Until recently, I thought I knew pretty much all the ways postdocs and graduate students are exploited. That was before I encountered an apparently common form of peonage that I had not even imagined: trainees ghostwriting peer-review reports that principal investigators (PIs) send to journals under their own names. When I heard of the practice at a conference this past winter, I was astounded. Could so patent (to me, at least) a breach of ethics really be, as they say, a thing—in fact, a twofer that simultaneously violates fairness and undermines the scientific literature’s integrity?
A recent bioRxiv preprint, which its authors believe to be the first study ever published on the topic, reveals the depths of my naivete. According to survey responses from nearly 500 scientists, primarily in North America and the life sciences, ghostwriting reviews is accepted practice—at least in the sense of being common knowledge, if not in the sense of meeting most people’s ideal standards of scientific integrity.
“When you talk about it, you realize it’s sleazy, but … people just do it without thinking,” says Rebeccah Lijek, an assistant professor of biology at Mount Holyoke College in South Hadley, Massachusetts, and one of the article’s authors. She and her coauthors hope that the study starts a conversation about culture change that will increase transparency of the journal literature and lead to justice and opportunity for trainees.
A learning experience?
Fully 79% of the postdocs and 57% of the graduate students who responded to the survey reported that they “contributed ideas and/or text to peer review reports” their PIs had been invited to write—a practice the authors term “co-reviewing.” Most of the trainees did so five or fewer times, but a handful reported 21 or more such experiences. “These data suggest that collaboration on peer review reports, especially by ECRs [early-career researchers] who are not the invited reviewer, is an academic norm,” the authors write.
Co-reviewing is not necessarily considered a problem by either the survey respondents or the study authors. In fact, it can provide a valuable training opportunity. “People learn how to do peer review through these co-reviewing experiences,” Lijek says. “If I’m the PI, I’m helping you, my trainee, by giving you this opportunity to train in this critical scholarly skill.”
But busy PIs also help themselves by handing off a time-consuming task. About half of respondents reported doing the analysis and writing of a peer-review report entirely on their own, without any input or feedback from their PI—hardly proper “training.”
Even more striking, 46% of co-reviewers said they knew the finished product went to the journal with only the PI’s name on it. This is what the authors define as ghostwriting.
When this happens, the senior scientists are deceiving the journal editors about whose ideas and words appear in the reports, so the editor is not getting what she or he bargained for. Anonymous and unbiased evaluation of journal submissions by expert peers supposedly forms an essential foundation of science. If unidentified junior scientists perform the sacrosanct task of critiquing new research instead of specialists expressly invited for their knowledge, experience, and judgment, who is actually vouching for the soundness of scientific publications? I don’t mean to suggest that postdocs and even many graduate students are incapable of writing worthy reviews. Indeed, they are often more expert in their specific area than their PIs. But they necessarily lack a seasoned lab chief’s historical memory and experience, which may have motivated the editor to invite a particular PI to do the review in the first place. Without knowing who contributed, the editor cannot properly evaluate the review. Just as importantly, submitting ghostwritten reviews robs trainees of the credit they deserve for their work.
The survey “shines a light on a practice that should be abandoned,” writes Ivan Oransky, co-founder of the influential website Retraction Watch, in an email to Science Careers. “It’s one thing if PIs use reviews as a teachable moment, guiding ECRs who may want experience writing them, and then ensuring that they receive credit from journals. It’s another to simply have an ECR write a review and then not indicate that to the requesting editor. That would seem to be plagiarism.” The U.S. Office of Research Integrity (ORI) adds that “ghost authorship is ethically unacceptable” because it misleads the reader “as to the actual contributions made by the named author.”
“Survey respondents strongly agree with the ORI, and specifically believe that ghostwriting of peer review reports is an unethical practice,” the article notes. The power dynamic between PI and trainee, however, often makes trainees feel unable to refuse such requests. “Believing ‘I guess it’s part of my job,’” trainees generally comply, Lijek says.
Beyond forcing trainees into ethically dubious behavior, failure to acknowledge their contributions robs them of opportunities to enhance their CVs or even receive invitations from journals to do reviews themselves. Third-party systems for tracking scientists’ contributions, such as Publons, don’t give credit without proof of who did a peer-review report, says article co-author Gary McDowell, executive director of the advocacy organization Future of Research. Only co-reviewers acknowledged as contributors can provide such proof.
Recommendations for reform
Journals appear inadvertently to encourage ghostwriting by routinely requiring invited reviewers to keep manuscripts confidential. PIs understand that sharing the manuscripts with their trainees likely breaches the journal’s policy. Loath to reveal the deception, “they won’t give credit to the other persons who contributed significantly to the process,” Lijek says.
Moreover, journals’ online platforms rarely provide a clear way to indicate an additional author for a review, implying that editors assume their invited reviewers work alone. However, Oransky says, it’s a “misperception” that invited reviewers have no way to inform editors of reviews’ true authors. “Even if the peer-review form doesn’t include an explicit field,” one can always “include a comment in the confidential notes to editors. I do that frequently.”
Instead of maintaining a rule of confidentiality that is routinely ignored in practice, journals should recognize the reality of widespread, clandestine collaboration by creating and publicizing policies that clarify trainees’ roles in reviewing, the article recommends. A journal could, for example, let invited reviewers openly share manuscripts with trainees who contribute ideas or text, with all participants’ names and contributions disclosed when the review is submitted. The authors favor this approach, writing that “adding co-reviewer names does not diminish the quality of the peer review report nor does it diminish the important role of the invited reviewer, akin to a senior author on a manuscript.”
A journal that does not want junior scientists to contribute could, on the other hand, choose to allow them to read manuscripts for training purposes but restrict writing the review to the invited reviewer. Or a journal could continue to require strict confidentiality. Whatever policy a journal chooses, stating it explicitly and prominently will clarify both ethics and authorship.
For trainees, PIs, and journal editors alike, Lijek says, “it would be quite a freeing experience”—and a whole lot fairer and more honest—“to have the policies and the culture change.”