Is the fact you reviewed this manuscript before confidential?

Apr 15 2016 Published by under Peer Review, Science Publication

Interesting comment from AnonNeuro:

Reviews are confidential, so I don't think you can share that information. Saying "I'll review it again" is the same as saying "I have insider knowledge that this paper was rejected elsewhere". Better to decline the review due to conflict.

I don't think I've ever followed this as a rule. I have definitely told editors when the manuscript has not been revised from a previously critiqued version in the past (I don't say which journal had rejected the authors' work). But I can't say that I invariably mention it either. If the manuscript had been revised somewhat, why bother. If I like it and want to see it published, mentioning I've seen a prior version elsewhere seems counterproductive.

This comment had me pondering my lack of a clear policy.

Maybe we should tell the editor upon accepting the review assignment so that they can decide if they still want our input?

28 responses so far

  • Chris lbs says:

    Just this week my husband just told the editor that he had rejected a piece for a different journal. I think he named the journal and said that it was only slightly revised (improved??) from the last version he read.

  • Pinko Punko says:

    I think reviewers should disclose to editor for sure- I'm not sure I buy the confidential argument. I think it is possible conflict (not necessarily a conflict)- it is information that editor should have.

  • qaz says:

    Do editors care if their journal was the first try? I've never seen any evidence of that.

  • Morgan Price says:

    If you thought it was flawed science the first time and it doesn't look like it has changed much, why review it again? In this situation I just send the editor a note saying this and attaching my previous review.

  • genomicrepairman says:

    We've always said that we have reviewed before but we don't mention where or mention the other reviewers comments. I'm not sure that it matters too much.

  • Busy says:

    I see re-reviewing a paper as a time saver for all involved. In the vast majority of cases I deal with, suitable changes have been made and the new venue is more modest which results in rapid acceptance, all while less work on my side. I.e. a win-win situation.

    If the paper has not been changed, I also indicate this and attach a new review with minimal changes. Again, a work saver. In my experience, nothing kills a paper faster than a re-submission in which minor corrections were ignored.

  • AnonNeuro says:

    >> "Better to decline the review due to conflict."

    I meant to say, "if you're going to decline, better to cite a conflict than say it's because you've already reviewed it."

    Just to clarify -- I don't have a problem with reviewing the same paper twice. But I don't think you should tell the editor you've seen it before and/or make comparisons to a previous submission. Even if you don't mention the other journal, you're coloring the submission with a rejected tone.

  • Lisa says:

    I'm trying to think of a time when I was asked to re-review a paper and the editor did not already know that I had reviewed it the first time. In my field, at least, when a paper is rejected by a primary society journal, the editors will give the authors the option of revising and submitting to a secondary society journal. Often, the same reviewers are asked to review again.

    Looking through my records, I can find a couple of instances where this has happened. I don't remember if I mentioned it to the editor before reviewing. Hmm.

  • DJMH says:

    AnonNeuro, how do you make sense of the fact that there are whole consortia of journals which will forward their reviews on to the "next" journal's editors? Obviously in that case everyone chooses to have reviews/reviewers shared among journals. I don't see how this hurts the process.

  • drugmonkey says:

    Those are opt-in are they not, DJMH?

  • MoBio says:

    If I've reviewed it previously and declined it, I generally decline to review it for another journal without comment--other than 'sorry too busy right now'.

  • qaz says:

    l don't understand people who decline to see a manuscript a second time. Presumably you rejected it because it was wrong. Don't you care if it's still wrong?

  • DJMH says:

    DM, of course. But the point remains that given that people choose this system, obviously there's not a ton of concern about the "taint" that the manuscript, in some previous form, was rejected somewhere else.

    Personally, I'd rather have an editor know the reviewers were used before--that way it's clear that this review is not wholly independent to the review at Journal A. In fact, I think there's a case to be made that it is unscrupulous NOT to tell the editor you've reviewed before. I wouldn't make that case, just saying.

  • drugmonkey says:

    I do, qaz!

  • AnonNeuro says:

    > "AnonNeuro, how do you make sense of the fact that there are whole consortia of journals which will forward their reviews on to the "next" journal's editors?"

    As DM said, this is the author's choice to disclose. From my experience, one always has the option to resubmit to another consortium journal without transferring, in which case the information isn't shared.

    As an aside, how common is it for authors to use these "transfers"? I know PLOS will transfer any rejected paper from PLOS Biology, Genetics, etc. to PLOS ONE, but I'm guessing most people shop laterally first (for better or worse). Do papers that get reviewed by rejected at Nat. Neuroscience often end up in Nat. Communications?

  • drugmonkey says:

    DJMH- you know that I remain intensely skeptical that the peer review consortium is used in large numbers. Have I missed any data on that? Anyone familiar with these numbers?

