Before the Restriction: Reading Chemical Regulatory Signals

Chemical regulation is increasingly being shaped by systems designed to identify potential concerns before they become conventional restricted-substance requirements. In some countries and regions, such as the EU and Korea, new data infrastructure, hazard classifications, and regulatory frameworks are creating earlier signals about where chemical scrutiny may be headed.

The shift extends beyond traditional chemical registration. Emerging hazard criteria, chemical grouping, targeted information requests, and exposure data are giving regulators additional ways to prioritize substances and uses for closer attention.

For chemical manufacturers, product stewards, and EHS teams, the result is a broader horizon-scanning challenge: watching not only which substances appear on regulatory lists, but also which hazards, datasets, uses, and exposure questions regulators are beginning to emphasize.

EU and Korea: Horizon Scanning Embedded in Law

In the European Union, Regulation (EU) 2025/2455 establishes a common data platform on chemicals, managed by the European Chemicals Agency (ECHA), designed to bring together information generated under numerous EU laws and held by agencies including ECHA, the European Environment Agency (EEA), European Food Safety Authority (EFSA), European Medicines Agency (EMA), and the European Agency for Safety and Health at Work (EU-OSHA).

The regulation does more than consolidate databases. It requires the European Environment Agency to establish an early warning and action system for emerging chemical risks by January 2, 2027. Signals can come from human biomonitoring, national early-warning systems, scientific literature, EU agencies, regulatory information, and other datasets.

The EU’s official summary of the regulation says the early-warning system is intended to improve the scientific basis for decisions and enable emerging chemical risks to be identified. The full Common Data Platform is scheduled to be established by January 2, 2029, which essentially makes it regulatory horizon-scanning embedded in law.

ECHA’s existing Assessment of Regulatory Needs process operates on a similar principle. Authorities can evaluate individual chemicals or groups using existing hazard, use, exposure, and information from the Regulation on the Registration, Evaluation, Authorization, and Restriction of Chemicals (REACH) and Classification, Labelling, and Packaging (CLP) Regulation to determine whether additional data or eventual risk-management measures are warranted. Grouping makes the system potentially more powerful. Instead of waiting for extensive evidence on every individual chemical, authorities can examine structurally related substances together, potentially identifying concerns earlier and reducing the likelihood that one restricted chemical will simply be replaced with a similar alternative.

The EU’s CLP Regulation now incorporates additional hazard classes covering endocrine disruption; persistent, bioaccumulative, and toxic or very persistent and very bioaccumulative properties; and persistent, mobile, and toxic or very persistent and very mobile properties.

These additions expand the universe of characteristics capable of signaling regulatory concern. A substance does not necessarily need to produce the kinds of conventional toxicity signals historically associated with chemical restrictions to draw greater scrutiny. Persistence, mobility, and endocrine activity themselves can become regulatory signals.

Korea has similarly reorganized its chemical-management system. Revisions to K-REACH and the Chemical Substances Control Act that took effect in 2025 replaced the former broad toxic substance category with more differentiated categories addressing acute human hazards, chronic human hazards, and ecological hazards.

These changes matter because expanding and refining hazard classifications creates new signals that can feed prioritization and risk-management decisions.

“Europe is also increasingly taking a category approach (although they call it grouping),” said Dr. Neil Hunt, principal at Yordas Group by 3E. While PFAS is the best-known example, regulatory activities on bisphenol A analogues and phthalic anhydrides have both resulted in potentially significant regulatory action, he added.

“Knowing whether a substance falls within one of these groups is essential to avoid having an unexpected surprise later on in the process. One effective way of avoiding regulatory restrictions is to be able to argue that your key substance should not belong within one of these groups,” Hunt advised.

Think Category, Not Substance, for Horizon Scanning

Dr. Maria Doa is the senior director of Chemicals Policy at the Environmental Defense Fund (EDF). Having spent 30 years working at the U.S. Environmental Protection Agency (EPA), including serving as director of the EPA's Chemical Control Division, she has thought a lot about chemical hazard classifications and how chemicals are utilized. Doa suggested that companies watch chemical families and categories rather than individual substances. A company that sees regulatory pressure building around several phthalates, PFAS, or brominated flame retardants should not assume structurally similar chemicals are necessarily safe regulatory havens.

