Samizdat for Science

The Atlantic

10-02-2025

How can we prevent our suicidal patients from killing themselves? That's an important question for a primary-care physician like me. I am often in the position of trying to assess—in 15 minutes or less—which patients need urgent treatment. The type of guidance that might help me can be found in a paper that was published in 2022 on PSNet, the Patient Safety Network, a federally funded initiative. 'Few considerations are more critical,' the authors wrote, 'than identifying a person at risk for taking their own life.'
On January 31, however, the authors of that paper received a notice that their peer-reviewed article had been struck from the PSNet website. Apparently, it violated Executive Order 14168, 'Defending Women From Gender Ideology Extremism and Restoring Biological Truth to the Federal Government,' signed by Donald Trump on his first day in office.
In addition to being a physician, I happen to be a woman, so I was curious why women needed defending from an analysis of how health professionals might better help suicidal patients. In the paper, the authors reminded clinicians to keep in mind which patient groups are known to be at higher risk, citing peer-reviewed data: ' High risk groups include male sex, being young, veterans, Indigenous tribes, lesbian, gay, bisexual, transgender, queer/questioning (LGBTQ).' The acknowledgment of transgender people, however peripheral, was apparently enough to invite the ax.
The memo came out on a Wednesday, and agencies had until 5 p.m. on Friday to scrub their websites—as well as their agencies, grants, contracts, and personnel—of anything that might 'promote or inculcate gender ideology.' As a result, hundreds of government websites were shorn of articles, pages, and data sets about transgender issues, along with information on contraception, HIV, and abortion.
Much of the information that was stripped came from the CDC website, but even pages on the Census Bureau and the National Park Service sites came down. The tech-news publication 404 Media has estimated that more than 2,000 data sets have disappeared from government websites since Trump took office.
Coupled with other recent actions— pulling out of the World Health Organization, muzzling communications from government health agencies, stopping funding for overseas programs that treat HIV and malaria, drastically cutting NIH research funding —the Trump administration is signaling its contempt for evidence-based science and doing so in a way that demonstrates its sweeping disregard for human health and life.
Federal agencies and employees may be required, for the moment, to follow these guidelines. But the path for nongovernmental medical and scientific organizations is clear: Every hospital, university, professional medical organization, residency program, scientific organization, and nursing and medical school needs to insist that these data remain accessible to the public.
The science and health-care communities must also work together to make available all of the expunged data. This is beginning to happen: Individual researchers, doctors, students, and self-declared data hoarders have been racing to download as much of these crucial data as possible. Efforts such as the Internet Archive, the Library Innovation Lab Team, the End of Term Archive, and other groups to archive and host public data can prevent the erasure of years of scientific progress, and, by preserving this information, create a kind of scientific samizdat.
As of yet, major medical and scientific organizations have not formally stepped into the void. Doctors for America has filed a lawsuit over the expunged data, claiming that the actions were unlawful and endanger the lives of Americans. But bigger groups such as the American Medical Association and the American Association for the Advancement of Science remain on the sidelines.
A week after the initial purge, the agency that hosts PSNet was informed that the paper could be reposted on condition that the words transgender and LGBTQ be removed. The senior author rightly refused, stating that the researchers would remove those terms only if the Trump administration could cite verified data demonstrating that LGBTQ and trans communities did not have a higher risk of suicide. In that case, they would issue a correction. The fact the government would interfere with scientific work at this level at all is startlingly authoritarian.
The Trump administration may feel that winning the election grants it the authority to alter science to its liking. It may even get nongovernmental institutions to temporarily parrot the party line by threatening to withhold funding. But the scientific community needs to stand its ground. Doctors and nurses have a particular responsibility, because we have sworn oaths to put patient welfare first. As Dr. Steven H. Woolf succinctly put it in a recent editorial: 'We must draw the line when the science is clear that a policy will increase the risk of disease, complications, or premature death.'
The legal challenges to Trump's executive orders are piling up, though it will take time for these actions to grind through the courts. Fortunately, the medical and scientific communities need no such delay to determine our course of action. Commitment to patients and to scientific inquiry is our unequivocal guiding principle.