA requests comment on the strengths and weaknesses of the MOE approach and whether there are other approaches (e.g., use of hazard indices, use of probabilistic risk assessment) that might better suit the TSCA Risk Evaluation Program. EPA specifically asks whether there are other approaches that provide quantifiable non-cancer risks.
While EPA has included a systematic review approach in the past, and states that it intends to continue to do so, EPA requests comment on the need for regulatory text prescribing a specific systematic review approach for hazard identification, including the appropriateness of elements that might be included or concerns about codifying an approach.
According to the proposed rule, EPA anticipates that some chemical substances prioritized for risk evaluation were manufactured by persons who possess unpublished information that could impact the chemical’s risk determination. For chemical substances prioritized for risk evaluation, EPA generally expects to exercise, as needed, among other authorities, its information-gathering authority pursuant to new TSCA Section 8(a) and (d), likely very early in the process. EPA specifically requests comment on approaches to using its information gathering authorities to assure that it has the most complete information to make its risk determination. For example, according to EPA, one option might be to incorporate its new TSCA Section 8(a) and 8(d) authority into the “Information and information sources” section of the proposed rule to allow EPA to require, by notice in the Federal Register, manufacturers with information subject to new TSCA Section 8(a)(2) and 8(d) to submit that information to EPA for use in a risk evaluation. EPA requests comment on this option and on any more effective alternative methods to exercise this authority within the rule to assure the completeness of the information relevant to the risk evaluation.
EPA states that it also anticipates the possibility that one manufacturer could request a risk evaluation while other manufacturers of the same chemical who have not joined in the request also possess relevant unpublished information. For manufacturer requests for risk evaluation, the burden is on the requester to include or reference all information that is necessary for EPA to conduct a risk evaluation. EPA acknowledges that it could use its data collection authority to access information, including unpublished studies, held by entities other than the requestor, but states that it “intends to deny requests for risk evaluation if the requester does not have access to the information necessary for risk evaluation.”
As discussed in both the OMB Bulletin and in the EPA Peer Review Handbook, there are specific exemption criteria for information that does not necessitate peer review, even if it might be considered to be influential or highly influential. A number of specific circumstances where peer review is not necessary are discussed in Section 3.3 of the EPA Peer Review Handbook. According to EPA, examples include information involving a health or safety issue where the EPA determines that the dissemination is time-sensitive or where an application of an adequately peer-reviewed work product does not depart significantly from its scientific or technical approach. In addition, EPA states that it expects that there will be individual circumstances where a chemical substance is found not to present an unreasonable risk or that findings are similar or the same as other jurisdictions (states or countries) that have reached similar conclusions based on the same information, such that EPA could determine that peer review is not necessary for that chemical risk evaluation.
EPA expects that many of the risk evaluations conducted under TSCA will necessitate peer review. In cases in which a chemical substance is determined to present an unreasonable risk, EPA must promptly move to manage the risk, a circumstance that would typically qualify the assessment as “influential scientific information” under current guidance and practice. EPA states that it also expects that some risk evaluations would also be highly influential scientific assessments, e.g., contain novel, controversial, or precedent-setting science with significant interagency interest. EPA also expects that peer review will be warranted in many cases where it determines a chemical substance does not present an unreasonable risk. According to EPA, aspects of the evaluation may qualify as influential scientific information or highly influential scientific assessment, and thus warrant peer review. Other circumstances where EPA may determine that peer review is warranted could include circumstances where there are existing private sector standards suggesting concern for a given chemical substance, where existing state assessments differ from the EPA evaluation, or where the public has expressed general concern about the chemical substances effects.
As required under new TSCA, chemical substances must be prioritized as either low or high. Those categorized as high are subject to a risk evaluation, and those determined to be low are not. EPA notes that the bar for prioritizing a chemical as a low priority as required under new TSCA “is fairly high.” As such, EPA expects that, as an increasing number of chemical risk evaluations are completed, those chemical substances that present risk to human health or the environment will be managed accordingly, leaving an increasing number of chemicals that do not present an unreasonable risk. EPA “questions whether all future risk evaluations warrant peer review.”
EPA specifically requests public comment on whether there are circumstances where conducting peer review may not be warranted. EPA asks what circumstances might qualify, and whether the regulatory text should be adjusted to require it to make a case-by-case determination of whether and to what extent, consistent with the EPA Peer Review Handbook, peer review is warranted for the chemical substance undergoing a risk evaluation. In all cases, the rule would require that this determination, and any peer review activities that are conducted, be documented for each chemical evaluation, starting with the scope document.
