April 2005

The Office for Human Research Protection (OHRP) Compliance Activities: Common Findings and Guidance (July, 2002) states that there is often a failure to promptly reportunanticipated problems involving risks to subjects or others to the IRB as required by HHS regulations at 45 CFR 46103(a) and 46.103(b)(5). The failure to report is considered non-compliance under federal regulations.

Although most research personnel are familiar with adverse event reporting, many are unfamiliar with the requirement of reporting unanticipated problems involving risks to subjects or others. To clarify, unanticipated problems involving risks to subjects or others is not synonymous with adverse event. Although the two may overlap, some unanticipated problems involving risks to participants or others are not adverse events and vice versa. It is important to note, that both unanticipated problems and adverse events can occur in either clinical or social/behavioral science research.

To increase understanding of this requirement it is helpful to break the information into several parts. The first part, prompt reporting, is not clearly defined in the FDA and DHHS regulations although it implies that information will be delivered without delay or that an individual will be “quick to act or respond.” Essentially, when the event is first noticed and brought to the investigator’s attention, it should be reported.

The next word unanticipated, denotes an unforeseen event; in other words, the event was never intended to occur under any circumstances. The phrase, problems involving risks to subjects, refers to an incident where the event actually results in harm to the participant. This is a reportable event. The statement also refers to those problems that may not result in “harm” to the participant. Rather, the phrase implies only that the possibility of harm may occur. For example, records stolen during the course of a research project that contain identifiable, private information would be a reportable event. In this example, actual harm to the participants from a breach of confidentiality may or may not occur -- there is only the possibility of harm to participants.

What about the phrase risks to others? Although the focus of reportable events is often on the participants themselves, DHHS and FDA actually require that unanticipated problems involving risk to others must also be promptly reported. For example, if a study coordinator is inadvertently exposed to the HIV during a study, this would be a reportable event. Others can also be interpreted to mean family members or other study participants.

Although unanticipated problems involving risks to subjects or others is often misinterpreted as adverse event reporting, investigators must not make the mistake of only reporting the latter. The Principal Investigator is in the best position to carefully assess the event and determine whether the event is an “unanticipated problem involving risks to subjects or others.” After careful analysis, the PI can then decide whether the protocol requires modification to minimize risk, consent form should be revised, or whether participants should be re-consented. 

*Prentice, E. (2002). IRB Review of Adverse Events. In R. J. Amdur and E. A. Bankert (Eds.), Institutional Review Board Management and Function (pp. 297-300). Sudbury, MA: Jones and Bartlett.