TSCA fluoride lawsuits: two legal routes against the EPA

Federal fluoride litigation turns on a legal split that is easy to miss: a challenge to a final EPA decision usually asks a court to review the agency’s existing record, while a petition under Section 21 of the Toxic Substances Control Act can lead to a different kind of case, with de novo judicial review. The distinction shapes what evidence a court may consider and what plaintiffs must establish.

The numbers at the center of the dispute do not settle the legal question. The Department of Health and Human Services recommends community water fluoridation at 0.7 milligrams per liter. The EPA’s enforceable maximum contaminant level for fluoride is 4.0 mg/L, with a separate, non-enforceable secondary guideline of 2.0 mg/L for cosmetic and aesthetic effects. Those standards serve different functions. Treating them as interchangeable is a shortcut, and in litigation shortcuts become liabilities.

Two routes, two different records

A federal court challenge to water fluoridation can arrive through more than one procedural door. The route matters because it determines what the court is being asked to review.

In ordinary administrative-record review, the challenge targets a final agency action. The court generally examines the record the agency compiled when it made that decision. The fight often centers on whether the agency followed the governing law and adequately supported its action using the material before it. The record is the battlefield; arguments about material outside it may face substantial limits.

Section 21 of TSCA sets up another route. Under 15 U.S.C. § 2620, citizens may petition the EPA to begin rulemaking to restrict a chemical they believe presents an unreasonable risk. If the EPA denies the petition, the petitioners can bring a civil action in federal district court. The statute provides for de novo review.

That phrase carries weight. De novo review does not confine the court to asking whether the EPA’s denial was justified by the administrative record in the same way as a conventional record-review case. It allows the court to consider the statutory question anew. In the fluoride litigation, the plaintiffs used that feature to put the risk question before a district court rather than leaving the dispute solely inside the boundaries of the EPA’s petition-denial record.

That is the core comparison behind any TSCA fluoride lawsuit EPA litigation strategy. One route tests an agency decision as an administrative decision. The other uses a statutory citizen-petition mechanism that can move the dispute into a broader judicial examination. Neither route guarantees a win. They give litigants different procedural terrain.

The route determines what the court can examine. That is not a technical footnote; it can decide what the case is about.

Food & Water Watch v. EPA: the Section 21 test case

The central case began with a petition, not with a direct challenge to a national fluoridation rule. In 2016, a coalition including Food & Water Watch and the Fluoride Action Network petitioned the EPA under Section 21. They asked the agency to prohibit adding fluoridation chemicals to public drinking water supplies.

The EPA denied the petition in 2017. The groups then sued in federal district court in California. The case, Food & Water Watch v. EPA, became a test of how far Section 21’s de novo review can reach when petitioners challenge the agency’s position on a chemical risk.

On September 24, 2024, Senior U.S. District Judge Edward Chen ruled that water fluoridation at 0.7 mg/L posed an unreasonable risk of neurodevelopmental injury to children under TSCA Section 6(a). He ordered the EPA to initiate regulatory action.

That ruling was consequential. It was also narrower than some accounts suggested. The court did not enact a nationwide ban on fluoridation. It ordered the EPA to begin regulatory action under TSCA. The difference matters: an order directing agency action is not itself a completed rule, and it does not erase the separate drinking-water standards already in place.

The legal theory also matters. The case did not establish that every route for challenging EPA action works like a Section 21 suit. Its procedural force came from the statute’s citizen-petition provision and the de novo review it supplies. A litigant challenging a different final agency action may face a different review framework, a different record, and a different set of deadlines and legal questions.

For readers tracking the epa citizen petition legal process, the sequence is concrete:

1. A citizen group submits a Section 21 petition asking EPA to initiate rulemaking.

2. EPA grants or denies the petition.

3. If EPA denies it, petitioners may file a civil action in federal district court.

4. The district court conducts de novo review under Section 21 and decides whether the statutory basis for relief has been shown.

5. Any resulting order remains subject to appeal and further proceedings.

That sequence offers a route into court. It does not make the outcome automatic. The petition must identify the requested agency action and the claimed risk; the resulting litigation must survive procedural scrutiny; and a favorable ruling can still be appealed.

The Ninth Circuit’s reversal: procedure takes the wheel

The 2024 district court decision did not remain the final word. The EPA filed a notice of appeal on January 17, 2025. On May 21, 2026, the Ninth Circuit vacated Judge Chen’s ruling and remanded the case.

The appellate court’s stated concern was procedural. It found that the district court violated the party-presentation principle by holding the case in abeyance for more than a year while awaiting a National Toxicology Program monograph that had not been submitted by the parties, rather than ruling on the record from the initial bench trial.

