Updated on March 16 at 2:11 p.m. ET

In 2007, two senators—one Democrat and one Republican—proposed a bill to accelerate research in human genetics. The young Democrat, a relative unknown named Barack Obama, saw a future in which regular genetic testing allowed for “personalized medicine for all Americans,” including better medical treatments, more accurate diagnoses, and safer drug prescriptions. But first, he had one condition. “I don’t want [the bill] to go anywhere until GINA passes,” he told an aide.

GINA—the Genetic Information Non-Discrimination Act—was written to stop DNA from becoming a tool of discrimination. It prevented employers from basing employment decisions on someone’s genes, and it stopped health insurers from denying coverage or charging higher premiums if someone had, say, a gene that upped their cancer risk. First proposed in 2003, it went through several incarnations before finally being passed in 2008, with nigh-unanimous bipartisan support. It was a symbolic deterrent, designed to protect against a form of discrimination that didn’t really exist yet, but was deeply feared.

With protections in place, then-President Obama finally got his wish in 2015, and launched the Precision Medicine Initiative—an ambitious plan to collect genetic and health information from one million Americans by the end of 2019. This hoard of data would give scientists an unprecedented look at the links between genetics, lifestyle, environment, and health, and allow them to start truly tailoring medical treatments at a personal level. This year, the initiative, now rebranded as the All of Us cohort, will start recruiting its first volunteers.

Which is ironic, because the GOP has just put their genetic privacy at risk by launching a devastating attack against GINA.

Last Wednesday, as reported by Stat’s Sharon Begley, a House committee approved a bill called H.R. 1313, which would change GINA’s applicability to “workplace wellness programs.” These programs, introduced under Obamacare, are meant to encourage healthy living; employees provide extensive health information in exchange for a discount on their insurance premiums. But if H.R. 1313 passes, bosses could also pressure their employees to undergo genetic tests, and demand to see the results. They could even ask to see test results and medical histories for family members. And if employees refuse, they’d end up paying 30 percent more for their insurance—an extra $5,443 a year, on average.*

A spokesperson for the House Committee on Education and the Workforce said in a statement: "This bill simply reaffirms wellness program policies reflected in the Affordable Care Act and provides uniformity to the regulations governing those programs, which have been in place for years. Opponents who fundamentally disagree with wellness programs altogether are spreading false information in a desperate attempt to undermine an option for workers to lower their health insurance premiums. Under this bill, workers enrolled in wellness programs—which are completely voluntary—would still be protected by anti-discrimination and confidentiality provisions in GINA.”

But others see it differently. The bill “basically guts GINA,” says Jennifer Wagner, a lawyer at State College, who specializes in genetics. “It means that an employee has virtually no control over their own genetic information.” Although H.R. 1313 ostensibly applies only to the wellness programs, employers could easily ask about genetic testing done in other settings, from private consumer tests like 23andme, to large research studies like the All of Us cohort.

GINA isn’t perfect. As my colleague Sarah Zhang noted, it didn’t cover life insurance or education. “But at least it offered people some sense of safety—and that safety net may disappear,” says Alondra Nelson, a sociologist from Columbia University. “GINA is the ethical infrastructure that we need if we’re to make any advances for genetic research. Without it, we’re in the wilderness.”

All of this is happening at a crucial inflection point in modern genetics. Direct-to-consumer genetic tests are increasingly popular. Advocacy groups have argued that patients and research volunteers should have full access to their genetic data. And researchers are looking to set up larger and more ambitious studies. “At the same time that All of Us is talking about a visionary way of enrolling people into a large-scale research study, suddenly this bill appears,” says Robert Green from Brigham and Women's Hospital. “It’s a terrible irony, and it’s clearly going to hurt research writ large.”

His worry is that people will be less likely to agree to any form of sequencing if there’s a risk that their data can be used against them. Those risks are already a significant deterrent. Green’s team is running two large studies to see if genomics can help to guide the medical care of adults and babies. In both cases, between 15 and 20 percent of people  who initially express interest eventually decide not to enroll specifically because they fear insurance discrimination. Another 25 percent or so decline because they’re worried about the privacy of their data.

If GINA protections fall, the reticence will rise—and reasonably so. “Even when GINA wasn’t under attack, it was difficult for researchers to allay the concerns that people had,” says Wagner. But the new bill “creates a context where it makes it hard to even articulate what the risks of participation are.”

