THE BLOG
06/02/2016 11:30 am ET Updated Jun 02, 2017

You're Only as Good as Your Co-author's Integrity

One unscrupulous person on an author team can wreck your reputation.

Recently a major journal in my discipline asked me to provide peer-review for a manuscript. With a half-dozen manuscripts from my lab open and splayed across three oversized computer screens, I certainly didn't have time to take on additional reviewer duties. But the subject warmed my interest. I was glad to give a little service to the journal, and the paper's authors are recognized for their compelling work. I decided to critically evaluate the submission.

I read the paper over a one-and-a-half-hour flight from Atlanta to Detroit, taking careful notes in the margins. The data were presented clearly and were interpreted conservatively. They tested a sound hypothesis with a reasonable approach. The scope and quality were consistent with the journal's typical articles. I only picked up a few simple errors and a noted handful of suggestions. As wheels touched runway I felt positive about writing up a recommendation for publication with minor revisions.

The next morning the first email in my inbox was a note from the Associate Editor, indicating the paper was declined without my review. It turns out that another reviewer recognized exact figures from a paper that was submitted, reviewed and declined months before. The same figures were cropped and re-labeled for this new, unrelated manuscript. Manufactured data. The journal issued a harsh rebuke to the authors, and cc'd the reviewers, thanking us for our time.

I was shocked and saddened. The PIs on the project were top notch in their area. These are competent scientists and noted scholars. These were not sleazy scientists intent to pollute science-space with misleading information, so I could not imagine that they'd deliberately falsify data. My heart sank for them, as I suspected a deeper problem must exist. One of the other authors must have fabricated the figure, jeopardizing the integrity of the entire work.

You'll never believe what happened the next day.

A little backstory. Several weeks before I conducted the aforementioned review, I received an email from a journal indicating that I was named as an author on a manuscript submitted to a journal in my discipline. This was indeed curious. I didn't submit a paper, and while I knew some folks on the author team, I did not see the manuscript or even discuss the project. I contacted the corresponding author and demanded they withdraw the paper and allow me to at least see what I submitted.

The group submitting the work was not in the USA. The lead author was a brilliant student that spent a wonderfully productive year in my lab as part of her degree, taking her coursework in her home country and using time under my direction to forward her research. It turns out that upon returning home she discovered something new in our data, and since she developed the data set under my advising and input, she appropriately included me in the authorship team when publishing the new finding.

But to submit without my endorsement was a massive lapse in protocol, and the manuscript was withdrawn from the journal's review process without hesitation. I soon received a copy of the manuscript and began my critical evaluation. The data looked good but the writing was not acceptable, featuring over-interpretations of what were suggestive, but not conclusive results. It took me two full days to rewrite the manuscript and tone down the hyperbolic language, making it an acceptable paper for publication. It was not big-new science, but certainly an incremental advance. My feeling was that the work would not be acceptable for the journal they chose.

I was right. The day after it was submitted, it was declined. But my heart sank as I read the rejection notice on my smartphone screen. The Editor-in-Chief, a leader in our field and someone I deeply respect, wrote that the paper was rejected because a figure in the submitted work was identical to one that the same group published the year before, only with different labeling. Between cold sweats and a sick stomach, I cradled my head in my hands contemplating this chilling coincidence.

Only the day before the paper I reviewed was declined because of a rigged figure. Now a paper where I was an author was disgracefully disqualified for the same reason.

The queasy feeling quickly gave rise to deep gratitude to the Editor-in-Chief. I don't know that it was standard practice to go back through a group's previous publications, but in this case he did, and he eagle-eyed the doctored figure. I am so grateful that this never was published, or even wasted a reviewer's time. These days I find myself in the cross hairs of activist groups that would celebrate a retraction with my name on it. This editor's oversight saved me tremendous embarrassment and hassle, let alone a permanent asterisk on my reputation.

As for the co-authors, I don't know what happened and I don't care. When they contacted me I did not respond and never will. Someone on that author team put my reputation at risk for a paper of marginal significance. My former student was always trustworthy and I could not fathom this being her doing, but someone on that author team massaged a false figure. When I look at the two figures side-by-side, it is clear, this was no mistake -- this was boldfaced misconduct. I don't know who to blame, and frankly I never will bother to find out. That relationship is over.

These two events in scholarly publishing fraud illuminate several important points. First, for this to happen two times in two days is an incredible coincidence, but underscores that this kind of misconduct must be more rampant than we think. Second, be grateful to the reviewers and editors that keep this deceit out of print, and we all should strive to make such cross checks part of our evaluations. Most of all, know your collaborators, everyone on the author team. In both cases above, it is almost certain that such malfeasance was perpetrated by a single person that took the shortcut of cut-n-paste rather than actually doing the work, without anyone else being wise to the fraud.

Next time, I'll just ask to be thanked in the paper's acknowledgements section, or maybe just ask to be left off altogether.