TSCA “Conditions of Use” Require Use-by-Use Risk Determinations and Real-World PPE Assumptions; Precautionary “Whole-Chemical” Bans Demand Substantial Evidence

East Fork Enterprises v. EPA (5th Cir. Sept. 15, 2026) — Jones, J.

I. Introduction

Parties. Industry Petitioners (East Fork Enterprises, Epic Paint Company, and the American Chemistry Council) challenged EPA’s May 2024 TSCA rule regulating methylene chloride (“MC”), a widely used solvent. The Sierra Club filed a separate petition asserting the rule was not protective enough for certain vulnerable groups.

Regulatory background. After Congress amended TSCA in 2016 to require modernized risk evaluations and risk management, EPA issued a June 2020 MC risk evaluation and later (under a new Administration) a November 2022 “revised unreasonable risk determination.” That revision made two pivotal policy moves: (1) it shifted from use-by-use determinations to a single “whole chemical” risk determination, and (2) it assumed workers do not use personal protective equipment (“PPE”). EPA then promulgated the May 2024 final rule (the “MC Rule”), prohibiting 40 of 53 conditions of use and imposing strict exposure limits for 13 remaining uses through a Workplace Chemical Protection Program (WCPP).

Key issues. The Fifth Circuit confronted whether EPA complied with TSCA’s statutory structure and evidentiary demands when it: (a) abandoned condition-of-use-specific risk determinations; (b) ignored PPE as a “condition of use”; (c) set exposure limits using selective science and highly conservative uncertainty factors; and (d) effectively banned most uses based on uncertainty and without properly considering statutory constraints on “to the extent necessary” regulation and substitutes.

II. Summary of the Opinion

Core holdings. The court held EPA committed legal error by adopting a “whole chemical” risk determination rather than making determinations tied to each “condition of use,” and by excluding PPE from the risk evaluation despite TSCA’s “conditions of use” definition. On the merits of the science, the court found EPA’s acute and chronic exposure limits unsupported by substantial evidence under TSCA, faulting EPA’s selective reliance on limited studies and failure to grapple with contradictory human data. Finally, the court criticized EPA’s near-ban as inconsistent with TSCA’s “to the extent necessary” limitation and as improperly driven by uncertainty and burden-shifting to regulated parties.

III. Analysis

A. Precedents Cited (and How They Shaped the Decision)

1) TSCA’s evidentiary rigor and the Fifth Circuit’s asbestos benchmark: Corrosion Proof Fittings v. EPA

The court repeatedly anchored its review in Corrosion Proof Fittings v. EPA, 947 F.2d 1201 (5th Cir. 1991), the Fifth Circuit’s seminal TSCA decision vacating EPA’s 1989 asbestos ban for failure to satisfy TSCA’s demanding evidentiary and analytic requirements (including consideration of less burdensome options). Here, Corrosion Proof Fittings supplied three controlling themes:

• Substantial evidence under TSCA is “more rigorous” than APA arbitrariness review and requires consideration of record evidence that detracts from EPA’s conclusions.
• TSCA is not a “zero-risk” statute; “unreasonable risk” implies toleration of some risk (“No More, No Less”).
• EPA must not justify sweeping prohibitions with conjecture (“musings and conjecture” are not substantial evidence).

2) “Best available” data as an anti-speculation mandate: Bennett v. Spear

By citing Bennett v. Spear, 520 U.S. 154 (1997), the court framed TSCA’s “best available science” requirements as a statutory command against speculative administration. The court used Bennett’s warning about “speculation or surmise” to emphasize that TSCA’s scientific and procedural guardrails constrain EPA’s discretion even when pursuing environmental objectives.

3) Post-Loper Bright statutory interpretation and the “prudential baseline” of “unreasonable”: Loper Bright Enters. v. Raimondo, Groff v. DeJoy, and Texas v. EPA

The court treated “unreasonable risk” as a legal standard rather than a purely technical finding insulated by agency expertise. Citing Loper Bright Enters. v. Raimondo, 603 U.S. 369 (2024), it stressed that courts independently interpret statutory language. It analogized to Groff v. DeJoy, 600 U.S. 447 (2023) (construing “undue hardship”), and to Texas v. EPA, 137 F.4th 353 (5th Cir. 2024), where the Fifth Circuit interpreted “unclassifiable” and vacated an EPA regulation due to problematic scientific analysis.

