On a Friday afternoon in mid-March, a bunch of (presumably) non-hackers showed up to participate in a new kind of “hackathon.” It was Pi Day, so pie was served, but the mood wasn’t celebratory. Students, researchers, and members of the wider public health community were there, in a lecture hall at the Harvard T.H. Chan School of Public Health and on Zoom, to rescue data. Whatever they could. According to the flier for the Preserving Public Health Data Hackathon, the current Trump administration was trying to undermine research on everything from climate change to systemic racism, and saving data from government websites meant it would be archived and republished in the event federal agencies tried to remove it. The antiauthoritarian theme came up often: don’t obey in advance.
Everyone in the room and on the Zoom got a crash course in identifying at-risk information, collecting it, and storing it once they did. Backing up certain pieces seemed critical.
For those who work in public health, protecting research pertaining to gender identity and diversity issues has been a focus since the inauguration. Last year, then-candidate Trump repeated a lot of anti-trans rhetoric on the campaign trail, and in the first weeks of his presidency signed executive orders essentially barring transgender people from serving in the military, proclaiming that the US government would only recognize “two genders, male and female,” threatening to withhold federal funds in what is an attempt to bar trans women from sports, and attempting to block gender-affirming care for people under the age of 19. Since Trump’s inauguration, hundreds of words—from “transgender” to “Latinx” to “accessible”—have been removed from federal agencies’ websites. In early February, thousands of websites went missing as agencies raced to comply with the executive orders.
In recent weeks, the targeting of gender-related material has become even starker, after Trump told a joint session of Congress that the so-called Department of Government Efficiency (DOGE) had uncovered that the federal government had spent $8 million on “making mice transgender” (it hasn’t), and the Department of Defense performed a purge of “DEI” materials that included flagging, but ultimately not removing, images of the Enola Gay.
One particular dataset, the Youth Risk Behavior Surveillance System (commonly shortened to YRBS), seems particularly at-risk, notes Ariel Beccia, an epidemiologist at the Chan School’s LGBTQ Health Center of Excellence. The YRBS, which tracks scores of health-related issues in young people, is one of the only nationally representative surveys that regularly collects data on transgender kids. Beccia and other public health researchers fear the information in the YRBS may not be available forever.
And even if the YRBS can be backed up, the integrity of its data can’t be completely protected from the influence of the new administration. President Trump’s recent moves have also left some in public service afraid to participate in new research, like this year’s survey. “Because of this ‘comply in advance’ strategy, school boards are hesitant to participate in the YRBS,” Beccia told attendees at the hackathon.
Beccia would know. Her research focuses on race and ethnicity as well as LGBTQ+ young people. She’s recently been looking into LGTBQ+ inequities in eating disorders. Her work relies on YRBS data. Now that the Centers for Disease Control and Prevention (CDC), which hosts the survey’s results, has reportedly stopped processing data on trans Americans, and fewer schools are participating, the data Beccia uses, even if it stays online, will be incomplete.
When asked about this, CDC spokesperson Melissa Dibble confirmed that, in compliance with Trump’s executive orders, “the transgender identity question was removed from the national Youth Risk Behavior Survey,” but no other changes were made. Dibble added that the change shouldn’t delay the survey’s results.
It’s created a “double whammy” situation, Beccia says, where government employees are trying to comply with Trump’s executive orders and school districts are afraid to participate in any study related to gender or LGBTQ+ health, even if it is data that in turn guides how they run their schools.
“We are living through a pretty scary time with the administration,” Beccia tells me a few days after the event. “This is obviously impacting the mental health of everyone in the country, especially queer and trans people and queer and trans youth, and we’re not going to have data on this.”
