When Stephen Rosenfeld and Patricia Seymour founded the not-for-profit North Star Review Board, they’d become disillusioned with the scientific oversight system in America. IRBs, or institutional review boards, are supposed to have one priority top of mind: the rights and welfare of human research participants. But after decades as research ethicists, Rosenfeld and Seymour thought the industry had lost its way.

The problem, they said, isn’t just theoretical. Over the last couple of decades, IRB scandals have cast a spotlight on a fundamental problem in the system, experts said. In one such example, the Government Accountability Office in 2009 tested whether a group of for-profit IRBs could successfully detect a bogus unethical research protocol and reject it. One of the three companies tested, Coast IRB in Colorado, approved the fictitious application, even though the fake device had specifications and matched several examples of “significant risk” per FDA guidance. The sting operation revealed what the GAO called a vulnerability to “unethical manipulation” in the IRB system.

Though Coast was shut down, experts told STAT that major reform didn’t follow that scandal or others. The central problem in the system persisted, ethicists said.

“I felt that the people running IRBs should be focused on its oversight and ethical mission, not necessarily investor returns,” Rosenfeld said. “It’s very hard to figure out how to do this review, pay for it, and avoid structural conflicts of interest. It’s not about people doing bad things, but how the system is set up.”

Before any research can get started, it has to first pass ethical review. IRBs are meant to be a safety net for research participants: an assurance they can take part in science without the fear of exploitation or intentional harm. For the most part, experts said, these boards cover major, particularly physical, harms. But ethical oversight of science has also evolved over the years to prioritize facilitating speed and meeting legal requirements, rather than ensuring what might be best for research participants.

That’s mainly because of financial interests and outdated regulations, experts told STAT, which has now left holes in the IRB system where science that is both legally compliant and ethically dubious can proceed. Those concerns are increasing now that research institutions are experiencing historic financial instability and fractured trust between researchers and the federal government.

“The pressures are not new, they’re just intensified,” said Holly Fernandez Lynch, an associate professor of medical ethics at the University of Pennsylvania. “What we see now, sometimes, is a customer service model of the IRB, where they view the investigator as the customer. It’s about turnaround time, and how can we do this faster. It becomes easy to say, we’ve got this set of regulatory requirements, and so we will force the IRB into a state where they approve research where it’s minimally legally acceptable.”

That can be a major problem, particularly as science and technology progress and create new participant protection concerns, like privacy or data exploitation, that current regulation doesn’t adequately address, Rosenfeld said. There’s a widening gap between the standards set by the law and the highest ethical standards for research.

“And what’s at stake is trust in science. People who need or want to enroll in trials or fill surveys, they should be able to do that without the fear that they are providing information that will be exploited,” he said.

He thinks creating a system of independent nonprofit IRBs is a step toward more ethical research oversight. To him and Seymour, North Star could lead the way for that change.

NIH expansion carried ‘hot potato’ liability

The groundwork for the current IRB model, where research institutions and companies have their own local IRBs, was laid in the mid-20th-century. Much of what shaped that system of ethical research, said Laura Stark, a science historian at Vanderbilt University and the author of “Behind Closed Doors: IRBs and the Making of Ethical Research,” was actually a desire to pass the buck.

The National Institutes of Health had expanded rapidly under the leadership of director James Shannon, who served from 1955 to 1968 and oversaw the growth of the NIH budget from $65 million to over $1 billion. This growth also came with a risk. If the NIH could be held fiscally responsible for unethical behavior of researchers in its now vast extramural program, that could be a problem for the agency.

“You can think of liability being a hot potato,” Stark said, and the federal government wanted to pass it off just as much as anyone else.

On top of that, research organizations and leaders were strongly against any kind of external regulation, Stark said, and resisted the establishment of IRBs in the first place. Instead, they created industry guidelines that scientists could choose to follow. “Professional associations were writing ethics codes to do something the AI industry is doing now,” she said. “Creating the equivalent of voluntary agreement as in ‘These are the rules we will follow. We don’t need regulation. We’re the experts.’”

