Keep working on it. That’s the reaction of U.S. science watchdog groups to the first attempts by federal health agencies to flesh out a promise by President Joe Biden to restore trust in government by ensuring that government scientists are free to do their jobs without political meddling.

One week after taking office in January 2021, Biden ordered a review of agency policies on scientific integrity, and 2 years later the White House issued guidelines for agencies to follow in revising existing policies. In July, the Department of Health and Human Services (HHS) was first across the line with a 22-page draft policy. Last month, the National Institutes of Health (NIH), a component of HHS, did likewise.

But advocacy groups, while praising portions of the draft policies, worry they could limit the ability of scientists to speak openly about federal actions on such sensitive topics as fetal tissue research and climate change. The draft policies, part of a wider initiative to strengthen scientific integrity protections across federal agencies, are “praiseworthy” in some ways and “problematic” in others, says Anita Desikan, a senior analyst for the Union of Concerned Scientists (UCS). Liz Borkowski of the Jacobs Institute of Women’s Health adds that, without specific penalties in place, political appointees will feel free to violate the policies “with impunity.”

A coalition of 11 different advocacy organizations wrote to HHS on 1 September suggesting revisions to its draft policy. Borkowski who served as the letter’s lead author, says the policy fails to provide protections for federal grant recipients and is ambiguous when it comes to allowing scientists to speak publicly on their areas of expertise without prior approval and protecting them from reprisals by their superiors.

The policies are partly a response to the administration of former President Donald Trump, which was widely criticized for attempting to stifle scientists and distort data that conflicted with its political agenda. They are intended to go further than policies created under former President Barack Obama to prevent such scientific integrity violations, which were largely ignored under Trump.

Watchdog groups say problematic language in the HHS and NIH policies stems from what they see as flaws in a 66-page framework policy issued in January by the White House Office of Science and Technology Policy. They include a provision that prohibits agency scientists from making statements that “could be construed” as criticisms of federal policy. Desikan says the intention was likely to standardize how scientists can communicate with the public across agencies and prevent them from influencing policy decisions outside their realm of expertise. But she warns the ambiguous wording could be weaponized by “bad-faith actors” and used to silence researchers working in controversial fields.

“Virtually any statement could be construed as a judgment of federal policy,” Desikan says. Under the Trump administration, for example, HHS stopped some NIH-funded research involving fetal tissue. If a scientist were to factually state that these bans halted work on HIV drug testing, Desikan says, that might be viewed as criticizing agency policy.

Another major gap in the HHS policy, critics note, is its failure to extend protections to nongovernment scientists receiving grants. In 2017 under Trump, for example, the HHS abruptly axed funding for dozens of scientifically-based teen pregnancy prevention programs. HHS’s scientific integrity policy could prevent this type of interference by prohibiting the termination of grants for reasons other than breach of conduct, abusive conduct, or gross mismanagement, Borkowski argues.

Many advocates say these agency policies should ultimately be codified in legislation. Otherwise, a hostile administration could easily ignore or wipe out the entire framework. “Scientific integrity continues to be an issue under both Republican and Democratic administrations,” Desikan says. UCS points to George W. Bush–era attacks on climate science and stem cell research and the Obama administration’s age restrictions on emergency contraception, which went against U.S. Food and Drug Administration advice.

One such bill prohibiting violations of scientific integrity drew bipartisan support in 2019 and was reintroduced in the House of Representatives this July. But “the odds are very low that this goes forward,” says Roger Pielke, a policy specialist at the University of Colorado, Boulder, who has testified before Congress on the issue.

An HHS spokesperson said the agency is considering edits to its policy, which should be finalized in early 2024. NIH is taking comments until 9 November and will use them to determine whether changes are needed.

The advocacy groups are also watching for draft policies from other agencies, including the Environmental Protection Agency and the National Oceanic and Atmospheric Administration, where research in areas such as pesticide risks and climate change has in the past been subject to political pressure. They say standardization is important so that scientists in different agencies are not subject to inconsistent guidelines.