AE protocol helps clarify what should be reported

Reporting rules spelled out for PIs

While each IRB has its own policies and procedures for reporting adverse events (AEs), there are some strategies that appear to make it easier for investigators to decide what should be reported and simultaneously keep a check on the AE workload for IRB members. At the same time, IRBs should keep in mind that not all AEs involve a physical or medical problem; therefore, their policies also should address AEs related to psychological, financial, and social harms.

"Some people are focusing on AE drug events and are missing other issues that involve unanticipated problems, which should be handled in the same way," says Jeffrey A. Cooper, MD, of Washington, DC. For example, a protocol may involve genetic research that, in the course of study, discovers a problem with a particular subject’s paternity, he says. "If investigators disclose that information it might cause problems when a person finds out that the person who he thought was his father is not his biological father," Cooper says. "This situation involves unanticipated risk, although there are no drugs, biologics, or devices involved."

An investigator may fail to file an AE report on such an incident, so it’s the IRB’s job to have some sort of process established to deal with these types of unexpected adverse events, he states. Likewise, there may be an AE that results in potential, if not actual physical harm. For instance, there could be a study in which an investigator is monitoring a subject’s drug levels only to discover that the lab has made a mistake in its analyses and the last 20 drug levels were incorrect, Cooper says. "This needs to be reported to the IRB as an unanticipated event, a failure of monitoring procedure and not an adverse drug event," he adds. "Think about risk — psychologically, economically, and socially," Cooper says. "IRBs should consider what additional steps should be put in place to minimize those risks and to inform participants."

A prime example of this sort of outside-the-box thinking involves how an IRB might handle an AE that sometimes occurs with participants of HIV vaccination trials, says Kathleen Gifford, RN, a research compliance coordinator with Penn State Milton S. Hershey Medical Center, Penn State College of Medicine in Hershey, PA. "People are going into AIDS vaccine trials where they receive vaccinations for a period of time, and it will appear on standard HIV tests that they’re HIV-positive," Gifford says. "More sophisticated blood tests tell you that they’re not HIV-positive."

However, these subjects could unwittingly be tested as part of a life insurance exam and then denied coverage based on the flawed findings. When HIV researchers and IRBs began to realize this AE could occur, they were able to react with safeguards in the protocols, Cooper says. "We had a trial like that and had participants issued cards from the drug company that explained their participation in the trial," he explains. "The cards said they’d be HIV-positive for antibodies, and they would need to be tested for HIV in a different way. Participants were told they’d be given this card, but it would not limit their ability in the future to get life insurance and other insurances."

So what began as an unanticipated AE became anticipated and properly handled to better ensure subject safety. Here are some other issues and strategies that IRBs need to consider when revising AE reporting policies and procedures:

Make AE reporting guidelines and education easily accessible.

All of the Hershey Medical Center’s adverse events instructions and reporting forms are available on the center’s web site at www.hmc.psu.edu/hmc-irb/downloads.htm. Investigators can easily download instructions, forms, and reporting logs that define what needs to be reported, Gifford says.

The Edward Hines Jr. VA Hospital in Hines, IL, research service has a guidelines packet and a large coordinators group that has helped to communicate the AE policies to staff, says Della Herzog, RN, human studies subcommittee coordinator. "Our coordinators group — and most, but not all, are nurses — has been a major way of getting our policies across," Herzog says. "But we also have all-day educational sessions, such as a good clinical practice conference that we held last December."

Whenever there are new rules or regulations, investigators and physicians also are notified through e-mail and staff meetings. "Our chief of research has an e-mail group so any time we want [to send] an alert to doctors we give it to her, and she sends it out to all MDs, PhDs, and other investigators," Herzog says. "And we have a research web site where we have already posted all of our forms, updated consent templates, and primary investigator guidelines on research."

Create an AE reporting form that collects enough information to provide perspective.

Catholic Medical Center, a community tertiary hospital in Manchester, NH, has an IRB that has created a three-page IRB report that details AEs and asks investigators to assess the risk-benefit ratio based on study results at the time of the report. "We want to know the number of participants screened, the number enrolled, the number withdrawn from research and why, the number of deaths, and the total number of adverse events to date on the local and national and international level," says Donna L. Bennett, RN, MS, CIM, vice president of research and program development at Catholic Medical Center.

The report form asks investigators whether the event resulted in death, a life-threatening situation, persistent or significant disability/incapacity, hospitalization or prolonged hospitalization, an overdose or protocol error, or a congenital anomaly/ birth defect, or cancer. One question asks, "What is the likelihood that this adverse event was caused by the drug/device or procedure?" Investigators may answer that it is a definite yes, probably yes, probably no, definite no, or unknown. The report form also asks whether the AE has been reported to the sponsor or a federal agency. Additionally, the form asks the investigator whether the consent form needs to be changed as a result of the reported AE.

Ultimately, the IRB makes the determination if there is a change in the risk-benefit ratio as a result of the reported adverse or unanticipated event, necessitating a change in the informed-consent document. An outcome of this type of reporting provides the opportunity to identify educational areas to strengthen with investigators and their staff, especially when the IRB’s determination is different from the investigators’, Bennett says.

Define unanticipated, serious adverse events and closely related research.

Hershey Medical Center’s AE definitions are outlined on the research web site. These include the following:

Unexpected adverse event: Any adverse experience that is not identified in nature, severity, or frequency in the consent form and is not due to a disease process.

Serious adverse event (SAE): Includes any experience that is fatal or life-threatening, is permanently or significantly disabling, requires inpatient hospitalization or prolongation of hospitalization, a congenital anomaly/birth defect, or any medical event which requires treatment to prevent one of the medical outcomes previously mentioned. The definition for serious adverse events was derived from the Office for Human Research Protections guidelines and is similar to the definition found in most sponsor-related protocols, Gifford says.

The web site also provides definitions of what closely related research in the reporting of serious adverse events means. It states that a closely related protocol is when all study procedures and drugs and the study population are almost identical to the protocol for which the SAE is being reported. A slightly related protocol, it states, is when one or two study procedures or drugs are the same as the protocol for which the SAE is being reported. The study population is usually different (e.g., different disease being treated, adult vs. children).

"There’s always the fear that you might be missing something, so you’re always re-evaluating your definitions to make sure you’re not," Gifford says.

In Bennett’s experience, the information an IRB may obtain from closely related or similar studies may be especially valuable in protecting research subjects. The Catholic Medical Center IRB had seen published research reports about a study involving a cholesterol-lowering drug that also was being investigated in protocols submitted locally. In the outside study, it was found that some patients were placed at a higher risk of sudden acute coronary events during a wash-out period in which they were taken off the medication, Bennett says.

"So the IRB at this institution asked one investigator what affect the wash-out period had on participants," Bennett recalls. "We questioned the sponsor about how appropriate this wash-out period was and whether we needed it and whether it would affect randomization?" she adds. "We didn’t want to put our patients who were going to enroll in a study that required a wash-out period to have a risk of a myocardial infarction."

The IRB’s inquiry and its pointing out the SAEs that occurred with similar research convinced the sponsor to change the protocol because it turned out that having a wash-out period really wasn’t necessary, Bennett says. "One pharmaceutical company mentioned that we were the only IRB in the country who asked that question," Bennett adds.