A tipping point came after the revelation of the Tuskegee syphilis study, where members of the U.S. Public Health Service deliberately withheld syphilis treatment to Black men to study the illness. Lawmakers had the choice of going down one of two paths, Stark said. One option was creating a single, federal IRB. Though, the problem there was that liability would land with the federal government.

Instead, Stark said, “the federal government was strong enough with the hot potato issue of ‘we’re not taking this on.’” Thus, when the National Research Act of 1974 became law, requirements for ethics reviews leaned on local IRBs and saddled local universities and research organizations with the financial and legal risks of reviewing research protocols.

The continuation of that, Stark said, was the creation of commercial IRBs, like the private-equity owned Advarra and WCG. “The reason why private boards are really advantageous is that they can also shift liability away from the university setting,” Stark said. “They’re paying for someone not only to do the work of review and hire staff; the underlying advantage is not having the financial risk of liability if something goes wrong.”

Commercial IRBs pushed back on this characterization. Representatives from Advarra told STAT that independent IRBs increase the quality and efficiency of ethics reviews by having a professional, expert workforce. “Irrespective of the modality, academic, not-for-profit, commercial IRBs, we’re all driven by the same mission, which is to safeguard the rights and well-being of research participants,” Gadi Saarony, the CEO of Advarra, told STAT. Advarra also said that all IRBs are held to the same regulatory standards.

There have been high-profile cases of IRB failures that resulted in settlements, said North Star’s Rosenfeld. He brought up two examples. In 2001, Johns Hopkins University settled with the family of Ellen Roche after she died in a trial testing an asthma medication, and in 2000 the University of Pennsylvania settled with the family of Jesse Gelsinger after he died in a clinical trial for gene therapy. Rosenfeld noted that such high-profile settlements reinforced the IRB system’s focus on compliance.

‘It’s basically an honor system’

Many IRB experts and ethicists said that this model created in the 20th century was a mistake. Placing the regulatory and ethical oversight of research within the same research institution automatically introduces an inherent conflict of interest. On top of that, IRBs have little serious oversight themselves.

“The entire way our research oversight system is set up is crazy,” said Carl Elliott, an ethicist at the University of Minnesota and author of “The Occasional Human Sacrifice.” “It’s basically an honor system. Oversight bodies, FDA, OHRP, they basically just trust researchers to do the right thing and be honest. That’s a crazy way to oversee this multibillion-dollar, multinational, increasingly for-profit research enterprise.”

Universities rely on research to bring in grant dollars, and pharmaceutical companies need to develop drugs. Delays run counter to institutions’ operating budgets, and red tape in the form of regulatory boards like IRBs gets in the way of that. Also, local IRBs are made up of the researchers’ colleagues, typically other employees or faculty members, creating personal conflicts of interest.

That all leads to an “implicit pressure,” Elliott said, for the ethics boards to ensure that reviews pass quickly. That can influence how the review gets done and the outcome, he said, regardless of how well-intentioned, competent, and professional the reviewers are. “A subtle pressure of, well let’s make sure these things are safe, but let’s really make sure the research is done. We’re on a noble mission, and research is a good thing,” he said. “At least one IRB I’ve been on, they rarely turn down a protocol because it’s unethical. They work and work and work to get it through somehow, compromise some way.”

Elliott became a whistleblower in the case of Dan Markingson, who killed himself after enrolling in an antipsychotic drug trial at the University of Minnesota.

For-profit IRBs face a similar conflict of interest, Elliott said. “They are being paid by the corporations they are supposedly overseeing,” he said.

To show how this combination of conditions can go awry, Elliott refers to the SFBC International scandal in 2006. “A contract research organization was recruiting undocumented immigrants to do Phase 1 trials, which was bad enough, but they were doing it in a condemned motel,” Elliott said. “All of their trials were going through an IRB.” That motel was later deemed a fire hazard by Miami-Dade County.

“This was a reputable company, allegedly. How could it be that you are recruiting undocumented immigrants in a condemned motel for 10 years with nobody catching that?” Elliott said. “They’re doing the study in Florida, reviewing it in Washington, and nobody sets foot in the trial site. If we regulated these contract organizations the same way we regulate restaurants, you’d have some kind of inspector coming in.”

