Retraction Watch

Tracking retractions as a window into the scientific process

Publisher discovers 50 manuscripts involving fake peer reviewers

with 26 comments

bmc logoBioMed Central has uncovered about fifty manuscripts in their editorial system that involved fake peer reviewers, Retraction Watch has learned.

Most of the cases were not published because they were discovered by a manuscript editor on a final pre-publication check. The five or so that have been published will go through some sort of re-review, which may result in expressions of concern or retraction.

The narrative seems similar to that in the growing number of cases of peer review manipulation we’ve seen recently. What tipped off the editor was minor spelling mistakes in the reviewers’ names, and odd non-institutional email addresses that were often changed once reviews had been submitted, in an apparent attempt to cover the fakers’ tracks. Those “reviewers” had turned in reports across several journals, spanning several subjects.

It would seem that a third party, perhaps marketing services helping authors have papers accepted, was involved.

The publisher has let all of its external editors in chief know about the situation. To prevent it from happening again, authors will not be able to recommend reviewers for their papers. Here’s a message from BioMed Central senior managing editor Diana Marshall that went out to a number of journal editors earlier today:

We need to inform you of a situation that has come to light on several journals involving potentially fake peer reviewers.

When completing final editorial checks on a manuscript an in-house Editor spotted irregularities in reviewer reports from author-suggested reviewers. Further investigation suggests that the reviewers have been fabricated. By searching systematically across our systems we have uncovered a number of cases of these potentially fake peer reviewers returning reports across several journals including a number in the BMC series.

There doesn’t seem to be an obvious link between the authors as they are different in each case. We are concerned that a third party is involved, possibly supplying the names of fake reviewers to authors to influence the peer review process. While we investigate further we will temporarily switch off the functionality which allows authors to suggest potential peer reviewers directly into our workflows. This change will come into effect in the next few days.

We appreciate this may cause concern but given the level of sophisticated fraud that appears to be taking place we hope you will understand.

We will be in touch with individual Editors if manuscripts you have handled have been directly affected and if your attention is required for individual cases.

BioMed Central tells Retraction Watch:

In completing our checks and balances on a manuscript, one of our in-house editors spotted problems in reports from author-suggested reviewers. Since this was flagged, we have searched our systems and found several potential false reviewer accounts that seem to have returned peer review reports to several of our journals. At present this amounts to around 50 manuscripts, the majority of which have not been published and are held in our systems.

We cannot see a clear link between the authors and believe that a third party may be involved, and influencing the peer review process.

At this stage we are investigating further and the manuscripts are on hold. We have informed our staff and external editors and are switching off the functionality that allows authors to suggest potential peer reviewers directly into our systems while we look into the issue in more depth.

Journals have retracted more than 100 papers in the past two years for fake peer reviews, many of which were written by the authors themselves.

Update, 2 p.m. Eastern, 11/25/14: Based on additional information, clarified headline and first line of post to reflect that most of the 50 manuscripts were somewhere in the editorial process, not accepted. BioMed Central also tells us that in many cases, editors did not use the author-suggested reviewers, but nevertheless held the manuscript after noticing the irregularities.

Written by Ivan Oransky

November 25th, 2014 at 1:05 pm

Comments
  • Dave Fernig November 25, 2014 at 1:13 pm

    So greedy – had they rejected on paper in 4, perhaps the scam might not have been detected?

  • JATdS November 25, 2014 at 1:20 pm

    “an in-house Editor spotted irregularities” = had this not be detected, then most likely BMC would never have detected this. This sounds like a total failure of the traditional peer review system.

    “There doesn’t seem to be an obvious link.” OK, this suggests that BMC actually doesn’t have a hold on this case, and doesn’t really know what’s going on. They only seem to know what they detected, but this suggestes that the problem might be more widespread.

    “given the level of sophisticated fraud that appears to be taking place” If BMC doesn’t even know what’s ging on, then how are they able to assess the sophistication of the fraud?

    “We appreciate this may cause concern.” You can bet your bottom dollar on that. What the scientific community now wants to know is:
    a) how wide is this reviewer fraud?
    b) which journals exactly did it affect?
    c) which papers got through based on fake peer review?
    d) how have papers that have passed under the peer reviewer radar influencing (or going to influence) the impact factors of BMC journals?

  • Paul Brookes November 25, 2014 at 1:53 pm

    This is why we can’t have nice things!

    There’s nothing inherently wrong with authors being allowed to recommend reviewers, so when a few selfish people abuse the situation, the solution should not be to punish everyone else. But of course, proper vetting of reviewers would cost time and money, and we all know what matters to journals the most. Shame on BMC for taking the easy route.

