Fifteen years ago BioMed Central was a pioneer of openness and transparency in publishing. Since then, openness is increasingly becoming the norm in all fields of research, as evidenced by, for example, the AllTrials campaign and the open data movement.
We were ahead of our time when Fiona Godlee championed open peer review on all medical journals of the BMC series. Under that model of peer review authors know who the reviewers are and, should the manuscript be published, readers know too as the entire pre-publication history including all previous versions of the manuscript is published. Over the years, many of the non-BMC series journals, such as Trials, GigaScience, and Biology Direct to name but a few, have also adopted fully open peer review.
During the last 15 years, we have tackled many unforeseen problems associated with this level of transparency and to date have found a way to deal with them all without compromising our ‘total transparency’ ethos. However, there is one area where total transparency could conflict with our other policies.
Imagine you came into work one morning to be met by excited colleagues telling you that you had been on television the previous night. You had known there was a television crew filming at your place of work, but had no idea that you too had been filmed and that the film had been broadcast on national television. You had not given your consent for this to happen. While ultimately this may seem harmless and be seen by friends as a positive and exciting thing, the fact that your wishes had been ignored could leave you feeling angry and undermined.
Imagine how much worse someone would feel if it was details about their or their children’s health that were published for the world to see without their consent.
Everybody has the right to privacy and this ideal is enshrined in legislation to protect personal data.
Everybody has the right to privacy and this ideal is enshrined in legislation to protect personal data such as the European Data Protection Directive, for example. The right to confidentiality is equally important and requirement for consent to publish any individual’s data is a fundamental aspect of ethical research.
Our journals have clear policies on this and in order to maintain privacy and confidentiality we have occasionally had to remove sections of previous versions of submitted manuscripts before publication in cases where authors have not been able to obtain consent to publish. This is not total transparency. So, we have an uneasy balance between championing total transparency on the one hand and respecting confidentiality on the other.
A further complication is the small risk that confidential information could be inadvertently published in earlier versions of the manuscript.
We are aware that our current systems cannot provide a 100% guarantee that this won’t happen but, where confidentiality is concerned, this is the minimum assurance we aim to provide. In an attempt to openly address this we have made the decision that for all articles published from 30th March 2015 onwards, the online prepublication history will contain only the reviewer’s signed reports, and will no longer contain earlier versions of the manuscript or authors’ responses to reviewers.
Reviewers’ reports, including their names, will still be posted online alongside the published article. Moreover, any readers wishing to view the complete pre-publication history (including earlier versions of the manuscript and the authors’ responses to the peer reviewer comments) can still do so by contacting us. However, we will remove any identifiable details from earlier versions of the manuscript if we do not have ‘consent to publish’ from the participants of the study.
By continuing to make this information available (albeit on request), we will ensure that the prepublication history remains a valuable resource as a window into the peer review process, to assist with training of peer reviewers, and for further research into peer review. At the same time, we will be providing a more secure means of protecting confidentiality.