She isn’t convinced that the bill will have an impact on research. So little data gets returned to study participants that they’d have little to disclose to prying employers, she says. But ironically, many geneticists and advocates have been pushing to change that. They argue that people should be treated as active participants in research, and trusted with their own personal data. “You get to learn about us… so it seems both intuitive and only fair that we should get to learn what you’ve learned, however uncertain,” wrote Misha Angrist from Duke University, who was among the first people to have his genome sequenced and made public.

“But returning results is predicated on the notion that the information will be transmitted securely, and without anyone else having access to it unless individuals choose to share it,” he says. If employers can demand those results, the debate changes. “I’m less sanguine about the unfettered sharing of that information. If I’m conducting a research study, what can I say to my participants about the legal protections that I can afford them?”

To be clear, the stakes aren’t about any particular study, or any scientist’s career. If people are put off from taking part in research, medical progress will slow. That would be particularly disastrous for rare genetic diseases, where research into cures often depends on finding enough patients who share the same disorder. “Anything we can do to get clinical trials off the ground is a complete non-starter without participants,” says Sonia Vallabh from the Broad Institute

Five years ago, Vallabh learned that she has a lethal genetic disorder called fatal familial insomnia, inherited from her mother. “I decided that I was willing to take the gamble of putting my privacy aside, but I vividly remember the early days when that decision was desperately unclear,” he says. “We now face such an uphill battle in helping other families figure out whether they want to be tested. Even with GINA solidly in place, when someone asks if I can guarantee that their info will never be used against them, I’m not able to say yes. I’m so worried about people getting scared off.”

The erosion of GINA also comes at a time when researchers are actively trying to recruit more people from underrepresented minorities. People of European descent still account for 80 percent of the participants in large genetic studies, while people of African or Latin American ancestry, Hispanic people, and native or indigenous peoples collectively account for just 4 percent. Many studies are trying to rectify this imbalance. Eric Dishman, the director of the All of Us cohort, said in a recent conference talk that they would actively “overrecruit the understudied,” including ethnic minorities, people from rural areas, and those from lower socioeconomic groups. It’s an laudable goal. But perversely, those are the same groups who stand to lose the most from H.R. 1313.

“People are already being discriminated genetically without mandatory genetic testing—that’s what sexism and racism is,” says Holly Dunsworth, an anthropologist from the University of Rhode Island. “This would result in a higher resolution of discrimination and more individualized ways of targeting people.” The dark mirror of personalized medicine is personalized discrimination.

For example, in 2001, the Burlington Northern Santa Fe (BNSF) Railroad was successfully sued for covertly testing its employees for a genetic condition that causes carpal tunnel syndrome. Last year, two Californian parents sued a school district for pulling their son out of class because he tested positive for a cystic fibrosis gene—but didn’t actually have the disease. The Council for Responsible Genetics has documented many similar cases. A boy was denied health insurance for carrying a gene that predisposes him to heart disease. A woman was fired after revealing that her mother died of Huntington’s disease, meaning that she has a 50 percent risk of developing the condition. In a post-GINA world, such cases would only become more common.

“It potentially opens up a perfect storm of employment discrimination, health discrimination, and racial discrimination, because genetic data holds inferences of all of these at once,” adds Nelson.

And for what? There’s little evidence to suggest that workplace wellness programs actually work, much less that adding genetic information to them will accomplish anything. And such information is difficult to interpret. A tiny number of genetic variants (like the one Vallabh has) will definitively and deterministically lead to disease; the vast majority of them affect our health in subtle and probabilistic ways. “There’s plenty of genetic information whose predictive value is ambiguous, low, highly contextual, and poorly understood,” says Vallabh. “We’d be opening the door to a lot of speculation and poorly grounded discriminatory conclusions.”

Holly Dunsworth agrees. She has used 23andme tests as teaching tools in her classes, and she has seen students take up jogging based on an incrementally higher genetic risk of heart disease. “I can’t imagine if someone else was looking at these data for me and making decisions about my healthcare and making judgements about my lifestyle, on what are kind of arbitrary numbers,” she says.

No one knows if H.R. 1313 will actually become law. The bill has already triggered opposition from both the American Society for Human Genetics and the Wellness Council of America. Even Tom Price, the new secretary of Health and Human Services, said that “it sounds like there would be some significant concerns about it.”

“It’s my hope that just having this bill in the house will force the important conversations we need to have about how society should be thinking about genetic information,” says Wagner.

* Update: This article has been updated to clarify H.R. 1313’s impact onGINA, and to incorporate a statement from the House Committee on Education and the Workforce.