EPA is sidelining its independent chemical referee – and that endangers public health

For decades, the U.S. Environmental Protection Agency has relied on an independent scientific program to answer two basic questions when chemicals come up for review: Does the chemical pose a threat to human health? If so, how much exposure is necessary before it becomes a problem?

The scientists involved in that program, known as the Integrated Risk Information System, or IRIS, served as neutral scientific referees.

Now, the Trump administration is dismantling the program and moving the scientific assessment role to policy offices, opening the door for political pressure. The administration is also making it easier for past IRIS assessments to be revisited or overturned.

This change is not merely bureaucratic: It reshapes whether future assessments of chemical dangers will be ignored, delayed by time-consuming legal fights, or understated by the federal government, potentially with real consequences for public health.

Numerous chemicals are hazardous to human health. For example, ethylene oxide is used to sterilize medical equipment. However, studies show ethylene oxide poses elevated cancer risks to people who live near facilities that release it. Chromium‑VI, used as a corrosion inhibitor and for metal finishing, can contaminate drinking water. Made famous by the Erin Brockovich case, it has been linked to cancer and other adverse health effects. Formaldehyde, found in building materials and household products, has long raised concerns about cancer and respiratory disease.

EPA scientists assessed each of these chemicals through the IRIS program. Now, the IRIS program itself, as well as many of its formal assessments of over 550 chemicals developed over four decades, is being challenged under the Trump administration.

What IRIS did – and what it didn’t do

In any high‑stakes game, the referee enforces the rules so the outcome rests on the facts, not on who shouts the loudest or has the most at stake.

IRIS played that role for chemical safety. It was part of the EPA’s Office of Research and Development, which was recently dismantled by the Trump administration. Its scientists assessed whether chemicals cause harm and weighed how health risks changed with a person’s increasing exposure to the chemical. These scientists did not estimate real‑world exposures, decide acceptable risk or make regulatory choices. Those functions were handled in policy offices.

I have worked with IRIS assessments from multiple perspectives — as a professor of environmental engineering, as a reviewer for the National Academies and EPA science advisory processes, and as assistant administrator of EPA’s Office of Research and Development from 2022 to 2024, where I oversaw the IRIS program.

IRIS assessments were written by EPA scientists and rigorously reviewed by independent external peer reviewers with experience in each specific chemical. The assessments have been used across EPA programs and by states, local governments and tribes, and internationally. Industry representatives, environmental groups, other federal agencies and members of the public all had opportunities to comment on the drafts of assessments before they were finalized.

When disagreements arose over IRIS assessments, independent scientific experts were asked to weigh the evidence and advise the EPA on how to move forward. That process, relying on scientists, not stakeholders, was meant to ensure that scientific judgments were grounded in evidence, not in policy preferences or financial interests.

The actual policy decisions to regulate chemicals were made elsewhere, by EPA officials and, in some cases, by states or other jurisdictions. IRIS provided the scientific foundation so those decisions could be informed by an evidence‑based understanding of chemical hazards.

IRIS assessments effectively set the standard for assessing chemical hazards internationally. Other agencies and countries rely on IRIS assessments precisely because they are comprehensive, transparent and independently reviewed.

Why critics wanted IRIS dismantled

That track record matters.

Some industry‑aligned organizations have argued that IRIS assessments are flawed or biased and have called for eliminating the program.

However, independent scientific reviews have repeatedly examined these concerns and found that IRIS methods reflect the current state of the science and have strengthened in rigor, transparency and consistency over time.

It’s true that IRIS assessments often took years to complete, but that was because extensive interagency review and limited staffing slowed the pace at which assessments could inform regulatory decisions. Delay is not the same as poor science.

What changes when the referee disappears?

With IRIS eliminated as an independent program, chemical hazard assessments will be overseen by regulatory offices that also weigh economic impacts, legal risk and policy priorities.

When scientific assessments are developed within offices responsible for policy decisions, it becomes harder to maintain a clear separation between evaluating evidence and weighing its regulatory consequences. That separation has historically helped ensure that scientific conclusions are grounded in evidence alone.

Courts generally give weight to agency expertise when decisions are supported by a clear and well‑documented scientific record. However, when agencies fail to clearly explain how the evidence supports their decisions, including when agencies depart from their own scientific assessments, courts can block those decisions under the Administrative Procedure Act or other laws, such as the Clean Air Act.

The result can be prolonged litigation and delays in developing or implementing regulations, with consequences for public health.

How communities are affected

Industries have long challenged scientific findings that show their products can cause harm – from tobacco smoke to particulate air pollution and greenhouse gas emissions from fossil fuels.

When public health is at stake, I believe independent referees are essential to ensure that facts are determined by evidence, not by the industries that would benefit. Shifting away from independent scientific review risks undermining that foundation.

I served as Deputy Assistant Administrator for Science Policy at the U.S. Environmental Protection Agency from 2021 to 2022, and as Assistant Administrator for Research and Development at the U.S. EPA from 2022 to 2024.