The Environmental Protection Agency is changing its approach to chemical toxicity oversight, according to a report issued recently by the Government Accountability Office. In the overhaul, the EPA reassigned staff from its Integrated Risk Information System (IRIS)—a program that conducts comprehensive scientific reviews—to duties related to the Toxic Substances Control Act, which has a narrower mandate. The agency has also reduced the number of its ongoing chemical toxicity assessments from 20 to three. Former EPA officials contend the shake-up takes chemical assessments out of the hands of career scientists, potentially to the detriment of public health.

The EPA also recently halted release of a long-awaited formaldehyde toxicity assessment. In testimony before a congressional oversight committee on April 9, EPA Administrator Andrew Wheeler said the study, which had already been completed by IRIS, will instead be reconducted under the TSCA program. Formaldehyde, which is used in manufacturing pressed wood, adhesives and insulation, has been linked to leukemia.

IRIS was created in 1985 to study chemicals’ toxicity to humans. The program’s assessments “are the preferred source of toxicity information used by the EPA,” according to the agency’s website, which says EPA program offices (units responsible for specific areas such as air pollution or water quality) use IRIS toxicity values to determine public health risks posed by chemicals. The TSCA, passed in 1976, more narrowly authorizes the EPA to review and regulate chemicals determined to pose an “unreasonable risk” to human health and the environment.

An EPA official, who spoke on the condition of anonymity due to not being authorized to talk to the media, says IRIS and TSCA are “very different” in their approaches to chemical safety regulation. “One could make the argument that this is political interference, in that high-level people are saying which methodology we should be using to assess the safety of a chemical,” the official says. “And the policy’s pretty clear that they’re not supposed to do that.”

Under the changes, EPA leadership also now requires a program office to make a formal request for a chemical toxicity assessment before IRIS can release it to the public. According to the GAO report, the EPA’s Office of Research and Development (ORD) informed IRIS officials in June 2018 of this new requirement. The report adds that at the same time, the EPA administrator (then Scott Pruitt, who was succeeded by Wheeler the following month) directed IRIS officials to request reconfirmations of 20 chemical assessments—which were then already under way—from program and regional offices. While those were being compiled, the report says, ORD leadership instructed IRIS not to publicly release any assessment documentation—including chemical assessment documents that were ready for agency or peer review or for public comment. Possibly as a result of these changes, IRIS did not release a new chemical assessment for the remainder of 2018.

Genna Reed, a science and policy analyst at the Union of Concerned Scientists, says TSCA has become politicized, and that “shifting IRIS scientists to a more political process to look at these chemicals is undermining the work of EPA’s own scientists.” As evidence of politicization, Reed points to the 2017 appointment of Nancy Beck—a former lobbyist for the American Chemistry Council—to deputy assistant administrator of the EPA’s Office of Chemical Safety and Pollution Prevention, which implements TSCA.

“I really see this as part of a restructuring of EPA in such a way that science will have very little to do with what EPA is basing its regulation on, and that we will end up with much weaker regulations in terms of protecting the public health,” says Bernard Goldstein, who served as EPA assistant administrator for research and development in 1983–85. “It’s troubling, in large part because it’s very consistent with an overall approach—a very astute approach—to take out inconvenient facts.” Thomas Burke, a former EPA lead science adviser and the Deputy Assistant Administrator of the ORD from 2015-17, says “any reduction” of the number of IRIS chemical assessments “is a loss for public health and, unfortunately, puts populations who are exposed at risk.”

The IRIS assessment of formaldehyde toxicity was reportedly ready to be made public as early as 2017, but its release was suspended in December 2018. On April 9 this year, EPA Administrator Wheeler told a Congressional Energy and Commerce Committee hearing that the EPA “will not be moving forward” with the assessment. Wheeler told the committee formaldehyde will instead be reviewed under the TSCA program; when asked whether the IRIS assessment would ever be made public, he did not answer directly. The EPA office of public affairs had not responded to repeated requests for comment by the time of publication.

“If any IRIS assessment has stood the test of review, formaldehyde is one of them,” Burke says. “I think it’s a shame to see that slow-walked and shifted over to [TSCA], where there is a much narrower definition of evaluating potential exposures … rather than providing a big, robust evaluation of the full body of evidence.”

“I am concerned that the EPA under Administrator Wheeler is not carrying out its fundamental responsibility to protect Americans from exposure to harmful toxic chemicals as outlined in the GAO report,” says Representative Mikie Sherrill (D–N.J.), who chairs the House Subcommittee on Investigations and Oversight. Sherrill says the committee “needs to ensure that political interference within the EPA, such as suppressing the formaldehyde report, does not interfere with sound science and our safety.”

Wheeler testified that formaldehyde was not one of the chemicals a program office had designated as high-priority during last summer’s review. According to Wheeler, the advantage of using the TSCA risk evaluation process is that it allows for regulation at the end of the process. “If we were to move ahead with the formaldehyde IRIS assessment, it would be a minimum of 18 months,” Wheeler told the committee. “And we decided it was more important to put formaldehyde through the TSCA program, because at the end of the day you can regulate formaldehyde under TSCA.”

Burke disagrees with this characterization. The EPA “can use the [IRIS] evidence base for a pervasive environmental contaminant and use the full extent of the statutes,” Burke says. “Moving it to the TSCA program, where the scope would be greatly narrowed, and the evidence base would be narrowed—I wouldn’t agree with that.” Rita Schoeny, who was a senior science advisor at the EPA until 2015, says it is accurate “on paper” that IRIS does not have specific regulatory authority—but that Wheeler’s characterization could lead to misinterpretation. “IRIS is not toothless; it carries a lot of weight,” Schoeny says. “The science, the risk assessment, is an enormous driver in terms of how regulations are set.”