Knowledge integrity issues flagged in 130 girls’s well being papers — all by one co-author

A workforce of scientist–sleuths has flagged data-integrity issues in 130 research authored by the identical biomedical researcher, a specialist in girls’s well being and gynaecology, and his colleagues. The sleuths revealed their findings in a peer-reviewed paper earlier this yr¹.

A number of the research that have been recognized as probably problematic have been cited by different researchers or included in analyses that might inform scientific observe. The variety of papers being questioned is among the many highest by a still-active life-scientist, say some specialists.

The 130 research have been revealed between 2014 and 2023 and report the outcomes of scientific trials and different analysis on maternal and girls’s well being. The highlighted issues embrace oddities in reported statistics, unfeasible outcomes and textual content that’s an identical to different papers. Ahmed Abbas, an obstetrician and gynaecologist at Assiut College in Egypt, is listed as a co-author or corresponding creator for all 130 articles. Abbas didn’t reply to Nature’s request for remark.

Biomedical paper retractions have quadrupled in 20 years — why?

A number of the papers stay a part of the literature. Eleven have been retracted. Earlier than it was retracted, a type of 11 was included in a 2019 meta-analysis on a therapy to forestall miscarriage. The retractions of the paper by Abbas and his workforce and one other, unrelated paper will in all probability change the conclusion of the evaluation, says one of many 2019 work’s authors.

The inclusion of a probably unreliable examine in a systematic assessment can have dangerous penalties, as a result of “it will probably instantly have an effect on how a surgeon or an [obstetrician–gynaecologist] is doing their job”, says James Heathers, a forensic meta-scientist at Linnaeus College in Växjö, Sweden, who was not concerned with the investigation that recognized the data-integrity issues.

Ladies’s well being specialists are actively growing methods to forestall the publication of questionable knowledge. However they are saying that when these papers are revealed, it’s tough to purge them from the literature.

Alaa Mohamed Ahmed Attia, the dean of the School of Drugs at Assiut College, with which Abbas is affiliated, didn’t reply to Nature’s request for a touch upon the issues raised about Abbas’s publications on this yr’s peer-reviewed paper.

Rejection and retraction

The 130 flaggedstudies have been described in a paper revealed in Could¹ within the Journal of Gynecology Obstetrics and Human Replica by obstetrician and gynaecologist Ben Mol, at Monash College in Clayton, Australia, and his colleagues.

In 2016, Mol peer-reviewed an unpublished manuscript co-authored by Abbas a few scientific trial of the hormone progesterone to forestall miscarriage. Mol seen discrepancies within the paper and notified the journal, he says. The journal rejected the work by Abbas and his workforce. However in 2017, a special journal, The Journal of Maternal-Fetal & Neonatal Drugs, revealed a model² of the manuscript that included adjustments to the sections that Mol had flagged, he says. The journal in the end retracted the paper in December 2019.

Based on the retraction discover, the journal’s editors-in-chief learnt that earlier variations of the manuscript “confirmed important adjustments to the underlying knowledge.” The discover additionally stated that when contacted, the authors couldn’t present the unique knowledge to confirm the outcomes. Based on the journal’s writer, Taylor & Francis, issues concerning the paper have been first raised in February 2019. The ensuing investigation led to the article’s retraction later that yr, the writer says. Abbas didn’t reply to Nature’s request for remark concerning the retraction.

Large database

Mol’s workforce determined to survey all papers by Abbas aside from literature evaluations, case stories and research finished as part of a global collaboration. They recognized 263 papers that included Abbas as an creator. These research collectively enrolled greater than 74,000 individuals between 2009 and 2022.

Of the 263 research analysed within the paper, 130 — virtually half — raised the sleuths’ issues. A number of the papers had statistics that appeared unfeasible. One used wording that was just like that of a beforehand revealed paper. The articles that the workforce flagged appeared in journals produced by a number of publishers equivalent to Taylor & Francis and Springer Nature, which additionally publishes Nature. Nature’s information workforce is editorially impartial of its writer. When requested for remark by the information workforce, Springer Nature didn’t reply.

The sheer variety of research that have been claimed to have been produced in such a brief time frame caught the eye of Mol’s workforce. Based on the reported registration and publication timeline of the papers, in Could 2017, Abbas would have been conducting 88 simultaneous scientific research. Catherine Cluver, a gynaecologist and obstetrician who leads the preeclampsia analysis unit at Stellenbosch College in South Africa, agrees with Mol’s workforce that it appears unfeasible to conduct such numerous research at one time. “Doing all the regulatory work, the ethics approvals, ensuring the trials are being run appropriately … I believe there is no such thing as a manner you possibly can do greater than 4 or 5, and even then, it’s a push,” she says.

Concern over numbers

A standard problem recognized by Mol and his colleagues was statistical oddities. One paper they flagged, revealed within the journal Proceedings in Obstetrics and Gynecology³, evaluated the impact of the remedy esomeprazole in girls with the being pregnant complication preeclampsia. The sleuths famous that the final digit of 31 of the 32 values in tables 2 and three, together with means and customary deviations, are even numbers (see ‘Even numbers abound’). In scientific knowledge, the digits of such measurements and statistical outcomes are typically extra equally distributed between odd and even numbers, so the possibility of getting so many values ending in even numbers can be low. The numbers are a “concern”, in response to the paper by Mol and his workforce.

