When IRBs and investigators deal with risks, they should identify reasonably foreseeable risks and have reasonable precautions to prevent harm.
Risks are the cornerstone of human subjects protection laws.
In fact, when you distill federal regulations regarding human research subject protection, it can be described with just three sentences, says Ernest Prentice, PhD, associate vice chancellor for academic affairs at the University of Nebraska Medical Center in Omaha. Prentice is the former chair of the Secretary’s Advisory Committee on Human Research Protection (SACHRP) for the U.S. Department of Health and Human Services.
“One would be that the risks to subjects are minimized; two, that the risks to subjects are reasonable, and three, that there is adequate provision for monitoring the safety of subjects,” Prentice says.
So if an IRB plans to revisit policies and procedures (P&Ps) regarding protocol review, they might want to revise or at least revisit their language related to risk.
“The common interpretation of a risk is that it’s a potential harm,” Prentice says. “But you need to look at it more narrowly.”
For instance, every medication has a list of side effects, and each of those is a potential harm, he says.
“Those side effects are all going to occur, and every time you pick up a prescription, you get an insert that lists all of those risks,” he adds.
From a research perspective, the risks described should be reasonably foreseeable, which might not mean they are possibilities, but that they are likely to occur in at least some participants, Prentice explains.
The IRB’s role is to make certain investigators have minimized those risks as much as possible, he adds.
“Take a procedure like venipuncture, which everyone has had at one time or another,” Prentice says. “What are the known potential harms? One, it will hurt; it’s painful. The second risk — depending on the size of the needle and difficulty of the phlebotomist — is that you might get some bruising.”
Other risks are rare, such as obtaining an infection or fainting from the procedure, he says. “Someone could faint. Someone could faint, fall out of a chair, crack his head, and suffer a concussion.”
While that scenario is possible, is it something that needs to be listed on the research informed consent form, Prentice says.
“We don’t put that stuff in there because in a research context, we have to narrow down our considerations of risks to those that are reasonably foreseeable,” he adds.
Risks also are situational and population-dependent.
“Some subjects are more vulnerable to a risk occurring than others,” Prentice says. “For instance, the risk of infection with a simple venipuncture: If you’re immunocompromised, there’s a greater chance of infection. And hemophiliacs have a greater chance of bleeding and bruising from the procedure.”
People who are elderly and frail are at greater physical risk in medical studies, he adds.
“When an IRB reviews risks, they have to look at not only the intervention, but the risk susceptibility of the subject population,” Prentice explains.
Also, when IRBs analyze risk, the questions they should always have in mind are: What can we do to minimize risk? What kind of precautions can we put in place to safeguard subjects?
“If you want to totally eliminate risk, then don’t approve the research project,” Prentice says. “But we’re not going to do that, so what we have to do instead is come up with reasonable precautions.”
The following are ways to develop reasonable precautions:
• Investigators start by thinking about subject safety. Investigators are the first to ask: What do I need to do to minimize risk?
“For example, and the regulations speak to this, can I go ahead and utilize an already scheduled clinical procedure and use data from that procedure?” Prentice suggests. “So if a subject is going to have a CT scan, can I use that CT scan, which is done for clinical purposes, rather than have another one done for research purposes?”
Subject safety should be a consideration when schedule procedures that might be duplicates of procedures conducted for clinical purposes, he adds. “Any time you can use information from a procedure already done for clinical purposes, do it.”
• Assess the monitoring plan. IRBs and human research protection programs (HRPPs) need to ask what sort of monitoring is necessary and what sort of follow-up procedures are necessary, Prentice says.
“If a subject experiences a certain level of symptoms, then maybe he should be withdrawn from research before he suffers irreversible harm,” he says.
IRBs monitor and analyze risk, but it’s up to investigators to list and clearly describe the known, reasonably foreseeable risks, Prentice says.
“I would not suggest they put in 10 pages of risks, which has become a problem with consent forms these days,” he says.
• Identify, define reasonably foreseeable risks. “Take reasonable precautions to prevent harm from happening,” Prentice says. “By reasonable precautions, you decide what’s reasonable, and if it’s not reasonable, then we shut down the research.”
Investigators decide which risks are reasonable, and IRBs decide whether the principal investigator’s description of risks is reasonable, he says.
“If subjects are more vulnerable, including elderly people, people who are chronically ill, including people with terminal cancer, then look at all these things and make an assessment of the risk,” Prentice says.
• Consider potential benefits. Benefits in research are anything with a positive value, Prentice notes.
“It’s something that may accrue to directly benefit the subject, or perhaps it will only benefit society in terms of advancement of knowledge,” he says. “All research must have benefit to society, but not all will have benefit to the subject.”
For example, Phase I in clinical trials involves toxicity testing, and there’s typically no prospect of therapeutic benefit to subjects, Prentice says.
“IRBs have to factor that in,” he says. “Does the investigator think there’s any prospect of direct subject benefit and, if so, is it worth it? Or, if not, why is the investigator doing research in the first place? How will it advance knowledge?”
A flawed experimental design won’t yield usable data, so it would be unethical to conduct it, Prentice says.
“The principle investigator has to make sure the experimental design is sound, and the IRB has to say, ‘Yes, right design, and if it’s carried out according to the protocol, it could potentially result in advancement of knowledge,’” Prentice adds.
• Think about risk-benefit equation. “We have reasonable precautions placed in the protocol to reduce the possibility of any subject experiencing harm,” he says. “We’ve examined the anticipated benefit to society, and now it’s time to look at the risk-benefit relationship.”
IRBs might consider the possibility of risks outweighing benefits and whether it’s ethical to do research where this occurs, Prentice says.
In situations in which there is the possibility of subjects experiencing direct benefit from research, then IRBs might find it acceptable to have a higher level of risk. They will make this assessment as they consider the risk-benefit equation, he explains.