The best retractions of 2021
SSince the start of the pandemic, journals have retracted more than 200 articles related to COVID-19 and counting, most of them in 2021. But those items only represent about 5% of the more than 3,000 retractions we’ve indexed this year in the Retraction watch Database. In what has become an annual tradition, here we feature the best retraction stories of the year.
1Like many people, Victor Grech, a pediatric cardiologist in Malta, is very fond of Star Trek. The problem is that Grech managed to turn an Elsevier journal called Early human development into a kind of science fanzine, publishing dozens of articles for the periodical that were in a galaxy far, far away from the scope of his editorial interests. The publisher learned of the problematic articles in late 2020 through Hampton Gaddy, an undergraduate student at the University of Oxford in the UK. Grech’s articles covered topics such as the role of nurses in Star Trek, the banality of evil in Star Trek, and the portrayal of doctors in, you guessed it, Star Trek. Grech finally lost more than two dozen articles to withdrawal.
See “When researchers sound the alarm on problematic articles”
2In 2015, officials at the University of Colorado at Denver concluded that one of its former faculty members, Hari Koul, had to correct or retract nine articles due to concerns about problematic images in articles. But six years later, most of these articles remained intact– and many newspapers involved said they had never heard of the investigation. After Retraction watch reported the delay, newspapers drawn three items by Koul, who had moved from Denver to Louisiana State University Health Science Center (LSU HSC) in Shreveport and eventually ended up at LSU HSC New Orleans. Then after local media reported on others allegations Retraction watch had mentioned, LSU HSC New Orleans said it was investigating, and Koul resigned from his position as department head.
3When the review Vaccine published a study in June claiming that COVID-19 vaccinations killed two people for every death they prevented, the scientific community was outraged. Two members of the journal’s editorial board resigned in protest against the article written by Harald Walach, described on his Wikipedia page as a “parapsychologist and advocate of alternative medicine”. Vaccine quickly issued a expression of concern for paper and beyond retracted it. Meanwhile, Walach, whose institution in Poland terminated his post in response to the controversy, defended his group’s analysis, saying the data, while imperfect, was analyzed correctly. He also has lost another paperin JAMA Pediatricson COVID-19 and masks for children.
4Last year, scientists began to express doubts on the validity of the data they had received from Jonathan Pruitt, a behavioral ecologist in a prestigious position at McMaster University in Canada, whose field research on spiders had helped underpin numerous publications. Pruitt’s papers soon began to fall, and the following year he lost a dozen papers. At the end of that year, Pruitt’s doctoral dissertation, which he had received from the University of Tennessee at Knoxville, has been removed. Pruitt was placed on paid vacation of McMaster and removed from the prestigious Canada 150 Presidents website.
5When Cyriac Abby Philips, a gastroenterologist in India, published an article in 2018 about a young woman who suffered from liver disease after taking herbal supplements, he didn’t think three years later , he would consider suing the newspaper for defamation. Philips’ legal troubles began when he and his colleagues published their case study in the Journal of Clinical and Experimental Hepatology, an Elsevier title. Herbalife, which makes dietary supplements, including those the patient took, lobbied the newspaper, which ultimately decided to withdraw the work for “legal reasons”, as stated in the original retraction notice. This review was later amended to say that “the scientific methodology, analysis and interpretation of the data underlying the article were insufficient” – a claim which Philips called “highly defamatory”. He threatened to sue the publisher and the newspaper for the equivalent of US$1.35 million. The retraction notice was soon amended again, and now cite legal pressures as it did initially.
6Retractions often take years, but not in this case. Barely a month after the publication of an article claiming that female scientists do better with male mentors, Nature Communication removed the item amid a storm of criticism. Written by a group from the Abu Dhabi campus of New York University, the document was lambasted as soon as it was posted online in mid-November. As one statistician tweeted, the article “doesn’t tell us much about the impact of gender on mentoring, but it certainly tells us that the statistical community needs to do a better job teaching scientists about correlation, causality and confusion”. The authors said they agreed with the journal’s decision and said they felt “deep regret that the publication of our research has both caused pain on an individual level and triggered such a response.” deep within many members of the scientific community.
7Pierre Kory, then of the University of Wisconsin School of Medicine and Public Health, testified before Congress in May 2020 that MATH+ – a critical care regimen that includes methylprednisolone, ascorbic acid, thiamin, heparin and co-interventions – reduced the risk of death from COVID-19 by 75% compared to other regimens. Then, last December, he and his colleagues published an article in the journal of critical care medicine about MATH+ (to which they later added the controversial drug ivermectin) saying as much, prompting other experts to question whether the approach’s effectiveness was overstated. These concerns seem justified. In November, the magazine retracted Kory’s paperciting inaccurately reported data from one of the study sites in the analysis.
8At the end of 2020, the magazine Eurosurveillance announced that in response to an international petition, it was while looking closer in an article she had published earlier this year on the validity of PCR testing for SARS-CoV-2 (then called 2019-nCoV). The news encouraged critics of the article, who argued that PCR tests were not able to identify the virus – and therefore positive tests were meaningless and should not be used to guide the public policy, especially economically damaging measures such as lockdowns. But two months later, the publishers published a declaration saying the paper would hold (or more accurately, “the criteria for a retraction of the article are not met”).
9Proponents of using ivermectin to treat COVID-19 have little solid evidence to support their belief that the deworming drug is effective against the infection. A study that many ivermectin fans referred to this year appeared in Virus in spring. The randomized controlled trial reportedly found that a single dose of the drug resulted in “fewer symptoms, lower viral load and reduced hospital admissions”. Except it wasn’t true. As BBC News reported, the study “turned out to contain blocks of details from 11 patients that had been copied and pasted repeatedly, suggesting that many of the apparent patients in the trial did not really exist.” The authors acknowledged that they had mixed up their data files and, in November, the review removed the paperbut not before the study became part of a meta-analysis on the virtues of ivermectin for COVID-19, which to date remains uncorrected.
tenFinally one of our favorites of the year. the Arabic Journal of Geosciences was forced to retract 44 items of a special issue after readers pointed out that they seemed like absolute gibberish. The first clue? The titles read like a group of graduate students playing Mad Libs drunk: “Neural network-based urban precipitation trend estimation and adolescent anxiety management”; “Seismic activity distribution in mountainous area based on embedded system and basketball fitness detection.” A guest editor for the journal, which is owned by Springer Nature, at one point blamed an email hack for the nonsensical articles. In fact, the 44 was just the tip of the sand dune for Springer Nature. More than 400 articles in company-owned journals and hundreds more in reviews owned by Elsevier – have been flagged for similar issues.
Correction (January 3): This article originally referred to LSU HSC Shreveport and LSU HSC New Orleans as campuses of the same university. Although within the same university system, they are separate institutions. The scientist regret the mistake.