IRBs have wide range of policies about AEs
IRBs often are inundated with hundreds of adverse event (AE) reports that are not always useful to their mission of protecting human subjects. However, some IRBs are hesitant to make policies that would limit these reports because of a fear of missing important information. Others find the federal regulations about reporting AEs to be ambiguous and somewhat confusing.
"The federal regulations are very general," says Kathleen Gifford, RN, a research compliance coordinator with Penn State Milton S. Hershey Medical Center, Penn State College of Medicine in Hershey, PA. "Serious or unexpected adverse events will be reviewed by the IRB, but then you must define serious and unexpected," she says.
The U.S. Food and Drug Administration (FDA) regulations do not require sponsors to send AEs involving drugs to IRBs, and IRBs don’t have to accept AE reports of drugs, says Jeffrey A. Cooper, MD, a research expert and physician in Washington, DC. However, the regulations do require reporting of unanticipated adverse device effects, and federal regulations do oblige IRBs to review unanticipated problems that involve risks to human subjects, he adds.
"IRBs are being inundated with AEs because the sponsors are sending all their AE reports to investigators and asking them to pass those on to IRBs," Cooper explains. "My sense is that many of those AEs are clearly unrelated or the AEs are expected." For example, a common complication involving patients infected with HIV is toxoplasmosis, an infection that takes advantage of an immunosuppressed body. Sometimes during HIV clinical trials, patients will develop the infection, he says. So every time an HIV patient got toxoplasmosis, a serious AE report would be filed. "When I was working with an IRB, we saw a lot of serious complications of AIDS that were expected complications being reported as adverse events," Cooper reports.
"I think IRBs are struggling with what to do with these reports," he says, adding that according to federal regulations, the anticipated AEs, such as toxoplasmosis, do not need to be reported. "IRBs should have procedures in place to get reports of unanticipated problems of risk to subjects or others. My message is to get people to look at AE reports as this issue of unanticipated problems of risk," Cooper says. "IRBs would meet 24 hours a day, seven days a week, if they reviewed every AE they got."
While some larger IRBs and research institutions have decided to review only AEs that occur within their own institution, others will review AEs reported during outside studies that use the same or closely related protocols, Gifford says. "We let the principal investigator [PI] determine whether it’s closely related, and we’ve opted to look at all serious and unexpected AEs that are closely related," she adds. "If it’s listed on the consent form, it’s expected, and if not, then it’s unexpected."
However, the institution doesn’t waste staff and IRB time on all of the unexpected AEs reported by outside studies. "We’re not looking, for example, at cold symptoms in a patient just because they’re not listed on the consent form," Gifford says. "That’s unless the cold symptoms happen routinely, and we have a category of trends of AEs that are not truly serious but happen so consistently that you start to get suspicious about them."
The research compliance office at Hershey Medical Center assists three separate IRBs and about 1,200 active protocols. The staff include three research compliance coordinators, an administrative assistant, and five staff assistants who are well trained in IRB administration.
Some IRBs and institutions have policies that take the decision of what is a reportable AE away from the PI and require that IRBs review all AEs. For example, the Edward Hines Jr. VA Hospital in Hines, IL, research service requires that all adverse events be reported, says Della Herzog, RN, human studies subcommittee coordinator. "Our IRB wanted to make the judgment of whether or not an AE is related," she explains. "We do expect the investigator to give his assessment, but there are times when we change the assessment and require him to call it a related situation."
For example, there have been incidents when PIs did not consider a series of AEs to be related to the study, but because there was a trend of serious AEs, the study was closed, Herzog says. "Now it’s a rare occasion when that happens," Herzog adds. "So whether this continues to be worth the additional workload is up in the air." Herzog says she spends a day and a half each month preparing AEs, and this includes assistance from secretarial staff who enter the information in the computer database.
At Hershey Medical Center, about eight hours a week is spent on handling adverse events, Gifford says. The IRB’s AE workload is high partly because many reported AEs come from outside studies. IRBs must determine whether to review only AEs that occur at the IRB’s institution or whether the board also will review all AEs that occurred at other institutions during identical or closely related protocols. "Some of those related protocols are so slightly related that you wonder why in the world you’re looking at those AEs at all," Gifford says.
These dilemmas may be more readily solved if the IRB clearly defines for investigators reportable AEs. For instance, the Catholic Medical Center in Manchester, NH, requires all unexpected adverse events that had not been previously observed or included in the informed consent process to be reported, says Donna L. Bennett, RN, MS, CIM, vice president of research and programs development. "They may be minor in which no severity is anticipated, but they were unexpected," she explains.
Severe AEs are defined as those that result in death, a life-threatening situation, hospitalization, prolonged hospitalization, persistent or significant disability, a substantial disruption of one’s ability to conduct a normal life function, a birth defect, and cancer. Serious AEs also can be a result of an overdose or protocol error, Bennett says. The hospital’s IRB has a three-page IRB report in which investigators discuss details of the protocol and AE.
While the Catholic Medical Center’s IRB can depend on its investigators to review AE reporting policies and procedures, there also are AEs being reported from outside studies, and these can take considerable time to review, Bennett notes. "Many of the industry sponsors are requiring IRBs to review AEs occurring in the same drug, but maybe in a different study," she says. "For example, there may be a cholesterol-lowering drug that may be involved in five different studies, all from the same pharmaceutical sponsor."
When an adverse event occurs at a study being conducted in France, for instance, the sponsor may want to make sure all IRBs and investigators are aware the drug has had the adverse event, Bennett explains. "We had one protocol where the investigator submitted a letter and a list of 40 adverse events from a different study," Bennett says. "The sponsor wanted the IRB to know these results occurred but did not provide the IRB with the criteria of the research study nor the number of participants involved."
This type of report sometimes can raise more questions than it answers, so even if the IRB wants the information, the report may not be useful, IRB coordinators say. The problem is that drug safety monitoring boards (DSMBs) have a wide variation in reporting and monitoring practices, and this is part of the reason that some IRBs, such as the one at Hines VA Hospital, require all AEs to be reported, Herzog says.
"We’re still not satisfied with the reporting coming back from some DSMBs — we frequently have to say we need more information," she says. "Certainly as DSMBs become more attune to reviewing adverse events and as we get better information back from DSMBs, the IRB will be more satisfied and start reviewing fewer AEs."
IRBs need the DSMB reports to provide perspective with more detailed listings of the types and frequencies of the AEs being reported, Herzog notes. "A lot of them just say they reviewed the AEs and no recommendations were made. We’re not satisfied with that answer," she adds. "So we have maintained our policy to review all of them."
Herzog is hopeful that DSMB reports will improve as they begin to follow new policies developed by the National Institutes of Health (NIH) and the FDA. "There are more specific criteria now described by the FDA and NIH of what they expect out of DSMBs, so that will help us a great deal once DSMBs start providing us with that information," she says. "It might help reduce some of the workload."