One of the defining aspects of the Trump administration is the shameless peddling of conspiracy theories. There’s the myth that Democrats installed a spy on the president’s 2016 campaign, for instance, and the one about the Chinese making up climate change. Over at the Environmental Protection Agency, political appointees—including former administrator Scott Pruitt and his replacement, Andrew Wheeler—are promoting a conspiracy theory of their own. They claim that the EPA improperly uses “secret science” to justify unnecessary regulations—and they want to forbid the agency from doing so.

Claims about “secret science” stem from the fact that many studies linking pollution to illness rely on medical, personal, or occupational records that aren’t public—not for any nefarious reason, but because of patient privacy laws or confidentiality agreements. In 2018, Pruitt proposed excluding these types of studies from future policy-making. Some 600,000 public comments poured in, many of them pointing out that the measure, by excluding a large body of epidemiological research from consideration, would make it difficult for the EPA to fulfill its core mission of protecting public health. But instead of backing down, the EPA is now pushing for a dramatic expansion of its initial proposal. A new draft supplement to the rule, published by The New York Times last week, indicates that the data disclosure requirement could also apply retroactively—meaning that the EPA could use it as justification to gut long-standing clean air and water protections if they were based, in part, on research that incorporated confidential data.

Trump administration officials claim that the purpose of the data disclosure rule is to increase “transparency.” But the history of the idea indicates otherwise: It was originally ginned up by the tobacco industry for the explicit purpose of hobbling environmental regulators. In 1996, after the EPA concluded that secondhand smoke “presents a serious and substantial public health impact,” Christopher Horner, a lawyer working with the R.J. Reynolds Tobacco Company, submitted a memo to his clients laying out a plan to fight regulation. “Our approach is one of addressing process as opposed to scientific substance,” Horner wrote. Specifically:

We propose creating, beginning with congressional oversight and a goal of enacting legislation, required review procedures which EPA and other federal agencies must follow…. This is important to your organization because, at some point in the near future, EPA will most likely be ordered to re-examine ETS [environmental tobacco smoke].

The goal, Horner continued, was “to construct explicit procedural hurdles the agency must follow in issuing scientific reports,” and to avoid any case-by-case consideration of public health threats by “focusing on the process by which EPA arrived at its scientific conclusions, avoiding to the extent possible specific scientific issues, contaminants, or industries.” He proposed several criteria the EPA should be forced to add to its scientific processes, including “transparency.” This approach, Horner suggested, could help the industry weaken regulations on mercury, hazardous waste, and air pollution.

Two years later, a lobbying firm working for R.J. Reynolds created a “Secret Science” working group. The group’s goal, according to a memo obtained by the Union of Concerned Scientists, was to “focus public attention on the importance of requiring the disclosure of taxpayer-funded analytical data upon which federal and state rules and regulations are based.”

The idea spread quickly from the tobacco companies to the energy industry, as Sharon Lerner reported for The Intercept. Of particular concern to energy companies was groundbreaking research on air pollution conducted by a team from Harvard University. Known as the “Six Cities” study, the research linked a type of particulate matter known as PM 2 to increased mortality, as well as asthma, lung cancer, and other illnesses. When the EPA moved to limit this type of pollution, the Koch brothers funded a front group, Citizens for a Sound Economy, which accused the EPA of concealing the study data. During a protest organized by CSE in 1997, protestors in white lab coats held signs reading, “Harvard, release the data!” Eventually, the scientists allowed two other research teams to reassess the data. Their conclusion? That rates of PM 2 did indeed correspond with mortality.

While the tobacco industry wasn’t successful in changing the EPA’s scientific review process, the idea was kept alive by conservative think tanks and people like Christopher Horner, who went on to work at the Competitive Enterprise Institute and then for the Trump administration’s transition team at the EPA. One of his colleagues there was Steve Milloy, a former tobacco and oil industry flack who helped to popularize the term “secret science” when he worked for a front group funded by Philip Morris in the late 1990s. Milloy, a climate change denier, has also claimed that “air pollution kills no one.” (In fact, researchers estimate that some 200,000 people die from air pollution in the US each year.)