The tables additionally function quite a few pairs of numbers which have an identical digits after the decimal level — for instance, 0.76. A number of the repetitious values are in the identical desk; some are break up throughout the tables. This, too, is regarding, says the paper by Mol and his workforce.

These uncommon numbers ought to compel the authors to current their uncooked knowledge, says Nicholas Brown, a psychologist and research-integrity specialist at Linnaeus College.

The editor-in-chief of Proceedings in Obstetrics and Gynecology, Donna Santillan, stated in a press release that every one inquiries about analysis or publication misconduct are investigated by the journal. Santillan, a reproductive sciences researcher on the College of Iowa in Iowa Metropolis, declined to touch upon whether or not this examine is at present being investigated, citing privateness issues.

Persevering with investigation

Different research flagged by Mol’s workforce describe apparently unbelievable outcomes. In a 2020 survey⁴ in The European Journal of Contraception & Reproductive Well being Care that assessed the attitudes of obstetricians and gynaecologists in Egypt in direction of abortion, for instance, the imply age of physicians surveyed was 42.6, and their imply variety of years in observe was 26.4. For these numbers to be right, the imply age at which these physicians began practising can be 16.2. The identical paper accommodates phrases which can be an identical to these of a examine⁵ revealed in 2009 by completely different authors (see ‘Textual echo’).

The journal’s writer, Taylor & Francis, says that it’s at present investigating the paper, after issues have been raised in December 2023. Abbas didn’t reply to a request for remark concerning the investigation.

Mol says that he’s not accusing the authors of information fabrication and it’s potential that the discrepancies are a results of unintentional errors. “We’re simply presenting the information after which different folks can draw a conclusion.”

Scientific-trial guidelines

Some publications focusing on girls’s well being instructed Nature that they’re actively working to maintain problematic analysis from being revealed. For instance, a bunch of journal editors are combating in opposition to knowledge falsification within the subject of obstetrics and gynaecology by sharing details about probably flawed papers. The group additionally drew up a guidelines of seven necessities that randomized managed trials should meet to be revealed, equivalent to ethics-committee approval. If a trial’s authors don’t fulfil these necessities, “we’re not going to publish it”, says Vincenzo Berghella, editor-in-chief of the American Journal of Obstetrics & Gynecology Maternal-Fetal Drugs and a maternal-fetal specialist at Thomas Jefferson College in Philadelphia, Pennsylvania.

If problematic research do find yourself in journals, investigating them post-publication is usually a “painstakingly tough” course of, says Žarko Alfirević, a specialist in fetal and maternal drugs on the College of Liverpool, UK. “The burden of proof must be enormously excessive” for journals to confess that fraud has been dedicated, he says.

To mitigate the injury of problematic research within the medical literature, Alfirević, who’s an editor at Cochrane, a bunch that evaluations medical proof, is pushing for the adoption of trustworthiness assessments of randomized managed trials as a situation for authors to incorporate them in systematic evaluations.

Downstream impact

The chance of flawed papers affecting medical care is actual, says Mol. One instance is the 2017 examine by Abbas and his colleagues on using progesterone to forestall miscarriage and the 2019 systematic assessment wherein the examine was included. That very same assessment by Cochrane additionally integrated a second examine, authored by a special group, that was additionally subsequently retracted. Each papers contributed to the assessment’s conclusion that progesterone dietary supplements would possibly scale back the danger of miscarriage in girls who’ve skilled recurrent miscarriages. The assessment has been cited in ten scientific pointers.

Now it’s clear that, regardless of what the retracted research steered, the dietary supplements will not be efficient for all girls who’ve skilled recurrent miscarriages⁶. The assessment’s corresponding creator, David Haas, an obstetrician and gynaecologist at Indiana College in Indianapolis, says that it’s “extremely possible” that the 2 retractions will change the assessment’s conclusion. He and his colleagues are actually working to publish an up to date model of the assessment wherein the retracted research have been eliminated. A discover on the present on-line model of the assessment says that the assessment authors have been suggested that the examine by Abbas and his colleagues is the topic of investigation and the assessment workforce has moved the examine from ‘included research’ to ‘research awaiting classification’.

How papers with doctored photos can have an effect on scientific evaluations

One other assessment that included a paper authored by Abbas and his colleagues can also be being up to date. The meta-analysis⁷, revealed in 2023, analysed papers on a technique combining progesterone and a process on the cervix to forestall pre-term beginning and concluded that the mixture may very well be profitable. Among the many papers analysed was a examine⁸ by Abbas and his co-authors, which was revealed within the Worldwide Journal of Gynecology & Obstetrics in 2020.

The journal retracted the paper in late 2023, noting that “inconsistencies have been discovered inside the dataset … which name into query the validity of the information.” The authors of the meta-analysis say they’re conscious that Abbas’s paper has been retracted and they’re about to submit an amended model that excludes the retracted work. “Thankfully, eradicating this paper from our meta-analysis has not influenced the first end result,” says corresponding creator Craig Pennell, an obstetrician and gynaecologist on the College of Newcastle in Australia.