“Some companies do this. They look at EPA regulating three phthalates and say, 'We're not going to deal with other phthalates because they have similar effects.' Or PFAS: When PFAS first became a big thing, a couple of companies got totally out of PFAS. Some went to alternative PFAS, and there's still a problem,” she told 3E.

Doa argued that substance-by-substance regulation can encourage regrettable substitution: modifying a molecule enough to escape an immediate restriction without eliminating the underlying concern. Her conclusion is that category-based thinking is both more efficient for regulators and a smarter forward-looking strategy for industry.

“I saw this when I was at EPA with brominated flame retardants. If you don't look at things as categories or try to address them that way, companies will go to something similar. If EPA looked to regulate a flame retardant, folks would change it by one bromine to another flame retardant that wasn't being regulated,” Doa revealed. “Looking at categories is much more efficient. Looking to tomorrow, I think that's a smart way for companies to look, plan, and try to move to alternatives that are effective but aren't in the family that's already been marked.”

Doa identified economic incentives and incumbent chemicals as a major obstacle to faster substitution. Even when safer substitutes exist, incumbent chemicals benefit from established markets, infrastructure, and profitability. She pointed to asbestos, trichloroethylene, and methylene chloride as examples where regulatory action has been protracted despite available substitutes. She also argued that society does not adequately price downstream costs such as PFAS cleanup and human-health effects into decisions about chemical use.

Doa said consumer education can also be challenging. No one enjoys getting grease from takeout food in their car or on their clothing, but there are a lot of things where consumers wouldn't really be affected as much, or there is efficacy from the alternative, said Doa.

“One example I like is from somebody from the state of Washington [related to] the paper or cardboard used for takeout. How long is your takeout? If you only eat half of your takeout, you may keep the rest a week or two in your refrigerator,” she said. “Do you need to use PFAS, which lasts for hundreds of years, for a product that has a shelf life of two weeks?”

To the argument that alternatives such as untreated paper do not work as well as containers treated with PFAS, Doa countered: “Well, if you want a paper wrapping that lasts hundreds of years, use the PFAS.”

In the U.S., Watch the Exposure Questions

Chemical categories and hazard classifications are not the only early signals. The questions regulators ask about how a chemical is actually used can reveal another dimension of emerging scrutiny.

The EPA’s August 18, 2026, draft TSCA risk evaluation for ethylene dibromide examined potential exposures across the chemical’s life cycle, including manufacturing and import, processing, distribution, industrial and commercial uses, aircraft refueling, and disposal.

The agency designated ethylene dibromide a high-priority chemical in 2019 and released draft human-health, environmental-hazard, and exposure assessments in June 2026. Using chemical-specific monitoring data, exposure modeling, and health information, the EPA's draft identified significant contributions to unreasonable risk for workers and occupational non-users from inhalation and/or dermal exposure under 10 of the 12 conditions of use assessed. The EPA did not preliminarily identify unreasonable risk to consumers, the general population, or the environment from ethylene dibromide itself. That distinction illustrates why increasingly detailed information matters.

Doa supports the premise that restricted-substance lists alone are not enough for forward-looking compliance, but with an important qualification: Prioritization lists themselves are not necessarily backward-looking. The larger problem is the long lag between identifying a chemical as a potential concern and taking regulatory action.

Doa views prioritization as a potentially useful horizon-scanning signal for companies, using the TSCA Work Plan as her prime example. The EPA effectively warned industry years in advance which substances were attracting regulatory attention, yet when regulatory action can take a decade or more, the warning may lose urgency and companies may have little immediate incentive to respond.

“Even for things that had been on the list since 2012, folks were saying, 'Who knew, who knew, who knew? Don't regulate me. I need time, I need time.' We saw that with a number of the PBTs (persistent, bioaccumulative, and toxic chemicals),” Doa told 3E.

