On September 23, 2025, the U.S. Environmental Protection Agency (EPA) proposed highly anticipated amendments to the procedural framework rule for conducting existing chemical risk evaluations under the Toxic Substances Control Act (TSCA). 90 Fed. Reg. 45690. When conducting an existing chemical risk evaluation under TSCA, EPA must determine whether a chemical substance presents an unreasonable risk of injury to health or the environment, without considering costs or nonrisk factors, including unreasonable risk to a potentially exposed or susceptible subpopulation identified as relevant to the risk evaluation, under the conditions of use (COU). EPA states that it "proposes to rescind or revise certain 2024 amendments to the procedural framework rule to effectuate the best reading of the statute and ensure that the procedural framework rule does not impede the timely completion of risk evaluations or impair the effective and efficient protection of health and the environment." Comments on the proposed rule are due November 7, 2025. EPA states that under the Paperwork Reduction Act, "comments on the information collection provisions are best assured of consideration if the Office of Management and Budget (OMB) receives a copy of your comments on or before October 23, 2025."
As reported in its September 22, 2025, press release, EPA states that the proposed rule includes the following proposed amendments to address targeted changes to EPA's process for conducting TSCA risk evaluations made in the 2024 risk evaluation rule:
- A requirement for EPA to make a determination of unreasonable risk for each of the COUs within the scope of the chemical's risk evaluation as dictated by Congress in TSCA, as amended by the Lautenberg Chemical Safety for the 21st Century Act (Lautenberg Act), instead of a single risk determination on the chemical substance as a whole.
- Clarifications as to how EPA will consider occupational exposure controls such as personal protective equipment (PPE) and industrial controls when conducting risk evaluations and making risk determinations.
- Clarifications regarding EPA's discretionary authority to determine which COUs, exposure routes, and exposure pathways it will consider in a risk evaluation.
- Revisions to certain regulatory definitions to ensure consistency with Executive Order (EO) 14303 Restoring Gold Standard Science and to ensure transparency and accountability in conducting risk evaluations.
- Revisions to the procedures and requirements EPA would follow when revising or supplementing risk evaluation documents to enable EPA better to meet the statutory deadlines to assess and manage risk.
- Adjustments to the process and information collection obligations for manufacturers (including importers) for requesting an Agency-conducted TSCA risk evaluation.
More information on the 2024 final rule is available in our May 14, 2024, memorandum. Before the Biden EPA issued the final rule, on June 30, 2021, it announced policy changes surrounding TSCA risk evaluations by the previous Administration and the path forward for the first ten chemicals undergoing risk evaluation. As reported in our July 1, 2021, memorandum, the policy changes included:
- Whole Chemical Approach: In the final risk evaluations for the first ten chemicals, EPA made separate unreasonable risk determinations for each COU for each chemical. EPA stated that it would withdraw the previously issued orders for those COUs for which no unreasonable risk was found, issue revised, single unreasonable risk determinations for these chemicals as a "whole substance," and seek public comment;
- Use of PPE: Unlike the previous Administration, which generally assumed in the final risk evaluations that PPE was always provided to and used by workers, the Biden EPA stated that it was revisiting this assumption; and
- Expanding Consideration of Exposure Pathways and Fenceline Community Exposure Screening Level Approach: The previous Administration chose not to consider air, water, and soil contamination because these exposures are already regulated under other environmental laws such as the Clean Air Act (CAA), Safe Drinking Water Act (SDWA), and Clean Water Act (CWA).
EPA proposes to amend the regulations at 40 C.F.R. Section 702.31 (general provisions) so that the changes to the procedures as part of the rulemaking would be applied to all risk evaluations initiated on or after the date of the final rule (May 3, 2024) and would be applied to risk evaluations that are in process as of the date of the final rule, but not yet final, to the extent practicable. EPA states in the proposed rule that it "is not currently aware of any significant reliance interests in the 2024 amendments to the procedural framework rule at issue in this proposal, which remain fairly recent and apply almost exclusively to internal Agency process." EPA seeks comment on the proposed changes, "including on whether stakeholders have any significant reliance interests on the 2024 amendments at issue and, if so, how such interests should be accounted for in any final action."
Commentary
The proposed rule is well crafted and the changes to the risk evaluation framework are unsurprising. The legal underpinnings of the single determination/whole chemical approach have been vigorously contested since the policy was implemented. We agree with EPA that it can make separate determinations for some COUs of a substance. The rejoinder that doing so ignores the risk of exposure from other pathways assumes that the risk evaluation itself ignored such risks. EPA can (and should) review the risks from individual exposures (e.g., workplace, drinking water, consumer) and determine if any one or some combination of exposures leads to an unreasonable risk. If EPA finds that cumulative exposure from drinking water, fugitive air, and consumer exposures are not an unreasonable risk, but that workplace exposure is an unreasonable risk (whether cumulative with other exposures or not), EPA can still make a determination that general population exposures are not an unreasonable risk and a separate determination that workplace exposures may be an unreasonable risk. Separate determinations are based on the same evaluation of cumulative risk. EPA's regulatory action would be focused on protecting workers because there is no need to issue restrictions to protect the general population.
We also agree that it is appropriate for EPA to assess workplace exposures while assuming compliance with U.S. Occupational Safety and Health Administration (OSHA) regulations. Detractors of this approach note that OSHA Permissible Exposure Limits (PEL) are outdated and not sufficiently protective. That may be true, but that is a question of fact for the assessment. EPA can question legitimately whether compliance with OSHA requirements is sufficiently protective without assuming that employers violate OSHA.