This triangulation supports a significant doctrinal move in the opinion: EPA’s scientific modeling does not define “unreasonable risk”; it supplies inputs. The statute supplies the baseline, and courts review whether EPA’s findings support the statutory conclusion.

4) Rejecting an implied precautionary principle: Me. Lobstermen's Ass'n v. Nat'l Marine Fisheries Serv. and Clean Air Act comparisons

The court relied on Me. Lobstermen's Ass'n v. Nat'l Marine Fisheries Serv., 70 F.4th 582 (D.C. Cir. 2023), to reject the notion that courts should infer a precautionary principle from unclear text. The opinion contrasted TSCA’s “unreasonable risk” with statutes that explicitly require precaution (e.g., the Clean Air Act’s “adequate margin of safety,” discussed with Lead Indus. Ass'n v. EPA, 647 F.2d 1130 (D.C. Cir. 1980)).

This contrast mattered twice: first, to criticize EPA’s conservative uncertainty stacking as functionally “precautionary,” and second, to treat “to the extent necessary” as a limiting principle that forbids ratcheting regulation toward “lowest possible risk” without statutory authorization.

5) The “single study” problem and substantial evidence: Gulf S. Insulation v. U.S. Consumer Prod. Safety Comm'n, Johnson v. Arkema, Inc., and Allen v. Pa. Eng'g Corp.

The court’s most concrete application of substantial evidence review came through scientific-evidence precedents. It invoked Gulf S. Insulation v. U.S. Consumer Prod. Safety Comm'n, 701 F.2d 1137 (5th Cir. 1983), to condemn “exclusive reliance” on a single animal study where human epidemiology exists and to warn that small studies have large margins of error. The court reinforced skepticism about translating animal studies to humans through Johnson v. Arkema, Inc., 685 F.3d 452 (5th Cir. 2012), and Allen v. Pa. Eng'g Corp., 102 F.3d 194 (5th Cir. 1996).

These cases were used not to announce a categorical bar on animal data, but to require EPA to account for contrary human data and explain why it is discounted, especially under TSCA’s “more searching” substantial evidence regime.

6) Reasoned decisionmaking and agency reversals: BNSF Ry. Co. v. Fed. R.R. Admin. and FCC v. Fox Television Stations, Inc.

In invalidating EPA’s pivot away from PPE assumptions, the court relied on BNSF Ry. Co. v. Fed. R.R. Admin., 105 F.4th 691 (5th Cir. 2024), quoting FCC v. Fox Television Stations, Inc., 556 U.S. 502 (2009), for the requirement of a “reasoned explanation” when an agency changes policy, especially when it disregards facts and circumstances underlying prior policy.

7) Duty to respond to significant comments: Perez v. Mortg. Bankers Ass'n and Fifth Circuit SEC review

The court criticized EPA for failing to engage expert critique (Dr. Borak) and epidemiological evidence, citing Perez v. Mortg. Bankers Ass'n, 575 U.S. 92 (2015), and Chamber of Comm. of U.S. v. SEC, 85 F.4th 760 (5th Cir. 2023), as “significant comments” that require reasoned response.

8) Agencies must follow their own rules: DOJ v. Fed. Lab. Rels. Auth. and interpretive context: Wyo. Outdoor Council v. U.S. Forest Serv.

The “whole chemical” holding was bolstered by the principle that agencies must abide by their regulations, citing DOJ v. Fed. Lab. Rels. Auth., 991 F.2d 285 (5th Cir. 1993). The court also cited Wyo. Outdoor Council v. U.S. Forest Serv., 165 F.3d 43 (D.C. Cir. 1999), for the preamble as evidence of contemporaneous understanding— but it ultimately emphasized that EPA’s operative procedural rule text demanded use-by-use determinations.

