The lines between research and medical innovation can be blurry. When does a new surgical practice cross from case study to a study that must adhere to human research protection regulations?

“This issue has hit our IRB,” says Stephanie Solomon Cargill, PhD, associate professor at the Center for Health Care Ethics at Saint Louis (MO) University.

“We’ve had several situations where we get minimal risk, retrospective studies coming into the IRB, and they were doing pretty risky and outside-of-standard-of-care procedures,” Cargill says. “But they were doing these studies under the auspices of medical innovation.”

When physicians say their procedure is medical innovation and then they later want to evaluate their results retrospectively, they are not scrutinized by an IRB because it is a retrospective study, Cargill says.

“A lot of times when retrospective studies come to IRBs, they don’t make a big issue out of it because it’s minimal risk,” she says. “But it’s unclear how an IRB would know if medical innovation were happening clinically.”

For example, if a researcher tells a journal that the submitted paper describes a prospective study, but the researcher received approval from the IRB for a retrospective study, then this is a discrepancy that needs to be resolved, she says.

IRBs do not want to be too restrictive, forbidding physicians from going outside the standard of care and engaging in medical innovations. But they also do not want researchers to take advantage of IRBs by submitting studies that clearly should have been subject to full board review as retrospective studies, Cargill says.

Those kinds of cases show that IRBs should have policies and procedures in place that clearly state how and when researchers can use medical innovation and retrospective studies when sending their protocols to the IRBs.

“We’ve reviewed literature of surgical innovation, and we’ve noticed it happening in radiology and other areas that are not just surgeries,” she says.

If a physician has performed an innovative procedure once or twice, then it is feasible and legitimate to consider this a medical innovation. The physician could write these examples as retrospective review or case studies that do not require IRB review.

Suppose the physician performed the medical innovation with multiple patients over a long period of time and wants to publish results as though these were clinical research outcomes. This is different and probably should not be submitted to the IRB as a retrospective study. Instead, the physician should submit the procedure for IRB review prospectively, Cargill says.

“Can we distinguish between having a legitimate space for clinicians to work things out and then saying, ‘Hey, let’s write this up as a study,’ as opposed to their thinking, ‘We know we want to get published for this, but we don’t want to go through the study and tell patients what we’re going to do,’” she explains.

Clinicians might think their medical innovations are low-risk when there is very little evidence for that view. These procedures are not endorsed by their profession, she says.

For instance, surgeons can access surgery innovation committees before performing a procedure that is further outside standard of care. This provides checks and balances.

“There should be some extra level of oversight,” Cargill says.

IRBs need to have some guidance on procedures performed for clinical purposes, benefiting a patient, or for research purposes when it might establish the effectiveness of a procedure, she adds.

The goal is to develop documents that help clinicians assess whether their procedures constitute medical innovation or research, Cargill says.

“We’re a working group with members that include a general counsel, a surgeon, an IRB staff person, and someone from compliance,” she says. “We sat down together to work on coming up with policies and recommendations that work for our institution.”

“We give the guidance to department chairs, telling them that if someone in their department is doing those procedures, then here are the questions they should ask them,” she adds.

“There are two paths: medical innovation path, which doesn’t have to be research, but there needs to be specific informed consent, and then there’s the research path.”

When an institution writes a policy to distinguish between medical innovation and research, the policy should be promoted as a way to enhance institutional and clinician legal protection and provide human research protection.

“Without any kind of oversight, if something goes wrong with a medical innovation and a patient sues, there’s no malpractice protection,” she says. “So it’s about whether they are consenting patients so patients know what they are getting into.”

Also, clinicians should have limits placed on what types of studies they can publish without proactive IRB review.

“We’ve discussed requiring any publication of results to be limited to structured case reports, proof of concept articles, or retrospective case studies — if under five cases,” Cargill says.

The goal is to not make patients unwitting research participants. “If it’s being sold to patients as something that is the best thing for them, and ‘I’m acting as your clinician,’ that’s a big ethical question,” she explains.