That is a sharp reversal in the case’s posture. The appellate ruling did not resolve the underlying scientific question by declaring fluoridation safe or unsafe. It vacated the district court judgment on procedural grounds and sent the case back. The distinction is central to reading the ruling accurately: the court addressed how the district court handled the litigation, not a final nationwide policy on fluoridation.

The party-presentation principle places responsibility on litigants to present the issues and material they want the court to consider. A judge may manage a case, but the court cannot casually substitute its own investigative agenda for the parties’ presentation. In this dispute, waiting for a document that had not been placed before the court became the procedural fault line.

For a TSCA fluoride case, the lesson is not that judges must ignore science outside a trial record. It is that the way scientific material enters the case matters. Parties build the evidentiary record. Courts decide within the bounds of the case presented. A late-arriving report, even one that may be relevant to the broader debate, cannot simply be treated as though litigants had introduced and tested it.

The remand leaves the next district-court outcome unresolved. The available facts do not establish whether the court will reach a new decision on the original trial record, what further proceedings it may order, or what regulatory action EPA would ultimately take if the district-court ruling were reinstated. Anyone reporting a settled end to the case is getting ahead of the docket.

The regulatory gap: 0.7, 2.0 and 4.0 mg/L

The figures recur in public debate, often stripped of their regulatory context.

The Department of Health and Human Services recommendation for community fluoridation is 0.7 mg/L. The EPA’s enforceable maximum contaminant level is 4.0 mg/L under the Safe Drinking Water Act. The EPA also maintains a secondary guideline of 2.0 mg/L, which is non-enforceable and concerns cosmetic or aesthetic effects.

These are not three competing versions of one standard. They do different jobs. The 0.7 mg/L figure is a recommended fluoridation level. The 4.0 mg/L figure is an enforceable ceiling under a separate federal drinking-water framework. The 2.0 mg/L secondary guideline is not enforceable and addresses a different category of effects.

That gap creates a regulatory argument, but it does not resolve it. A maximum contaminant level does not, on its own, answer every question about a recommended operating level. Nor does a recommendation automatically function as a binding federal limit. The TSCA litigation asked a distinct question: whether the challenged use presented an unreasonable risk under TSCA and whether regulatory action was required under that statute.

This is where legal framing can get slippery. A party may point to the difference between a recommended concentration and an enforceable ceiling to argue that regulators have left a gap. Another may argue that the existing standards belong to separate statutory schemes and should not be collapsed into one. The court’s task is to apply the statute before it, not to turn three figures into a single all-purpose safety verdict.

For a reader comparing an administrative petition vs. federal court, the practical distinction is that the first seeks agency action; the second asks a court to review the agency’s response through the applicable statutory pathway. Section 21 gave these petitioners a route to de novo review after EPA denied their request. It did not convert the HHS recommendation into an EPA maximum contaminant level, or automatically repeal the existing drinking-water standard.

What the case means for future TSCA challenges

The Ninth Circuit’s vacatur has changed the immediate legal position, not erased the Section 21 mechanism. The statute still provides for citizen petitions and de novo judicial review after an EPA denial. But this case now carries a procedural warning label: even a broad review pathway does not free a district court from the ordinary rules governing how a case is presented and decided.

Future plaintiffs considering a federal court challenge to water fluoridation will have to make several choices before the merits take center stage:

• Identify the agency action. A challenge to a final EPA action and a suit following denial of a Section 21 petition do not start from the same procedural premise.
• Build the record the case needs. Under ordinary administrative review, the agency record is central. Under Section 21, de novo review broadens the judicial inquiry, but parties still need to present their evidence and arguments properly.
• Keep the requested remedy precise. Asking EPA to begin rulemaking is not identical to asking a court to impose a particular nationwide restriction. The remedy must follow the statute and the posture of the case.
• Treat appeals as part of the litigation, not an afterthought. The 2024 district-court order drew a major appellate reversal before the underlying dispute reached a settled endpoint.

There is no basis to say the federal government has banned community water fluoridation under TSCA. The 2024 order was vacated in May 2026. There is also no basis to say the appellate court resolved the neurodevelopmental-risk question in EPA’s favor. It reversed on the district court’s procedure and remanded.

That leaves a case in motion, not a clean victory for either side. Section 21 remains an unusually consequential tool because it can move a denied citizen petition into de novo review. The price of that reach is procedural discipline. Every motion, evidentiary submission and requested delay becomes part of the litigation’s architecture.

I read this docket as a warning against treating a headline ruling as a regulatory endpoint. The trial court’s order mattered. The Ninth Circuit’s reversal matters too. The next answer depends on what happens on remand, and the record does not yet supply it. For now, the law has produced a route into court, a ruling that did not survive appeal, and a regulatory dispute still waiting for its next filing.