The YRBS is just one of many datasets researchers have sought to shore up in a long-running effort to preserve government-funded information as the new administration takes control. The University of Washington Information School held “Data Rescue” events in January aimed at collecting climate crisis information. The Environmental Data and Governance Initiative (EDGI), a network of research professionals that launched a tool during Trump’s first term to track changes to environmental information on federal websites, relaunched that tracker at the beginning of March. The Data Rescue Project, a consortium of data-rescue organizations, lists dozens of ways people can get involved if they want to help collect. A group of archivists has recreated the pre-Inauguration Day version of the CDC website and is now hosting it in Europe at RestoredCDC.org.
Other organizations are also looking to back up the YRBS, too. On January 30, soon after Trump signed the first of his executive orders aimed at trans Americans, Libby Hemphill, director of the Resource Center for Minority Data at The Inter-University Consortium of Political and Social Research (ICPSR), started getting calls. Word had gotten out that the CDC might be scrubbing data, including the YRBS, and people wanted to know how to stop it. Hemphill gathered some colleagues and started scraping it. Then there were requests for data from the National Oceanic and Atmospheric Administration (NOAA), from the Department of Education, from the National Institutes of Health.
Much like the data-preservation effort at the Chan School, Hemphill and her team coordinated ways for people to submit data they wanted to preserve and ways for researchers to collect it and store it. The ICPSR runs a repository known as DataLumos, where a lot of the preservation efforts are backing up their collections. When I ask Hemphill if she’s worried about the safety of the information in the DataLumos archive, she says “it’s absolutely something that we think about,” adding, “I can assure you that ICPSR has a non-US physical, non-US regulatory plan for data preservation.”
If you think this sounds familiar, you’re not wrong. When Trump assumed the presidency in 2017, scientists, archivists, and librarians at the University of Pennsylvania raced to save data published by the Environmental Protection Agency (EPA) and NOAA. Another group in Michigan, also fearing the EPA and NOAA websites would lose valuable information, made a similar move. Websites were backed up to the Internet Archive; large datasets were “bagged” for safe keeping.
At the time, the researchers weren’t sure the incoming administration would seek to erase any info. It was more like a hunch, one that proved prescient when, then being led by Trump appointee and agency administrator Scott Pruitt, the EPA began removing climate change information from its website in April 2017, “to reflect the approach of new leadership.”
Between 2017 and 2021, more than 1,400 pages related to climate change on government web sites were altered or made less accessible, according to data compiled by the EDGI. That, notes Gretchen Gehrke, who leads EDGI’s website monitoring program, is not “a comprehensive list of changes,” since some alterations—like removing “Climate Change” from the navigation page of EPA.gov—only get counted once but affect several other pages.
“I think there is a lot more awareness about the precarity of federal information after having experienced the first Trump administration,” Gehrke says. “Watching the Trump campaign become truly obsessed with trans people, and knowing the Trump administration’s history of information suppression, people were and are rightly concerned that that information is at risk.”
Which is why Beccia is concerned. Datasets like those in the YRBS are few and far between and losing it could be disastrous to those wanting to know about the health and well-being of trans youth in America.
Although the YRBS is currently live on the CDC’s website, it did briefly disappear, along with data on the Food and Drug Administration and Department of Health and Human Services web sites, earlier this year following an order from the Office of Personnel Management that it be scrubbed to comply with Trump’s executive orders.
The information returned in mid-February when US District Judge John Bates, responding to a lawsuit from Doctors for America, granted a temporary restraining order and the site was reinstated. A disclaimer at the top of the YRBS page now says “any information on this page promoting gender ideology is extremely inaccurate,” adding “this page does not reflect biological reality and therefore the administration and this department rejects it.”
Tazlina Mannix worked for the YRBS program in Alaska from 2015 until 2023, both as survey coordinator and data manager. She notes that even if the CDC keeps the data online, disclaimers like the one on the site now make it harder for researchers to do their work. Collecting public health data relies on relationships with people in health departments and school districts. Giving those people any reason to hesitate can “set you back to zero,” she says. “When I first saw [that disclaimer], I was so horrified. The language is so extreme, and it’s also just wrong.”