9) Standing and attenuated causal chains: TransUnion LLC v. Ramirez, Clapper v. Amnesty Int'l USA, Summers v. Earth Island Inst., and related cases

In rejecting Sierra Club’s ozone-depletion challenge for lack of standing, the court applied the familiar Article III framework from TransUnion LLC v. Ramirez, 594 U.S. 413 (2021), and rejected a speculative causal chain under Clapper v. Amnesty Int'l USA, 568 U.S. 398 (2013). It rejected “procedural standing” untethered to a likely substantive change, citing Summers v. Earth Island Inst., 555 U.S. 488 (2009), and applied Fifth Circuit procedural-injury doctrine from Nat'l Infusion Ctr. Ass'n v. Becerra, 116 F.4th 488 (5th Cir. 2024), and redressability logic associated with Massachusetts v. E.P.A., 549 U.S. 497 (2007).

B. Legal Reasoning

1) TSCA forbids “whole chemical” unreasonable-risk determinations

The opinion’s most explicit rule is textual: TSCA’s repeated focus on “conditions of use” requires risk determinations tied to the discrete circumstances in which a chemical is “intended, known, or reasonably foreseen” to be used. The court read 15 U.S.C. § 2605(a) and § 2605(b)(4)(A) together: risk is “presented” by activities/conditions of use involving the chemical, not by the chemical in the abstract.

The court reinforced that statutory reading with EPA’s own then-operative procedural rule: Procedures for Chemical Substance Risk Evaluations Under the Amended Toxic Substances Control Act, 82 Fed. Reg. 33726, 33752 (July 20, 2017), which required EPA to determine unreasonable risk “under each condition of uses within the scope of the risk evaluation.” EPA’s later “whole-chemical” interpretation was deemed inconsistent with both TSCA and EPA’s own rule.

2) PPE is a “condition of use” and must be included when known or reasonably foreseen

The court treated PPE use as an archetypal “circumstance” of use under 15 U.S.C. § 2602(4): it directly affects intensity, frequency, and duration of exposure. EPA’s 2020 evaluation assumed OSHA compliance and routine PPE use; the 2022 revised determination assumed the opposite (no PPE).

The court held the PPE reversal unlawful and unsupported: it ignored TSCA’s “conditions of use” definition, lacked substantial evidence, and constituted an unexplained departure from prior findings. In doing so, the opinion implicitly ties TSCA’s “reasonably foreseen” condition-of-use concept to ordinary expectations of compliance with other binding workplace protections (here, OSHA’s MC standard at 29 C.F.R. § 1910.1052).

3) “Unreasonable risk” is not “any risk,” and TSCA contains no implied precautionary principle

The court drew a sharp line between TSCA and statutes that demand near-zero risk. It criticized EPA for using uncertainty factors and points of departure in a way that effectively chased “NO” adverse effect and resembled precautionary regimes (contrasting, for example, Clean Air Act jurisprudence and the Second Circuit’s pesticide context in Nat. Res. Def. Council v. EPA, 658 F.3d 200 (2d Cir. 2011)).

Legally, the court characterized “unreasonable risk” as a prudential statutory threshold—a legal conclusion derived from the statute’s guardrails, not whatever number results from stacking conservative assumptions.

4) Substantial evidence review drove the scientific critiques

Applying TSCA’s more demanding substantial-evidence standard, the court found EPA’s acute and chronic limits untenable:

• Acute limit (16 ppm/15 minutes): EPA relied heavily on the “Putz study” (12 humans; modest, temporary peripheral-vision impairment), discounted the “Winneke study,” and did not meaningfully reconcile its approach with decades of OSHA’s 125 ppm acute limit and “immediately dangerous” concentration contexts. The court treated EPA’s exclusive reliance and conservative extrapolations as failing to account for record evidence that detracted from EPA’s conclusion.
• Chronic limit (2 ppm/8 hours): EPA relied on the “Nitschke study” (rats; liver irregularities at 500 ppm; NOAEL at 200 ppm), while discounting “acceptable” human epidemiology and failing to respond to Dr. Borak’s critique. The court analogized directly to Gulf S. Insulation: it is not good science (and thus not substantial evidence under the governing statute) to rely on a single animal experiment while ignoring relevant human studies.

5) “To the extent necessary” limits bans; uncertainty and burden-shifting are impermissible

Even setting aside the risk-determination defects, the court faulted EPA’s risk-management choice—near prohibition rather than alignment with or modest adjustment to OSHA— as inconsistent with TSCA’s requirement to regulate “to the extent necessary so that the chemical substance or mixture no longer presents such risk.”