Doa also warned of the distinction between hazard information and exposure information. Regulators may understand a chemical's toxicology reasonably well while knowing far less about how workers and consumers are actually exposed to the chemical. The EPA's ethylene dibromide information request illustrates the issue, she said.

“Some of the information they're asking for helps improve their modeling of exposure,” said Doa. “One of their questions is the number of aircraft [worker] refuels on a typical eight-hour day. That's good input. Or the average duration of refueling an aircraft.” Other important information for regulators, said Doa, concerns the transition involved in inserting the fueling mechanism into the plane and removing it. How many workers are exposed to that?

“Knowing that a substance has hazardous properties is important, but under TSCA exposure, it is also important for a regulator to determine which activities require intervention,” said Doa. Regulators also need to understand who is exposed, through what pathway, under which conditions, and at what level, she added. The regulatory outcome can therefore depend not merely on the intrinsic hazard of the chemical but on the information available about what actually happens throughout its life cycle.

“I think that's all needed information because exposure is the one area where information is often really poor,” said Doa.

This example raises an important question for horizon scanning: What happens when toxicological knowledge advances faster than regulators’ understanding of real-world exposure? Questions seeking detailed information about worker activities, exposure pathways, frequency, duration, and controls may themselves indicate where regulators are trying to fill gaps before deciding what intervention is warranted.

UK Shifts Toward More Targeted Chemical Data

The UK’s developing Alternative Transitional Registration Model (ATRm) for UK REACH offers a different variation on the trend toward more targeted chemical information.

The UK government initially proposed reducing the hazard information required for transitional registrations while placing greater emphasis on information about chemical uses and exposures in Great Britain. Following consultation, however, the government announced in March 2026 that it would not introduce the proposed additional use and exposure information requirements. Instead, transitional registrants will provide use and exposure information in line with existing UK REACH requirements.

The decision does not necessarily signal a retreat from information-driven chemical regulation. The government’s broader approach to the ATRm is intended to reduce the amount of hazard data companies must duplicate from EU REACH while allowing regulators to seek additional information in a more targeted way when new or emerging risks are identified. The government has described the approach as one that can be augmented by requiring further data for regulatory purposes as those risks emerge.

The approach illustrates an important evolution in the debate over chemical data: More information is not necessarily synonymous with better regulation. The regulatory question increasingly may be which information is needed, when it is needed, and how regulators can obtain it efficiently enough to make risk-management decisions.

That is particularly relevant to the EPA ethylene dibromide example. In both cases, regulators may already have substantial information about a substance’s hazards. The remaining questions may concern how the substance is actually used, who is exposed, how exposure occurs, and what controls are in place.

Reading the Signals Before the Restriction

The important question for horizon scanning is therefore no longer simply which chemical might be restricted next. Companies can also watch the systems regulators are constructing, the categories they are creating, the chemical families they are grouping, and the information they are requesting.

None of those signals guarantees that a restriction will follow. A request for exposure information is not itself a restriction, just as placement within a regulatory assessment process does not necessarily mean that a substance eventually will be prohibited.

That distinction is precisely what separates horizon scanning from conventional restricted-substance monitoring.

The EU’s emerging risk infrastructure, Korea’s more differentiated hazard categories, regulatory grouping of chemical families, and EPA’s detailed exposure questions all operate at different stages and for different purposes. Collectively, however, they demonstrate the range of information that can appear before a conventional restriction or compliance deadline.

For chemical manufacturers and product stewardship teams, the opportunity is not to predict with certainty which chemical will be restricted next. It is to recognize when classifications, categories, exposure questions, scientific concerns, and regulatory infrastructure begin pointing toward an issue that deserves closer attention.

By the time a chemical appears on a restricted-substance list, those earlier signals may have been visible for years.

(This is Part 2 of a three-part series. Part 1 is Horizon Scanning: Moving From Restricted Lists to Regulatory Intelligence, which examines the move from restricted-substance lists toward regulatory intelligence. Part 3 is Beyond the Chemical List: Building Regulatory Readiness Into Product Data, which examines how companies can connect those regulatory signals with product, supplier, facility, and exposure data).