To IRB or not to IRB?

The Committee on Publication Ethics (COPE), of which PLoS Medicine, and other PLoS journals, are members, has recently released guidance for editors on the thorny topic of "research, audit and service evaluations". This guidance aims to help editors decide how to handle their journal's requirement for ethical review in relation to these types of studies. As outlined by COPE, most journals require that for any research involving human subjects, the study has been approved by a properly constituted ethics committee.

However, for some types of studies and in some countries, ethical approval is not required (or, may not even be possible). For example, the
US Department of Health and Human Services exempts research from review if it involves studying existing data or specimens, providing data is anonymised and delinked from the participant it originated from; exemptions also apply for five other major categories of research. In the UK, the National Research Ethics Service issues its own guidelines to help people to work out whether what they are planning to do is research (which requires ethical review) versus audit or service evaluation (which don't).

The PLoS Medicine editors recently discussed the COPE guidance. That guidance suggests that editors think about four key components, quite apart from the "box-ticking" exercise of checking that ethical review and informed consent have been obtained. Those four areas are: 1) scientific validity; 2) minimisation of harm; 3) whether benefits outweigh harms; and to 4) the need to seek clarification from authors or ethics committees regarding local laws and regulations, as necessary.

At PLoS Medicine, our journal policy is that "All research involving humans and animals must have been approved by the authors’ institutional review board or equivalent committee...". We do however recognise that many studies may be exempt from ethical review, and therefore our overriding approach is that authors describe clearly in their paper how ethical oversight for the study was obtained, and how the informed consent of participants was handled. This means that reviewers, and readers, will be able to judge for themselves whether the protection of human participants in the study was appropriate.

At the same time, we recognise that "journals should not automatically defer to ethics committees", as Kamran Abbasi and Iona Heath have pointed out in a BMJ editorial, and we consider the key areas outlined by COPE as part of our editorial review process. If we do have concerns regarding the protection of human participants in a study, irrespective of whether that study was approved by an ethics committee (or exempt from ethical review), the editors reserve the right to follow up, under the guidance of our advisory group on publication ethics, as necessary.