Adverse Event Reporting To IRBs – Improving Human Subject Protection

Date: January 2009

Status: Final

Target Audience:

Anyone working on clinical trials of drugs, biological products, or devices at investigative sites (ie, clinical investigator, and clinical research coordinator), IRBs, or sponsors (ie, clinical research scientist, pharmacovigilence department, site monitor)

Laws and Regulations:

  1. Title 21 of the Code of Federal Regulations (CFR) part 56 (IRBs)
  2. Title 21 of the Code of Federal Regulations (CFR) part 312 (Investigational New Drug Application)
  3. Title 21 of the Code of Federal Regulations (CFR) part 812 (Investigational Device Exemptions)


This guidance was written to clarify the requirements for reporting adverse events (AEs) that occur during clinical trials to the IRBs.


The primary purpose of the IRB is to protect the human subject participating in a clinical trial. To do this, they evaluate the trial before it starts and during its conduct. To monitor subject safety, they must be given the appropriate safety information, such as reported AEs. However, even in the safest clinical trial, many AEs occur and most do not constitute a safety signal. Prior to the guidance, there was a lack of clarity regarding what types of AEs the FDA expected to be reported to IRBs. As a result, the volume of AEs reported to IRBs was overwhelming and it was difficult for them to effectively evaluate significant safety issues.


In general, it is recommended that all unanticipated AEs during clinical trials be reported to the IRB. The guidance defines an unanticipated AE as an event that is unexpected, serious, or would require a safety related change in the protocol or informed consent form (ICF). This is especially true if the event meets these criteria and is considered by the investigator to be related, or potentially related, to the drug, biologic, or device being studied.


In most cases, a single occurrence of an event does not constitute an unanticipated event. For multicenter trials, the sponsor is able to evaluate whether the event is unanticipated because they have the ability to review data from all study subjects. Therefore, it is recommended that the individual investigators conducting the study rely on the sponsor to report unanticipated AEs to the IRB. The sponsor would also be responsible for reporting an AE that was determined to be unanticipated to all participating investigators.

Of course, there are exceptions to every rule. Some events, such as liver damage, are considered serious events that are uncommon in the non-study population. It takes only one occurrence of an event like this to send up a red flag and be considered unanticipated. It is recommended in the guidance that these types of events be reported to the IRB even if it is the first time it occurred in the study.


In general, the guidance is very helpful, although it does create a concern.

On one hand, it makes sense that the sponsor, being the party that has access to all the data, should be responsible for assessing if an AE occurs. However, on the other hand, if a sponsor is inclined to hide unfavorable outcomes, this reporting strategy opens the process up to bias and under-reporting.

For more information on this issue, contact the Kulkarni Law Firm.