    • Vladimir Svetlov November 26, 2014 at 7:21 am

      What is exactly you are shaming BMC for? For detecting the attempt to defraud the much venerated “peer-review” system? You do realize there are millions of so-called “scientists” all over the world now and all of them wanting to publish their “findings”? And there are parts of this world where the culture of science simply doesn’t exist. Where they have money to buy the latest Illumina instrument but don’t have a word for “plagiarism”.

  • BB November 25, 2014 at 2:03 pm

    I never really grasped the concept of author-suggested reviewers. The public is also polarized about it, I know people who never suggest reviewers for their papers unless they are obliged during submission as they think its unprofessional, and then there are others who always do that as they are on the opinion that managing editors appreciate such help.

  • JATdS November 25, 2014 at 2:26 pm

    Although perhaps obvious to some, it might not be obvious to all. So, allow me to point out some important points, with help from Wikipedia [1] (direct quotes):
    a) owned by Springer Science+Business Media
    b) a United Kingdom-based, for-profit scientific publisher
    c) BioMed Central describes itself as the first and largest open access science publisher
    d) In 2002, the company’s business model evolved to include article processing charges, and these have since been the primary source of revenue
    e) BioMed Central was founded in 2000 as part of the Current Science Group (now Science Navigation Group, SNG [2]), a nursery of scientific publishing companies.

    The economic repurcussions may be substantial if BMC is unable to get a hold on this. If BMC fails to release details and be transparent about the actual papers and journals affected, their image could become tarnished.

    [1] http://en.wikipedia.org/wiki/BioMed_Central
    [2] http://sciencenavigation.com/aboutus.html

  • Michael Kenward November 25, 2014 at 2:28 pm

    “It would seem that a third party, perhaps marketing services helping authors have papers accepted, was involved.”
    Hmmm. In what way? Writing reviews of the papers they were pushing?

  • JATdS November 25, 2014 at 3:07 pm

    Michael, I think that is a very central aspect of this case and why we need to hold BMC and Springer accountable here. Because, if we are dealing with commercial companies that are providing such fraudulent services, then we may be entering a new chapter in science publishing fraud, in which companies may be selling services to allow some scientists to get ahead of the pack… if the price is right. If indeed, there is such a marketing service, in addition to the pertinent question you ask, other aspects exist:
    a) apart from the fraudulent aspect of writing reviews for others, this would be downright ghost-like services, as serious as ghost authorship;
    b) if indeed a service company has employees writing such reviews, then who are the writers? Surely they are scientists? If so, what ethics and corporate responsibility does this company and its employees have?
    c) In what country does the company operate and from what country do its “writers” originate?
    d) Do such companies exist as we speak? I would imagine that such companies might be particularly successful, and profitable, in countries like China that game the impact factor and IF-journals for financial gains and ranks. Such scientists would clearly benefit from having such a service that sets up fake emails, fake identities, fake reports.

    So, the topic of “caretaker businesses” in science publishing needs alot more investigation because they are at the fringe of publishing ethics.

    Of course, a)-d) are pure conjecture at this point, but well worth thinking about now that we are a this cross-road.

    As for author-suggested reviewers, in my field, at a certain point, it becomes unbearable because the field is narrow, the number of specialists are limited and the interaction with most eliminates their selection becase it would be a COI. The author-suggested system (aka the online submission systems) serves, I believe, simply to boost the “name stocking” of data-bases of the commercial publishers, making editors lazier and lazier. The system would work quite wel if the manuscripts were in a double-blind peer review in which at least one editor and at least 2 peer reviewers were overseeing the whole process.

  • Narad November 25, 2014 at 3:34 pm

    Most of the cases were not published because they were discovered by a manuscript editor on a final pre-publication check.

    As much as it would please me if were this the case, I don’t get the impression that BMC “in-house editors” are “manuscript editors” in the usual sense. It seems to be some sort of support role that is a hybrid of the usual managing editor and assistant editor positions.

  • Neuroskeptic November 25, 2014 at 5:08 pm

    BMC is a major publisher with the resources to deal with this issue and a motivation to do so. One dreads to think how many of the minor or predatory publishers have been victims (or maybe accomplices) of such a scam…

    Certainly I would say that if a publisher can’t even avoid typos on their own website and spam emails, they are unlikely to spot typos in reviewer email addresses…

  • JATdS November 25, 2014 at 5:40 pm

    Neuroskeptic, this is precisely why this story is so serious. It is precisely because BMC, Nature, Elsevier, SAGE and others have been duped that this indicates that their billions in resources have somewhat of a limited ability to detect this type of fraud, even less when perhaps the dude inventing the e-mails and identities may be from the sticks somewhere in a developing country, almost impossible to trace, even with an IP address. Many of the predatory publishers already run zero or pathetic review, so manipulated peer review in those venues doesn’t seem to change the overall game plan, which is to accept almost anything. The publishers think that they are evolving to meet the challenges, but I get the sense that the frauds are evolving as equally effectively to beat the system, and some scientists are taking advantage of this. I don’t know about others, but I have not seen, or felt, the degradation of science as bad as this in my entire career so far, possibly because the tools to detect the fraud and the problems has increased. Although we must request BMC to be fully transparent about this, despite any differences, we have the moral obligation of lending a hand and support in their fight, simply because they appear to have been cheated.

