Gutting research integrity agencies won’t eliminate fraud at universities

President Trump’s administration has launched a series of attacks on American higher education institutions, many of them of dubiouslegality and tainted by politics. It’s quite ironic, then, that in one area where universities really would merit scrutiny, the administration is scaling back the government’s efforts. If Trump were serious about cracking down on waste, fraud, and abuse at universities, he would be strengthening the federal oversight bodies that ensure scientific integrity in research. Instead, Trump has gutted them.

A federal role is crucial because universities can’t be solely responsible for policing themselves. As Ivan Oransky, cofounder of the Center for Scientific Integrity, told the editorial board, universities have little incentive to seek out fraud. “Until someone external actually has power and authority to deal with it, putting universities in charge of figuring out whether there’s misconduct, fraud, or real substantial risk to human beings isn’t doing the job,” Oransky said.

But under Trump, oversight agencies are being decimated. The US Department of Health and Human Services’ Office for Human Research Protections, which oversees federally funded research involving human subjects, has long been underfunded and understaffed. Trump fired its director, laid off staff, and dismantled an advisory committee to the secretary on human research protections, according to reports published by Stat and the Health Care Compliance Association.

Retraction Watch reported that top staff and investigators at the National Science Foundation’s Office of Inspector General, which oversees that foundation’s grants, were fired or left their jobs after Trump took office. Also gone are leaders and investigators at the HHS Office of Research Integrity, which investigates research falsification, fabrication, and plagiarism on behalf of multiple federal agencies.

There may be ways to improve federal oversight, like having government rely less on research institutions to conduct initial investigations. But even when federal oversight is robust, it leaves gaps. For example, there’s no mechanism to oversee nonfederally funded research that doesn’t relate to drugs or medical devices.

Ivor Pritchard, a retired senior advisor at the Office for Human Research Protections, said it simply isn’t possible for government to monitor labs’ daily work: “From a practical standpoint, scientific activity has to be self-correcting or self-monitoring, although external oversight mechanisms are also crucial.”

So while federal oversight is necessary, it is insufficient to ensure that researchers police themselves. To truly improve research integrity, universities themselves must get serious about oversight. Universities, research journals, institutional review boards, and funders all have roles to play in changing incentive structures so that getting research done correctly becomes more important than getting something published or into clinical trials quickly.

Jonathan Kimmelman, associate professor of biomedical ethics at McGill University, told the editorial board that the problem is cultural: When considering hiring, tenure, and promotions, universities prize innovative research. So do prestigious scientific journals deciding what to publish. They don’t value studies aimed at confirming or disproving existing research. “There are practices that can put the brakes on misconduct that we don’t reward well,” Kimmelman said. “The result of publishing negative studies or committing yourself to checking integrity is you get penalized.”

Luckily, there are those in the industry with power to change the culture — if they choose to. Kimmelman suggested medical and scientific journals could encourage or require researchers to preregister their research studies, publicly stating the hypothesis of an experiment before they conduct it. Both research institutions and journals could reward practices like inputting data into a public database so others can analyze it. Universities and journals need ways for whistleblowers to safely and anonymously report problems with research or published studies, including mechanisms to ensure thorough investigation.

Universities could do better at giving scientists credit, including promotions, for research that disproves prior studies. Journals could provide a more robust market for their publication. Kimmelman said funders that condition grants on starting clinical trials also contribute to an atmosphere where products move into clinical trials too quickly, without research being double-checked.

Institutional review boards — which approve and monitor research studies — could be made more rigorous. One former federal regulator told the editorial board that university-based boards today are often staffed by volunteers who don’t have time to thoroughly review and monitor every study. Giving professors time built into their schedule for board work could help address that.

There are also conflicts inherent in review boards funded by organizations they regulate. Many independent review boards today are for-profit entities owned by private equity investors. Stephen Rosenfeld, who founded a nonprofit independent review board called the North Star Review Board, said many boards are paid high prices by pharmaceutical companies, whose interest is in having the board do the minimal review necessary to ensure compliance. Rosenfeld has suggested that publicly funded nonprofit review boards are a better model to ensure boards prioritize the interests of human research subjects rather than the companies paying for the review. “You have to take out the structural conflicts,” Rosenfeld said.

In an era when science is increasingly viewed with skepticism, if universities, researchers, and regulators want to maintain the public trust, they need to ensure science is being done with rigorous methods and integrity.