Publication Fraud, Peer Review, and Preprints
Myron Yaster MD, Justin L. Lockman MD, MSEd, and Mark Schreiner MD
Editorial
Kharasch ED. Scientific Integrity and Misconduct-Yet Again. Anesthesiology. 2021 Sep 1;135(3):377-379. PMID: 34329383
Society for Pediatric Anesthesia 35th Annual meeting
DiNardo JA: The Future: Information Dissemination in a Rapidly Changing World, Are our Journals up to the Task? (Available online for registered attendees)
The whole point of the Pediatric Anesthesia Article of the Day (PAAD) is to alert, summarize and inform you of new information that can stimulate your thinking with new ideas and hopefully change and move your practice forward. We (the PAAD’s executive council) specifically scan and pick articles form major, peer-reviewed journals like Anesthesiology, Anesthesia and Analgesia, J Pediatric Anesthesia, JAMA, and NEJM. But what if the published information is fraudulent? And what about “preprint” information, which was the focal point of Jim DiNardo’s recent lecture at SPA’s 35th Annual Meeting last month.
For those of you, like us, who had never heard of a preprint (or if you were not able to attend the SPA meeting), preprints are full draft research papers that are shared publicly before they have been peer reviewed and are specifically designed to bypass the delays in publication in traditional journals. In some ways preprints are like abstracts – but on steroids. Unfortunately, by bypassing peer review, if the preprint information is fraudulent or just turns out to be wrong, it will still enter into the public sphere as if it was peer reviewed and become “truth.” Ultimately, these “truths” spread like wildfire in social media and in our journals and have the very real potential of harming people. A good example of this, as discussed by Dr. DiNardo, was a preprint about convalescent plasma for COVID-19 that resulted in early misinformation (and even treatment of some world leaders with the therapy) before peer reviewed data showed no benefit. In fact, the plasma article, still not peer reviewed, has now been published in seven different versions. https://www.preprints.org/manuscript/202004.0097/v4
Today’s PAAD, an editorial by Kharasch, the editor-in-chief of Anesthesiology, once again addresses the issue of fraudulent research in anesthesia publications. As discussed in a previous PAAD “What’s in a name? It’s Fraud… not ‘Publication Misrepresentation’ [September 23, 2021],” I (Myron) wrote that of the many lessons I Iearned from my mentor, Dr. Richard Traystman of blessed memory, perhaps the most important concerned research and personal integrity. Dick was always worried about scientific fraud and stressed the need to verify data and publications in research and for grant funding decisions. Indeed, I can still hear his voice in my head as I write this: “How can this be?” Unfortunately, the problem of research fraud is getting worse. Today’s editorial underlines the tsunami of scientific misconduct in our field and Kharasch asks the question “why does scientific misconduct exist?” He writes: “It has happened again…A special committee of the Japanese Society of Anesthesiologists (Kobe, Japan) recently found Dr. Hironobu Ueshima, a Japanese anesthesiologist, was guilty of fabricating data and other misconduct in 142 publications (12 original papers, 9 case reports, 1 in related fields, and 120 letters). Last year, worldwide, more than 2,300 articles were retracted, an increase from just 38 in 2000…the problem is particularly common in papers from China”.1 Importantly, these fraudulent papers have major impacts on systematic reviews and meta-analyses, from which many important policy decisions are made.
Why is this happening? As discussed in the PAAD in September as well as one from August 16, and reiterated by today’s editorial: “Academicians feel pressure by the processes of institutional appointment and promotion, which require publications for jobs, compensation raises, retention, or advancement—even for those whose occupation, skills, and interest are not scholarship but, for example, clinical practice [or education]. Institutions themselves may be just as culpable, as they too seek credit, reputation, status, and the financial gains therefrom in terms of extramural funding and donations”.2
Even when fraud is not involved, readers need to be aware that most published research findings are likely to be false.3,4 Even when the reported association is true, the effect size in early publications is likely inflated 5 and will subsequently be shown to be much smaller than initially reported. As clinicians, we should be especially wary of studies with small sample sizes, large effect sizes, and reports of new discoveries.
What can we do? To paraphrase Dr. DiNardo, the peer-review process is our goal line defense. So, if you are asked to review a paper by a journal, please take on this responsibly as a public service and as your commitment to integrity and truth. And for those of you on recruitment and/or promotions committees, please be sure to question everything and assume nothing! Remember the words of Justice Louis Brandeis: “Sunlight is the best disinfectant”.
References
Shi Q, Wang Z, Zhou Q, Hou R, Gao X, He S, Zhao S, Ma Y, Zhang X, Guan Q, Chen Y: More consideration is needed for retracted non-Cochrane systematic reviews in medicine: a systematic review. J Clin Epidemiol 2021; 139: 57-67
Kharasch ED, Avram MJ, Bateman BT, Clark JD, Culley DJ, Davidson AJ, Houle TT, Jiang Y, Levy JH, London MJ, Sleigh JW, Vutskits L: Authorship and Publication Matters: Credit and Credibility. Anesthesiology 2021; 135: 1-8
Ioannidis JP: Why most published research findings are false. PLoS Med 2005; 2: e124
Ioannidis JP: How to make more published research true. PLoS Med 2014; 11: e1001747
Ioannidis JP: Why most discovered true associations are inflated. Epidemiology 2008; 19: 640-8