  • DJMH says:

    Still, I think it's arguably *better* to disclose to an editor that you've seen it before so they can choose how to handle that--include you, not include you, add an additional reviewer, whatever.

  • A. Tasso says:

    As an editor, I appreciate it when the authors disclose that a paper has been submitted elsewhere. For example, last week I had a paper on my desk where the authors disclosed that their manuscript had been rejected from BMJ. They provided the BMJ reviewers' comments, along with their responses to the comments. Easy for me. So I farmed it out to a single reviewer rather than the customary 2-3.

    On the other side of the fence, when I am reviewing a manuscript, I typically only disclose that I have previously reviewed the manuscript *** when it is clear that the authors have not made any changes since the last time I saw it ***. If I get a manuscript a second time, and it is clear that the authors made a good faith effort to address the comments that were given to them in the rejection letter, then I do not disclose. I basically write my review as if it were a new manuscript. However, if it is clear that the authors simply turned the manuscript around again without modification, then I (a) disclose to the editor "hey I reviewed this manuscript at the Annals, and the authors didn't change it at all. So I am copying verbatim the review that I wrote once already"; and then (b) I cut and paste my review from the previous journal, just to make it clear to the authors that they are now on my sh*tlist.

  • Philapodia says:

    On a tangent, how often does everyone here accept review assignments? Do you accept every one you get, or are you selective about which ones you take? Does the journal matter (PNAS = yes; PLoS One = no)?

    Since reviewing is voluntary and unpaid, but considered service by a lot of Unis, how much is enough or not enough.

    Personally, if I have never heard of the journal before I will decline (lot from those Hindawi journals!), or if I have too much other shitte going on that I don't feel like I can spend the 1-2 hours doing a good review. I don't dump reviews on my staff because that's not fair to my staff or the authors. Sometimes I feel guilty about declining a review, but there's only so many hours in a day and so many other things that have to be done that this takes lower priority.

  • M says:

    The American Journal of Ophthalmology requires authors to indicate whether or not the manuscript has been reviewed elsewhere. If it received review, they want to know what was improved/changed. Not sure how common this is in other fields.

  • Busy says:

    When I was a junior researcher I would pretty much take any review assignment from a respectable journal. I recommend my grad students and postdocs do the same as it gets their name out there with AEs. Now that I'm quite a bit more senior I restrict my reviews to subjects that are a perfect match to my research interests. I still get to review quite a a few papers this way anyhow, since the more senior you are the more people think of you as a potential referee.

  • pablito says:

    I used to accept almost all relevant review assignments, especially those from famous journals. But I also took on reviews for less well known journals, which added to a fairly meaningless "trophy list" of journals for my CV (like keeping bottles of different beers). There were 35 journals over the years, everything from Cancer Cell and Genes & Development to PLOS ONE and Naunyn-Schmiedeberg’s Archives of Pharmacology . After retiring I stopped doing reviews.

  • drugmonkey says:

    I accept selectively. On Editorial Board of the journal? I will take all of them unless I am really really busy. Otherwise it depends on topic (my current interest- I'll take all of them) and my current workload or vacation schedule. Don't want to delay other people too long.

  • MoBio says:

    @Philopodia
    As I'm a handling editor (200-400 papers/yr) I'm pretty selective about anything that doesn't come to the journal simply because of time constraints.

    @QAZ: when I outright recommend 'reject' I'm typically extraordinarily negative (usually if it is fixable i'll recommend 'reconsider' or something like that) and believe they would be better served by having someone else look at it.

    I am surprised to see it at other journals even though what seem to be serious defects remain.

    Caveat emptor fellow scientists.

  • physioprof says:

    Now that I'm quite a bit more senior I restrict my reviews to subjects that are a perfect match to my research interests.

    I do the exact opposite, as it's much more interesting to learn something new than to review shitte that I already know about.

  • AcademicLurker says:

    I try to accept most requests if time allows. As an editor, I know how difficult it can be to find people willing to review a manuscript sometimes, so I try not to be part of the problem.

    I publish a lot of papers with "Important Protein X" in the title, and my work is very much on the molecular and biophysical end of things. Nevertheless, I'll occasionally be asked to review a paper on Important Protein X that involves a totally different set of methods, behavioral phenotyping of mouse models, for instance. Since I know f*ck all about behavioral phenotyping of mice, I reject those requests. My name must come up purely through word-association with Important Protein X.

  • dr24 says:

    This strikes me as classic overanalyzing. There is no meaningful overarching ethical principle here. Take it case by case.

  • A. Tasso says:

    @Philapodia: I used to accept _every_ single request, because that is what my supervisor told me to do ("never say 'no', journal editors have long memories"). Now that I am busier I have revised the rule slightly, and I usually say 'yes' unless it is a journal that I would never want to publish in. I also tell my postdocs to do the same.

Leave a Reply