Reliance on Existing Guidance and Procedures for Conducting Risk Evaluations
EPA reiterates that it intends to take advantage of existing guidance, tools, and models that are relevant and available for use in conducting a risk evaluation under this program. Since each risk evaluation is based on the specific circumstances surrounding the chemical being assessed, EPA has not attempted to codify any specific guidance, method, or model. EPA states that it believes that this is necessary to ensure that there is flexibility to address potentially unique circumstances on a chemical basis. EPA requests comment on this approach, and where there is any existing guidance that may be of particular interest for consideration in conducting these risk evaluations. Additionally, EPA asks if the current guidance documents are sufficient and whether there are additional guidance documents that should be relevant but may not be on the lists available on EPA’s website at https://www.epa.gov/risk/risk-assessment-guidelines. Finally, EPA asks it should consider requiring that a list of appropriate guidance documents be included on a case-by-case basis as part of the scoping document that undergoes public review and comment. As discussed above and in our commentary, we question whether EPA’s preferred approach of relying on existing Agency risk assessment guidance is sufficient to meet the legal requirements under new TSCA.
EPA states that it is committed to ensuring there is interagency engagement and dialogue throughout its risk evaluation process, and has chosen not the limit the potential interagency collaboration by proposing to codify any particular process. EPA is concerned that imposing a single, pre-determined consultation step might lead to an overly bureaucratic process that could limit or complicate ongoing collaboration efforts, and so is not proposing to codify any particular process in this regulation. EPA requests specific public comment on whether codifying this collaboration at a specific point in the regulation is necessary, however.
EPA states in the proposal that it is not proposing a new method of risk evaluation, but is building upon existing and proven methodologies for evaluating risk. EPA recognizes that there are several key differences between previous chemical risk assessments conducted under old TSCA and the new risk evaluation process mandated by the new TSCA amendments that is the subject of these proposed procedural regulations. Differences highlighted by EPA include considerations of conditions of use, timelines, and determination of unreasonable risk. EPA states that this proposed rule and the procedures described apply to risk evaluations conducted under TSCA, and do not apply to risk evaluations conducted by EPA pursuant to other statutes or programs. This is consistent with earlier speculation that new TSCA would lead to design and implementation of a program based on Agency-wide risk assessment guidance modified as needed by TSCA statutory requirements.
Despite the logic and reasonableness of the approach, EPA has declined in this proposal to wrestle with and introduce terminology that could more readily contribute to defining a TSCA framework for risk evaluation. Instead, EPA responds that many of the terms used in the proposed rule are not novel concepts and are already in use, and their meaning is discussed extensively in existing EPA guidance. For example, extensive descriptions for the phrases “best available science,” “weight-of-the-evidence,” and “sufficiency of information” can be found in EPA’s Risk Characterization Handbook and other existing EPA guidance. EPA states that it believes further defining these and other terms in the proposed rule is “unnecessary and ultimately problematic.”
We disagree with this statement and with EPA’s approach in the proposal of essentially relying on generic risk assessment approaches and elements as described by EPA’s risk assessment guidance. While these approaches and elements can be generally applied by EPA across its various program offices, differences between and specific requirements in each law can require the use of different approaches and elements. This is clearly the challenge that EPA is facing in implementing risk evaluations in new TSCA that include new and different risk assessment components and new science accountability requirements such as those in Section 26(h) and (i). This is also the case across new TSCA where Sections 4, 5, and 6 incorporate science-based decisions in each program, all of which are expected to use the same terminology, but under a different regulatory context. For example, there are several scenarios that call for WOE including prioritization, Section 6 risk evaluations (explicitly including the exposure assessment component), and Section 5 determinations. The information — on hazard and exposure — that supports science- and risk-based conclusions for each of these scenarios differs in quantity and quality.
We believe there is value in starting from a conceptual model, deriving characteristics for WOEs for each of these varied purposes in new TSCA, and then describing those characteristics to make clear how they differ. As an example, the Work Plan prioritization method assigns weights to different components — this is an ingredient leading to an approach to WOE for this step. The WOEs for other scenarios (testing, risk evaluation, new chemicals) will in many cases work differently and these differences can be described and discussed. Best available science could be defined starting with the data quality principles embedded in the Section 26(h) standards — common to all scenarios – and then given an operational component to differentiate the different regulatory environments under TSCA. The proposed risk evaluation procedural rule could have contributed a starting point for the development of this understanding, but largely failed to do so due to its reliance on EPA’s generic risk assessment guidance as providing the substance needed for the proposed rule. We believe that this existing guidance should merely have served as the launching pad for development of an approach that is elaborated to meet the specific requirements under new TSCA, initially for risk evaluation, and then subsequently for science-based decisions under Section 4, 5, and 6 as contemplated by Section 26(l)(1).
Overall, and as was seen with the proposed prioritization rule, the proposed rule for risk evaluation falls short of providing meaningful insight into the process that EPA will use in the future, except to indicate generically that it will use the same approaches, guidance, and policies as in the past. We believe that Congress intended for EPA to develop an approach and elaborate upon a procedural rule for implementing risk evaluations that went beyond the existing EPA guidance to explain how EPA would meet the science requirements and implement the new elements contained in the new law. If Congress thought that the status quo was adequate, silence on these topics would have made that point clear, but as discussed above that is not what happened.