A related — and critical — question is if EPA has extensive data documenting very low workplace exposures (as is the case in some of the phthalate risk evaluations), should EPA still assume high-end exposures based on its worst-case models? In the risk evaluation for diisobutyl phthalate (DIBP), EPA only identified health risk from high-end unprotected workplace exposures during spray application of coatings and adhesives in an automotive refinish shop, but EPA offered no evidence that this is a COU other than speculation that it might happen. If the only risk identified is a high-end, unprotected exposure under a hypothetical use scenario, does the substance really meet the standard that it "presents" an unreasonable risk? If any of EPA's stacked worst-case assumptions is not present, the unreasonable risk vanishes: if DIBP is not used in automotive refinish operations, or workers wear basic respiratory protection, or if the exposures are not at the 95th percentile, there is not unreasonable risk. EPA has to assume a great deal to get there; under the whole chemical and no PPE assumptions, DIBP is an unreasonable risk. Under a COU-specific approach with basic assumptions about OSHA compliance, DIBP is not an unreasonable risk (or is only a risk when sprayed over a certain exposure limit).
Another expected change is EPA's assumptions about exposures that result from pathways regulated by other authorities, such as the CAA and SDWA. As with OSHA protection, our view is that EPA should evaluate exposures from those pathways, assuming compliance with those programs. It may be the case that a substance undergoing risk evaluation is not restricted by another statute. In that case, EPA would reasonably assume no control.
The proposal also strengthens the view that Congress did not intend for TSCA to be the preeminent protective statute. If EPA finds a COU that is an unreasonable risk and that another federal authority can effectively and efficiently protect against that risk, EPA should consider seriously whether that other authority is more appropriate for protecting against the risk identified. For example, EPA identified workplace inhalation risk for methylene chloride and imposed a 2 parts per million (ppm) existing chemical exposure limit (ECEL — the TSCA equivalent of an OSHA PEL). But because TSCA does not apply to food, drugs, and cosmetics, and components thereof, the TSCA ECEL would not apply in a pharmaceutical facility. If EPA had instead worked with OSHA to lower the PEL to 2 ppm, the enforceable limit would apply to pharmaceutical facilities. Another example is 1,4-dioxane — a pollutant that is present in drinking water in some locations. Its presence is largely due to past 1,4-dioxane releases, not prospective activity. In that case, a ban (or some other limit) of 1,4-dioxane under TSCA will have little effect on drinking water exposure from past pollution. A more effective approach is to employ the SDWA to limit drinking water exposure to 1,4-dioxane regardless of the source. EPA requests specifically comments on this issue.
We are also pleased to see that requirements to use the best available science and weight of scientific evidence are being written back into the regulations. EPA's explanation in the 2024 rule that both were statutory requirements, so it was not necessary to embed the requirements in the regulations rang hollow. The discussion in the rule also mentions assessments performed by other competent authorities, including other EPA offices and other countries. EPA requests comments on whether EPA should or must consider such assessments. Our view is that the regulation should require EPA to review such assessments as part of the weight of scientific evidence. That does not mean that any particular assessment would be dispositive of an outcome, but it should not be ignored. We are especially supportive of EPA reviewing the quality of the scientific evidence prior to that evidence's use in a risk evaluation. Judging the quality of the science should not depend on how that science supports or refutes a particular conclusion about the risk of a substance.
EPA seeks comment on whether EPA should define "reasonably foreseen." This is a crucial question on which stakeholders are urged to comment. We have been urging EPA to wrestle with this term since June 2016. From about 2017 to 2021, EPA took the position that "Reasonably foreseen conditions of use will not be based on hypotheticals or conjecture." (See footnote 1 of the determination document for P-16-0510.) EPA no longer includes that explanation in its determination documents. Rather, the New Chemicals Program views anything that it has not reviewed as being foreseeable. There is significant disagreement in the TSCA stakeholder community about the term and, absent congressional action to define the term, our view is that EPA can and should define the term.
Other issues that EPA raises and upon which EPA requests comments are de minimis quantities and the question of byproducts and impurities. Our view is that de minimis quantities may be justified if EPA can confidently conclude that exposures, even combined exposures, are not likely to rise to a level of concern. Such a consideration will be necessarily substance-specific, but establishing a de minimis threshold both for the risk evaluation and in a risk management rule are potentially justifiable.
Whether and how EPA assesses a substance as a byproduct or impurity will also depend on the circumstances. A highly hazardous substance that may be present as an impurity in many substances is likely more appropriate for EPA to assess in a risk evaluation for the impurity substance, rather than the many substances in which it may be found. Another highly hazardous substance that is likely only present in a small number of products can be efficiently assessed in a risk evaluation for those few products. The same logic applies to byproducts. Byproducts that are formed during the manufacture or processing of many substances should probably be addressed during the risk evaluation for the byproduct itself, whereas a byproduct that is only formed during the manufacture of a single substance can be efficiently assessed during the risk evaluation for that substance.
We also support EPA's proposal to codify its procedures if it finds it must change either a final scope or final risk evaluation. This rule proposes to formalize steps for EPA to highlight and conduct notice-and-comment, if necessary, to ensure that the public is aware of changes to the final scope or final risk evaluation documents. Risk evaluation documents are long and complex; this way, EPA must call attention to changes it makes to its final documents rather than forcing stakeholders to review de novo or do tedious side-by-side reviews.
This proposal goes a long way in rightsizing the risk framework and we encourage stakeholders to comment on key issues.
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.