After the scandal made headlines, Elliott said, the regulations still weren’t updated to prevent similar situations from arising. “That is the pattern in general. Not only does nothing change, usually the people behind scandals get promoted or celebrated or win awards,” he said.

Increasingly, experts told STAT that there are also new ethical concerns rising in human research that regulations don’t cover. “In the last decade and a half, increasingly, the harms are informational. Privacy, exploitation of data, you can do almost anything in those realms that are IRB exempt because it doesn’t involve physical harms,” North Star’s Rosenfeld said.

There is an organization that provides accreditations to IRBs called AAHRPP, or the Association for Accreditation of Human Research Protection Programs, and there are some institutional policies that do “go above and beyond the regulations,” Penn’s Fernandez Lynch said. However, she said her theory is that the current system generally punishes reviewers that try to meet the highest ethical standards, and rewards reviewers that prioritize speed and meeting legal minimums.

“If you have IRBs going above and beyond, you don’t have a lot of reasons as a customer to stay with them,” Fernandez Lynch said. She added that universities face the same pressures because they need to keep research funds flowing. “If you care about building trust in research, that might fall by the wayside.”

Removing the profit motive

Rosenfeld and Seymour founded North Star Review Board because they hoped taking profit out of the equation would lead to a better system. Both had worked for large commercial IRBs in the past, and felt that the profit motive undermined the mission. Seymour said she remembered two key moments that drove her away from the for-profit industry.

“An IRB I worked for, a big for-profit, they were so happy when Covid hit. I won’t point fingers, but in the first quarter, they made their entire budget because of Covid. They were crowing about it,” Seymour said.

Another time, a team of investigators said they needed to modify the research protocol, but they didn’t have the budget for the required additional review. Seymour said this was putting profit over supporting the scientific mission. “They would call me and say, ‘My budget’s blown because I get charged for every single thing that I ask the IRB and now I have a safety issue, and I can’t make the change,’” she said. “Those are the kinds of things that made me really upset.”

Rosenfeld contacted her over five years ago, and proposed starting a nonprofit together. “A nonprofit IRB and take care of the regulatory issues and be ethical and take care of patients,” she said. “I was just thrilled.” To her and Rosenfeld, North Star felt like an opportunity to do things differently. They wanted the board to engage with the research and involve the investigators more deeply, rather than going through a regulatory checklist. It may take more time, but Seymour and Rosenfeld see it as a way to ensure more ethical science.

But North Star still collects fees and competes in a broader IRB marketplace. Ethicists like Minnesota’s Elliott argued that leaves them just as exposed to the same systemic conflicts of interests. Seymour pushed back on this, saying that she and Rosenfeld both receive “minimal” salaries, don’t own the organization themselves, and charge only a flat annual fee. Rosenfeld said that he felt the nonprofit status of North Star was a step toward a more ethical system.

Rosenfeld said an ideal system would be one in which the public is the primary stakeholder in research oversight. One example he offered was a system similar to how the Food and Drug Administration reviews applications, where user fees pay for ethics reviews. “You distribute those through a government entity that has measures of quality and trust to represent the interests of the members of a pluralistic democracy. That’s how it should’ve been set up in the first place,” he said. “We don’t have that now. It’s really problematic.”

Experts told STAT the research oversight system needs an overhaul, one that would redirect the flow of dollars paying for ethics review at best and rework of regulatory requirements at minimum. However, the Office for Human Research Protections, the federal body that regulates IRBs and research oversight, recently lost over half its staff under the Trump administration because of reductions in force, resignations, or early retirements.

“This isn’t a time I would consider redoing the regulations. We all know where that would go,” Rosenfeld said.

Until a proper overhaul can be done, Rosenfeld hopes that North Star, by removing the profit motive, can make space for more moral science. “IRBs were established to serve the research participants and the public,” he said. “In every existing market model right now, that customer has no voice.”

STAT’s coverage of bioethics is supported by a grant from the Greenwall Foundation and the Boston Foundation. Our financial supporters are not involved in any decisions about our journalism.