  • Adam Etkin November 25, 2014 at 6:44 pm

    Wait, the editorial office actually catches what appears to be an elaborate scam prior to accepting/publishing (some details still emerging) and somehow this shows they’ve done a BAD job or that somehow they’re to blame? I don’t follow this logic.

  • Adam Etkin November 25, 2014 at 6:45 pm

    Note, my previous comment directed to some of the other commenters here, not Retraction Watch.

  • Edward Ciaccio November 25, 2014 at 6:46 pm

    I am editor of Computers in Biology and Medicine (Elsevier). It is mid level. I do a Google search of title and keywords to find independent academic referees for each paper. Most are faculty although I allow a few postdocs. I prefer 3-5 referees for each paper. I will invite one suggested referee if the address is a known university, government, or industrial institution. I do all handling of manuscripts. We are at about 900 submissions per year now. My board and associate editors referee when I occasionally have a hard time finding reviewers for the topic. Reviews are due in 3 weeks, and our turnaround time is about 4 weeks. I have found this paradigm to work rather well to prevent the kind of problem outlined in the article.

    • Rafael M Santos November 25, 2014 at 8:03 pm

      Edward, I recommend you try using Publons to find some experienced reviewers.

    • Narad November 26, 2014 at 12:37 am

      I do all handling of manuscripts. We are at about 900 submissions per year now.

      This is simple free association from Dr. Ciaccio’s contribution, but I’m reminded of a remark from Benjamin Apthorp Gould:

      “Proofs of No. 385 came this morning, but not as I had so laboriously arranged them on Sunday. The printer thought he could not squeeze the matter onto one of the pages and so had left the last page six lines long and the intermediate ones very ‘fat’ to my disgust. By the omission of a few needless leads all would have been right. Too late now.”

  • Adam Etkin November 25, 2014 at 8:17 pm

    Follow up…..does anyone yet know if the suggested reviewers were the only reviewers used on these articles? I’m also still trying to puzzle out how the authors who suggested the reviewers could not know something odd was going on.

  • Nitin November 25, 2014 at 10:02 pm

    I will also like to point out that while submitting the paper, editorial manager ask a question “Are you willing to review the paper?” Most of the scientist will answer “yes” and the funny part is most of them are very novice and fresher, thus it gives me sleepless night after I submit the paper, fearing that my paper will go to such immature person.
    and yes I personally feel that suggesting the reviewer should be optional -currently it is mandatory by several journals.

    • Marco November 26, 2014 at 4:57 am

      I know of at least one journal that uses the suggested reviewers to expand its database of potential reviewers.

      Another journal in which I am involved demands you select suggested reviewers from its Editorial Board and its Advisory Board. My pending submission does give me some trouble with that one, as the most obvious people are also those who I have collaborated with very recently, or who are connected to either myself or my collaborator.

    • Gerald F November 26, 2014 at 5:47 am

      Well, the real problem is not choosing author-suggested reviewers. The problem is uncritically choosing those. There are certain guidelines to follow, though these vary from journal to journal. The obvious first step is to check the validity of the e-mail address. A reviewer should not be from the same institution, and should not be a recent collaborator, for instance.

      Seniority of reviewers has little to do with peer-review quality, at least I know of no study proving it. Reviewing has to do with a scientific, critical approach to the work that has come before me. Also PhD students can be reviewers, I think, given that they have published work on the topic, or are active in the field and known to the editor.
      The fact that most reviewers are faculty is probably due to the career paths. Many “junior” scientists disappear from academia, and not many “junior” scientists become peer-reviewers. Still, that does not tell anything about peer-review quality.

      “Novice and fresher” reviewers (whatever that means) are not necessarily immature! Many senior researchers commit fraud or are inclined to questionable research practices, as is frequently demonstrated on this site.

      At some point, everybody reviews a manuscript for the first time – so everybody is a “novice and fresher” once!

