Six ‘Superretractors’ Responsible for Large Number of Retracted Clinical Trials

A small number of “superretractors” was responsible for a significant proportion of retracted clinical trials, a retrospective cohort study showed.

Just six superretractors accounted for about a fifth of all retracted randomized controlled trials (RCTs), according to Ioana Alina Cristea, PhD, of the University of Padova in Italy, and colleagues.

Eighteen highly cited scientists who had more than 10 retractions during their career accounted for a quarter of all retracted trials; five of these authors were also superretractors, they reported in JAMA Network Open.

“The disproportionate role of superretractors and top-cited authors with multiple retractions, their interconnections as expressed by co-authorships, and the narrow concentration may provide useful leads to help unravel fraudulent literature at scale,” Cristea and colleagues wrote.

Of the six superretractors, three were in anesthesiology and three were in endocrinology and metabolism. All but one were affiliated with Japanese institutions. One superretractor, Joachim Boldt, PhD, was from Germany.

Concerns about one superretractor, Yoshihiro Sato, MD, had been raised almost 10 years before the first retractions occurred, coming in the form of letters to the editor and a systematic investigation of his work by other researchers that took almost 3 years to get published, the researchers noted.

Investigation of multiple articles from Sato led to the discovery of another superretractor, Jun Iwamoto, MD, who was a frequent co-author on Sato’s work. Another frequent co-author, Kei Satoh, MD, was among the top-cited authors who had at least 10 retractions in their career.

In an accompanying invited commentary, Chang Xu, PhD, of Naval Medical University in Shanghai, and colleagues said the study “exposes a notable feature that superretractors act as superspreaders in perpetuating research misconduct,” and that the high number of highly cited authors involved in retractions “sets a negative example for the global scientific community that repeated misconduct does not prevent someone from becoming a top-cited scientist and may drive other scientists to imitate such behavior for personal gain, leading to more scientific fraud.”

Xu and colleagues called for efforts to weed out potential superretractors before they get published, including more screening of authors and publishers adopting integrity checks. Additionally, long delays in the retraction process are not acceptable, they said.

But Ivan Oransky, MD, who runs the Retraction Watch “leaderboard” cited in the paper, noted that it’s not surprising that retractions would cluster this way, and that more effective prevention methods are needed.

“What tends to happen is once somebody has a couple retractions, everybody scrutinizes the rest of their work,” Oransky told MedPage Today. “Once people start investigating, especially in a relatively small field like anesthesiology, you just get a pile-on, and I mean that in a good way. People find the problems and a lot gets retracted.”

“Of course, it’s going to look like you’re having an outsized impact on citations and the literature in general,” he added.

Thus, screening authors may not be so effective, he said. Instead, adopting a program like INSPECT-SR, which assesses the trustworthiness of randomized controlled trials, and other uniform checks that aren’t gameable, would be a better move.

For their study, Cristea and colleagues examined data from the VITALITY cohort, which includes 1,330 retracted RCTs. They then developed three groups of researchers with multiple retractions: those with the highest number of retractions on the Retraction Watch leaderboard (30); top-cited scientists with 10 or more retractions during their career (163); and top-cited scientists in 2024 with 10 or more retractions (174).

The study was limited because final retractions are only a fraction of the literature that should be retracted, the researchers said. Additionally, the VITALITY cohort includes trials that were retracted for any reason, not just data fabrication or severe flaws, which precluded the use of retraction as a proxy for misconduct or fraud. However, they said, all three lists had authors with multiple retractions that were not due to publisher or journal error, “thereby increasing the likelihood that their accumulated retractions reflect at least inadequate research practices, if not overt misconduct.”