Addressing bias in peer review

How can we address bias in peer review? How can we make the system more effective? How can early career researchers get involved? These are just some of the topics discussed at the recent Academic Career Development Event at the 2016 Alzheimer’s Research UK Conference.

As part of a panel discussion chaired by Maria Kowalczuk, Biology Editor in BioMed Central’s Research Integrity Group, panelists gave their perspective on the current state of peer review, potential solutions to address bias and ways for early career researchers to get involved.

Maria was joined by:

  • Stella Hurtley: Senior Editor, Science
  • Niki Scaplehorn: Chief Life Sciences Editor, Nature Communications
  • Bettina Platt: Chair in Translational Neuroscience at the University of Aberdeen, Network coordinator for the Scottish Alzheimer Research UK (ARUK) network, member of the ARUK Scientific Advisory Board and Alzheimer’s Society Grant Advisory Board

Bias in peer review

Bettina opened discussions by highlighting problems with bias in peer review, referencing this article by Richard Smith, former editor of the BMJ. Peer review involves inherent unconscious biases such as being more (or less) positive about a manuscript based on factors such as who the authors are or where they are based.

While peer review is needed as a quality control mechanism, Bettina questioned how peer reviewers can be unbiased when assessing other people’s work, particularly when it may conflict with their own hypotheses.

Bettina called for more community discussion around published articles in an open forum following an initial ‘quality assessment’ by peer reviewers to look for flaws.

Addressing the bias

Attempts are currently underway to address bias in peer review, such as experimenting with the use of different models of peer review.

Niki spoke about initiatives being trialled at Nature Communication. One such initiative offers authors the option to choose double-blind peer review, a model in which neither the authors nor the reviewers know each other’s identities. It’s the model which has come out as favoured in large surveys of peer review, though in practice there are challenges in blinding reviewers to authors’ identities.

Another option being trialled at Nature Communications is transparent peer review, a model in which reviewer comments to authors and author rebuttal letters are made available online if an article is published. Both models will be evaluated, and I look forward to seeing the results of this.

Another model to make the peer review process more transparent is open peer review, used by a number of journals including the medical journals of the BMC series and the BMJ. Going one step further than transparent peer review, authors are aware of the identity of the reviewers and if the article is published, the reviewer’s signed reports are made available online.

Stella spoke of ‘Cross Review’, an initiative at Science whereby reviewers are able to see and comment on each other’s completed reports (anonymously) before they are sent to authors.

How can early career researchers get involved?

As well as ways to address bias, there’s a real need to get early career researchers involved in peer review. Discussions focused on how early career researchers can get involved and the benefits this brings to writing up their own research.

There was universal encouragement from panelists to the audience to get involved. Bettina gave the audience the following advice – “Understand and engage in the process – talk to your supervisor and let them know you’re willing to review – they’ll probably be relieved to have some help. Try to learn strategically and follow other people’s thoughts.”

These thoughts were echoed by Stella who reminded the audience that peer review should be used as a learning experience, adding that at Science they allow and encourage reviewers to share manuscripts with one or two trusted members of the team so long as they let the Editor know.

Understand and engage in the [peer review] process – talk to your supervisor and let them know you’re willing to review

Bettina Platt

In addition to getting involved in reviewing ‘real life’ manuscripts, Niki highlighted that with transparent and open peer review, the reviewer reports and older versions of manuscripts that are available online can also act as valuable learning tools.

Once you’re starting to build up your publication history, what can you do to get invited as a peer reviewer in your own right? Stella suggested making yourself and your work known to the journals board of reviewing editors – speak to them at meetings.

If you are invited to peer review, you may wish to look at our blog series on how to peer review.

Where do we go from here?

While there were lots of different opinions in the room, one thing was agreed – peer review isn’t perfect, but it provides an important quality control mechanism that we can’t do away with.

As always, there’s tweaking that can be done and improvements that can be made. We are seeing a time of great innovation in peer review and there are many different models emerging. As new innovations emerge, they need to be tested in a scientific way. Time will tell whether there will be a ‘one size fits all’ model or whether different fields require different models.

View the latest posts on the Research in progress blog homepage