Two limiting principles are prominent:

• No bans based on uncertainty alone: EPA cannot prohibit a chemical because it is “uncertain” whether industry will meet EPA’s chosen limit; TSCA authorizes regulation to eliminate identified “unreasonable risk,” not to eliminate uncertainty.
• No burden shift to regulated parties: EPA cannot treat a lack of comments demonstrating compliance feasibility as support for prohibition; TSCA places the burden on EPA to justify both the risk determination and the necessity of the chosen remedy.

The opinion also criticized EPA’s “alternatives” analysis under 15 U.S.C. § 2605(c)(2)(C) as overly narrow (function-only) and insufficiently attentive to economic and technical feasibility—especially given TSCA’s express instruction to factor in benefits and reasonably ascertainable economic consequences during the risk-management phase.

C. Impact

The decision is poised to be a major constraint on aggressive TSCA § 6(a) rulemakings in three ways:

• Structural constraint: EPA must tie unreasonable-risk determinations to discrete “conditions of use,” limiting attempts to regulate “as a whole chemical” where some uses are low-risk or readily controlled.
• Real-world workplace controls matter at the risk stage: PPE (and by implication, other reasonably foreseen engineering/administrative controls) must be integrated into risk evaluations when supported by record evidence and regulatory context.
• Evidence discipline: Under TSCA’s substantial evidence review, EPA must grapple with conflicting human data and cannot rely on a single preferred study—especially animal-only evidence—to drive exceptionally low exposure thresholds.

On remand, EPA will likely need to (i) restore use-by-use risk determinations; (ii) rebuild exposure assessments reflecting reasonably foreseen PPE and compliance; and (iii) re-justify any prohibitions with a record-supported explanation that the chosen measures are “to the extent necessary,” including a more rigorous substitutes and economic-consequences analysis for any use effectively eliminated.

IV. Complex Concepts Simplified

• “Conditions of use” (TSCA): The real-life ways a chemical is made, used, and disposed of—including the circumstances of use that are “intended, known, or reasonably foreseen.” The court held this concept requires EPA to analyze risk use-by-use, not at a “whole-chemical” level.
• Risk evaluation vs. risk management: TSCA separates (1) determining whether risk is “unreasonable” (without considering cost), from (2) selecting controls to eliminate unreasonable risk (where EPA must consider benefits, economic consequences, and alternatives).
• Substantial evidence (TSCA): A tougher standard than ordinary APA arbitrariness review; EPA must consider the whole record, including evidence cutting against its position, and reach a conclusion a reasonable mind would accept as adequate.
• Point of departure (POD), LOAEL, NOAEL: A POD is the study-based exposure level used as a starting point. LOAEL is the lowest level where an adverse effect is observed; NOAEL is the highest level where no adverse effect is observed. EPA favored NOAEL-like conservatism; the court criticized doing so in a way that effectively imposed near-zero risk.
• Margin of exposure (MOE) and uncertainty factors: MOE compares a POD to actual exposure. EPA then applies “uncertainty factors” to account for unknowns (e.g., differences among humans). The court viewed EPA’s stacking of conservative assumptions—paired with selective study use—as a move toward a precautionary regime TSCA does not authorize.
• PPE (personal protective equipment): Respirators, gloves, protective clothing, and eye protection. The court held EPA cannot pretend PPE does not exist at the risk-determination stage when it is reasonably foreseen, especially in OSHA-regulated workplaces.

V. Conclusion

East Fork Enterprises v. EPA re-centers TSCA regulation on statutory structure, evidentiary rigor, and real-world conditions of use. The Fifth Circuit’s key takeaways are that EPA (1) must evaluate unreasonable risk by “conditions of use” rather than by “whole chemical” fiat; (2) must account for reasonably foreseen workplace protections like PPE in its risk evaluation; (3) cannot sustain exceptionally stringent exposure limits on selective or one-study science under TSCA’s substantial-evidence standard; and (4) must tailor risk management “to the extent necessary,” without banning based on uncertainty or shifting the burden of proof to industry. Sierra Club’s bid for stricter protections failed because TSCA does not require EPA to regulate “elevated” rather than “unreasonable” risk, and because the ozone-depletion challenge lacked Article III standing.