The proposal finally made its way into legislation in 2014, when then–Texas Representative Lamar Smith introduced a bill to prohibit the EPA “from proposing, finalizing, or disseminating regulations or assessments” if any of the underlying research relied on confidential data. The bill passed the House but died in the Senate; similar legislation failed again in 2017. After Trump was elected, Smith pitched his idea directly to the EPA, hoping to circumvent Congress. An agenda obtained by the Environmental Defense Fund for a meeting between then-administrator Pruitt and Smith (who has since retired) describes the objective of the meeting as, “to find a way to have EPA implement [Smith’s legislative] objectives outside of the legislative process since it is unlikely to pass in the Senate.” One of Smith’s staffers who worked on the bill, Richard Yamada, had joined the EPA, and there he helped to advance the first version of the proposal to exclude research based on non-public data.

“The era of secret science at EPA is coming to an end,” Pruitt said when he rolled out the proposal in early 2018, echoing the tobacco industry language. The draft rule itself was “a hasty hot mess,” according to John Walke, a former EPA attorney who now directs the clean air, climate, and clean energy program at the Natural Resources Defense Council. The rule relied on bits and pieces of eight different environmental statutes for its legal authority, which many attorneys found incoherent. The EPA received hundreds of thousands of public comments in opposition to the “censored science” rule, as scientists and environmental advocates dubbed it, arguing that it attempted to solve a nonexistent problem, and would prevent the EPA from considering the best available research in decision-making.

Things only got worse. The revised supplemental proposal now under consideration is “a breathtaking expansion of the original,” Walke said, applying the public data requirement to all science used by the agency for policy-making. The draft is so broad that it “target[s] EPA’s core mission to protect public health, air quality, water quality, and the safety of toxic chemicals and pesticides,” said Walke. Applying the rule retroactively could put a number of existing health protections in jeopardy when they come up for renewal. For instance, key studies linking lead in paint dust to developmental disorders in children might be considered inadmissible, as could research on the effects of mercury emissions from coal plants. The agency could also potentially avoid crafting new regulations for emerging risks, like those posed by per- and polyfluoroalkyl substances, a class of chemicals used in various consumer products from nonstick pans to clothing.

According to the supplemental proposal, the EPA has abandoned its original legal justification for the rule and is now basing the proposal on a two-sentence “federal housekeeping” law that doesn’t mention the EPA—because it was written in 1966, before the agency existed. When, at a hearing held last week by the House Committee on Science, Space, and Technology, Democratic Representative Jennifer Wexton pointed out that the EPA wasn’t included in the list of “executive departments” covered by the housekeeping law, the EPA official who’d be sent to testify responded awkwardly, “You raise a good point.”

“EPA cannot find any coherent plausible authority under existing federal statutes to adopt the censored science rule. They’re flailing,” said Walke. But, he added, “I think we still must treat it as a threat, because the Trump administration is obviously devoting extensive resources to it.”

The real-world impact of limiting EPA’s use if epidemiological research would be felt most acutely by children, whose developing bodies are more vulnerable to pollutants, and by low-income communities and communities of color, which often bear a greater burden of heavy industry in their neighborhoods. “They are the ones relying on EPA to do chemical risk assessments that define how harmful pollutants are,” said Gretchen Goldman, the research director of the Center for Science and Democracy at the Union of Concerned Scientists. “This rule would completely devastate EPA’s ability to do those chemical assessments. Without that, communities won’t know how harmful things are, and they’ll lose mechanisms by which they can advocate for themselves.”

Aside from their conspiratorial insinuations, supporters of the data disclosure requirement haven’t presented any evidence that it’s actually needed. The EPA already has extensive quality and integrity standards, including peer review, and a lengthy process for crafting regulations, much of which is transparent to anyone who wants to watch hours of meetings or read through thousands of pages of documents. “It was never about transparency—it was always about finding a way to halt or delay inconvenient science that was showing the need to protect people,” said Goldman. “We know that from the tobacco industry documents.”