The increasing challenges to publication integrity in medicine and basic sciences
Briefing: As this article will demonstrate, there is increasing concern in science circles about the integrity of the academic publication process. Over the last month the subject appears to have reached new levels of attention, as suggested by several important articles in prominent journals. This article summarizes those opinions.
It surely looks like things are getting worse by the month, maybe even by the week, as medical and basic science journals are full of articles lamenting a large variety of problems with scientific publishing. Nature magazine, for example, recently noted that “investigations by journals of potential unreliable research are often superficial, opaque, and prolonged.”1 The article further suggested that “changing the guidance of the Committee on Publication Ethics could tighten up the process.”
An article in the BMJ suggested that “urgent steps must be taken to reduce misconduct, restore trust, and protect patients.” The article also noted that many countries have no laws or legally enforceable standards that govern research integrity, therefore, relying on national voluntary codes of conduct, quite obviously not sufficiently effective. An article in JAMA suggested that proposals from the Office of Research Integrity (in the U.S. Department of Health and Human Services) should prepare the academic community for more regulatory directives and less autonomy in reviewing data integrity allegations.3
The Wall Street Journal recently covered the subject on its frontpage by reporting extensively about a debunked cancer study.4 Under the leadership of a very prominent physician scientist from a reputable academic institution, it was published several years ago in Nature magazine (likely the world’s most prestigious science journal) and reverberated through oncology as well as microbiology because it claimed that cancers have distinct microbial signatures, which should permit diagnoses through blood tests. It also reverberated through the finance community because a start-up company (Micronoma) raised $17.5 million under the assumption that – based on the article – blood tests could, indeed, be developed to diagnose specific cancers.
Unsurprisingly, the paper was cited in other publications in the ensuing four years no fewer than 600 times before finally being retracted in June of this year. As one scientist was quoted in the Nature article, “this paper polluted the literature.” The fall-out was considerable: journals have started reviewing eight other studies published since the original paper was published because they all, of course, relied on the original paper. Further retractions can be expected. And the startup shut down!
Another recent article in Nature reemphasized the point that bad science often gets propagated through the literature.5 The author – described by the magazine as a “scientific sleuth” and professor of computer science at the University of Toulouse, in France, in his article suggested that readers often are not aware when a paper cannot be trusted, a point we, of course, fully agree with. He furthermore suggests that adopting some easy-to-use technological solutions can help researchers, publishers, and referees “to clean things up.” For example, he points out that nonsensical phrases in papers can sound alarm bells.
And demonstrating how seriously Nature these days is taking this subject, another article in this magazine that pointed out how blind some journals (or their editors) are to citations of already retracted papers.6 Heavily building on some of the suggestions in previously cited paper,5 the authors put together a list of the 10 published papers with the highest percentages of withdrawn papers cited in their list of references: Astonishingly, it went from 21% to 65% of the respective papers’ reference list.
And then there was yet another, even harder-to-believe news item that crossed the wires and it, again, appeared in Nature magazine: research integrity “watchers” recently discovered schemes that sell fake citations, which allows researchers to artificially inflate their citation counts, which academic institutions increasingly consider in promotion assessments.7
Related and clearly reflecting the same disturbing mindset is also the subject of ever increasing so-called predatory scientific conferences. This, too, has recently been a repeated subject of concern in the literature. Again, an article in Nature magazine – this time even in the format of an unsigned editorial –addressed this issue in some detail, making the point that these kind of exploitative meetings especially target vulnerable early-career researchers and warrant much more awareness by the scientific community and possibly even legal redress.8
In short, the scientific enterprise must start cleaning house.
References
1. Grey et al., Nature 2024;632:26-28
2. Bouter L. BMJ 2024;386 doi://https://doi.org/10.1136/bmk.q1595
3. Caron et al., JAMA 2024;332(3):193-194
4. Subbaraman N. The Wall Street Journal. August 31-September 1, 2024. https://www.wsj.com/health/healthcare/cancer-study-retracted-research-fallout-9573f842
5. Cabanac G. Nature 2024; 632:977-979
6. Van Noorden R, Naddaf M. Nature 2024;633:13-15
7. Singh Chawla D. Nature 2024;632:966-967
8. Editorial. Nature 2024;632:7