An old friend and colleague of mine and I recently had an exchange over Facebook where he asked me the question “what is your take on a reviewer sharing a manuscript which they have accepted to review with their lab group for educational purposes?” My first thought was “the whole lab group? No way, that’s just not right.” But I do know that people asked to review papers in biomedical science regularly get colleagues to read some or all of the manuscript and this request can take many forms:
- Asking a colleague for advice on one particular experiment
- Co-reviewing the manuscript with a trainee in the lab (i.e., two independent assessments merged into one joint review)
- Delegating the review partially (or completely!)
I’ve had variable experiences during my own training ranging from one extreme to the other – in some cases my supervisors never passed on manuscripts at all, and in others it was clear that they wanted me to review it for them (not with them). While I learned a lot from the latter, there is also a labour exploitation angle that people might (fairly!) complain about.
For non-scientists reading the article, it is probably important to briefly explain that research labs are generally comprised of a number of people (anywhere from five to 50 depending on where you are) with distinct but related areas of expertise. The research that emerges therefore regularly has large numbers of co-authors reflecting the importance of each individual’s contribution to the work – a team sport, so to speak. This is in contrast to research in the arts and humanities where projects are much more individual and the supervisor guides/advises rather than constructs/directs.
Is it ethical/legal?
The first question that pops into my head is whether or not one should share a paper, but an interesting extension to that is whether or not it is actually legal to share or consult on a paper.
For example, Nature Publishing Group has the following snippets in its confidentiality policy:
- “Editors, authors and reviewers are required to keep confidential all details of the editorial and peer review process on submitted manuscripts.”
- “Reviewers must maintain confidentiality of manuscripts. If a reviewer wishes to seek advice from colleagues while assessing a manuscript, the reviewer must consult with the editor and should ensure that confidentiality is maintained and that the names of any such colleagues are provided to the journal with the final report.”
- “Nature Research reserves the right to contact funders, regulatory bodies, journals and the authors’ institutions in cases of suspected research or publishing misconduct.”
Also, the Committee on Publication Ethics (COPE), has published guidelines on appropriate conduct for peer reviewers and runs along similar lines:
Respect the confidentiality of the peer review process and refrain from using information obtained during the peer review process for your own or another’s advantage, or to disadvantage or discredit others… Do not involve anyone else in the review of a manuscript (including early career researchers you are mentoring), without first obtaining permission from the journal.
Despite some egregious examples of breach in confidentiality that are passed around anecdotally, I’ve never heard of anyone being prosecuted for sharing a paper, but there are definitely movements to crack down on poor reviewer integrity. The course of action currently – if any were ever to be taken – seems like it would be a slap on the wrist at best, I’ve certainly never heard of someone losing their job over sharing papers around…. but maybe they should.
Are group leaders and lab members distinct?
My colleague made the point that a laboratory has expertise that extends beyond the individual sitting at the top and, as such, what the journal requesting the review would benefit most from is the expertise of the lab as a whole (i.e., let the PhD student or postdoc who does “that stuff” weigh in on things because chances are reasonable that they would actually be better than the group leader). They suggested that sharing within the lab group, and making it clear that things were to be kept confidential, might be a good way to get the full range of expertise and also to help junior scientists learn quite a lot about how to review. This would run counter to Nature’s guidelines and it would also require an incredible amount of confidence in each and every one of the lab members to respect the confidentiality that the group leader is ultimately responsible for protecting. That said, as explored below, there are other ways to involve people in your lab in peer review.
Young scientists do need to learn how to peer review
That point about young scientists is a critical one – and it needs to occur before a scientist is asked to review a paper independently. However, I would argue that there are far more effective and ethical ways of achieving this goal rather than a reviewer unilaterally assuming that it’s ok to share the confidential work of others. First, the reviewer could simply write the editor and ask whether a person (or multiple people) could be consulted on some or all of the manuscript. Second, people could sign up for group peer review projects like PreREVIEW journal clubs. Third, supervisors could run the “training” exercise on a paper after it is accepted (rather than during the review process).
Give credit where credit is due
A far too common practice is when a senior investigator completely delegates the task of reviewing a paper to a subordinate (postdoc, grad student, etc) and doesn’t even actually read the manuscript. And then, they have the gall to sign their name to the review without acknowledging the person who actually wrote it (or for that matter, actually bothering to check if the review is accurate). More often, the younger researcher does the nitty gritty rooting around in the paper (did they perform the experiments correctly, are the statistics appropriate) and the senior investigator weighs in on the overarching theory and helps craft the review document. In the latter situation, it may be fair to sign one’s name to it, but really it should be a jointly authored affair.
Overall, I suspect confidentiality is breached far more frequently than we would like to think and sometimes this can have very serious consequences, including dire career implications for people whose work has been unfairly circulated. Reviewers need to respect the work of others’ as they would want their work respected. We’d love to hear stories from readers (anonymously or otherwise) on their experiences.
A key challenge with reviewing manuscripts is finding the time. When I do get to this task I want to start and finish it promptly. I’m not interested in an interruption to seek and receive approval for a reasonable consultation.
I have a long-term research associate who’s more familiar than me with emerging techniques and I will sometimes seek his assistance to assess methodology or interpretation. If a manuscript describes a study that is closely aligned with a project of a senior graduate student or PDF I may invite them to take a look and provide their impressions. They are often up-to-date in tracking the literature and provide another view, and this also provides a training opportunity.
I’ll consider their input with my review and I’ll then report the consultation in my ‘confidential comments’ to the editor. This recognizes their effort moreso than seeking authorization. I’ve never had concern following this approach, and as an associate editor I’ve seen similar explanation from some other reviewers. (As with the Hopper quip, ‘it’s easier to seek forgiveness than permission’.)
I don’t think that delegation is appropriate. If you don’t have the time or interest you should decline the review request.
Thank you for this even-handed take on sharing reviewing responsibilities. One elaboration that would have been helpful is a discussion of _why_ it is important to maintain confidentiality. After all, the manuscript authors probably don’t care if a few trainees see a pre-publication draft of their work (which may already be available on preprint servers anyway). What they _do_ care about is when the reviewer is a competitor, who either sits on the manuscript to allow their own work to get out first, or writes such a damaging review that it doesn’t get published. Perhaps worse is when the reviewer steals the ideas in the manuscript, does the experiments, and claims them for themselves. The concern over these behaviours probably outstrips their actual occurrence, but data are sparse on the extent of the problem, and there are certainly plenty of anecdotes to go around. Unfortunately, this is one of those situations where journals’ guidelines (adherence to which is difficult to monitor or enforce) have no impact on the bad actors who engage in these behaviours. Rather than worrying about which lab members get to see manuscripts, it may be more productive to think about mechanisms for detecting and reporting the unethical behaviours described above.