  • Dikran Marsupial November 26, 2014 at 3:46 am

    Journals should not request authors to nominate reviewers. The editorial board should have sufficient expertise to identify suitable reviewers for themselves and if this is not possible then either the editorial board should be expanded to include someone who can identify reviewers for that particular topic or the particular topic should be considered outside the scope of the journal.

    Asking reviewers to nominate reviewers is a recipe for a failure of the peer-review process, it is not difficult to find examples of this. It is not unusual for there to be a small group of scientists holding a non-standard view (e.g. that the rise in atmospheric CO2 over the last century is natural rather than being due to fossil fuel emissions). Are they going to nominate other members of this small group, or are they going to nominate mainstream scientists who they think are mistaken on a fundamental issue? I am not saying that this is in any way unethical, just that it is natural for a researcher not to nominate a reviewer that they think doesn’t understand their work, even though ideally that is what would be best for them and their work (if you can only convince those who already agree with you, your work is not ready for publication yet).

  • Scott Edmunds November 26, 2014 at 5:28 am

    I on no way speak on behalf of BMC, but as an external editor working with them (and as a former employee) I have some insight into how their editorial processes work. The initial headline posted here was very misleading when only a few papers have made it through the cracks, and the majority have been detected during peer review. How are they getting criticized for dealing with this correctly? Some external academic (unpaid) editors may have been not been doing their job checking author suggestions, but the in-house team were and are obviously now changing their procedures and doing the necessary detective work. I think BMC has done a good job to detect and quickly let their editors, RW and others know about this, as if you systematically went through other journals this is likely the tip of the iceberg. If 8% is the estimate for medical papers ghost written by the pharma industry in the West (see: http://www.nytimes.com/2009/09/11/business/11ghost.html?_r=0), there is a good chance it is even higher in countries with more imbalanced and skewed academic incentive systems. The publishing industry doesn’t make it easy for people to datamine, but clever use of machine learning algorithms could detect a proportion of the papers that are ghost written by “publication bazaar” companies (see: http://www.sciencemag.org/content/342/6162/1035). My gut feeling is that these peer review rings are probably just part of the packages they offer.

  • Ruth Francis November 26, 2014 at 8:26 am

    We have a blog about the issue of fake peer reviewers. See http://blogs.biomedcentral.com/bmcblog/2014/11/26/who-reviews-the-reviewers/

  • Joachim November 28, 2014 at 3:06 am

    JATdS
    Michael, I think that is a very central aspect of this case and why we need to hold BMC and Springer accountable here. Because, if we are dealing with commercial companies that are providing such fraudulent services, then we may be entering a new chapter in science publishing fraud, in which companies may be selling services to allow some scientists to get ahead of the pack… if the price is right. If indeed, there is such a marketing service, in addition to the pertinent question you ask, other aspects exist:
    a) apart from the fraudulent aspect of writing reviews for others, this would be downright ghost-like services, as serious as ghost authorship;
    b) if indeed a service company has employees writing such reviews, then who are the writers? Surely they are scientists? If so, what ethics and corporate responsibility does this company and its employees have?
    c) In what country does the company operate and from what country do its “writers” originate?
    d) Do such companies exist as we speak? I would imagine that such companies might be particularly successful, and profitable, in countries like China that game the impact factor and IF-journals for financial gains and ranks. Such scientists would clearly benefit from having such a service that sets up fake emails, fake identities, fake reports.

    If you use a search engine and type in peer review service, one of the first companies offers “independent peer review.” The idea being that authors can save a lot of time, if they have their manuscripts reviewed independently and professionally prior to submission. That seems to be a legitimate business idea so far. The authors pay for the time saved. But what if one of this independent peer reviewers makes a little extra money on the side by serving also as a suggestible reviewer for the “real” peer review process?

  • Don August 11, 2015 at 12:05 pm

    “It is simply no longer possible to believe much of the clinical research that is published, or to rely on the judgment of trusted physicians or authoritative medical guidelines. I take no pleasure in this conclusion, which I reached slowly and reluctantly over my two decades as an editor of The New England Journal of Medicine.”
    Harvard Medical School’s Dr. Marcia Angell

    nytimes.com/2012/03/20/science/a-drumbeat-on-profit-

  • Don August 11, 2015 at 12:08 pm

    “The World Health Organization has now admitted that its policies governing the publication of conflicts of interests of its “expert advisers” have “inconsistencies” and that safeguards “surrounding engagements with industry” need to be tightened. They point out that their advisors where receiving money from the drug industry. ”

    bmj.com/content/340/bmj.c3167.full

    Where there is profit, you will find inequality, greed and envy trumps all. Always.

  • Post a comment

    Threaded commenting powered by interconnect/it code.