A study purportedly of scars left by caesarean sections included women yet to undergo the surgery, say sleuths. But an investigation into the research by the author’s employer and the journal that published it found no evidence of research misconduct.
The paper, published in Wiley journal Ultrasound in Obstetrics and Gynecology, was flagged on PubPeer in February by Ben Mol, an ob-gyn researcher at Monash University in Australia whose efforts have led to scores of retractions and corrections, and Jim Thornton, emeritus professor of obstetrics and gynecology at the University of Nottingham, in the UK. The study looks at how the scar left behind on a woman’s uterus after a c-section is affected by the dilation of her cervix at the time of the procedure.
One of the senior authors of the article, Baskaran Thilaganathan, director of fetal medicine at St George’s University Hospitals NHS Foundations Trust, in London, is currently editor emeritus at the journal. He did not respond to Retraction Watch’s requests for comment.
The other senior author on the paper, Sherif Negm, and the corresponding author Rasha Kamel, who both work the maternal-fetal medicine unit at Cairo University in Egypt, did not reply to requests for comment either.
Mol and Thornton began looking at the study in detail when they noticed its peer-reviewed advance online version reported more than half the 400 study participants had a c-section after previously giving birth vaginally, which is relatively uncommon. So they asked authors to share their data.
Upon inspection of the data, Mol and Thornton found 84 of the more than 400 women involved in the study had ultrasounds to look at their scars before the operation took place, according to the researchers’ post on PubPeer. “Some [of these cases] could be date formatting issues, but this does not explain all,” they wrote.
The pair point out other potential irregularities in the data, including the fact that the information reported in the method section of the paper about the time between c-section and ultrasound scan often did not agree with what was reported in the data. More than half the women also fell into groups who not only had c-sections on the same day but also their subsequent ultrasound scan on the same day as each other, which Mol and Thornton call a coincidence.
Mol and Thornton said they shared their concerns with the authors, who explained some women were mistakenly classified as having given birth before when they had not, but they did not correct this error in the published paper.
The authors also addressed some of their other concerns by making changes to the description of the study design in the final version of the published paper. Mol and Thornton said on PubPeer these changes made the coincidental surgery and scan dates “even less plausible.” The data set in the paper omitted “the most obvious markers of a problem” – the procedure and scan dates, they added.
Anthony Odibo, the editor of the journal, said the publication takes research integrity seriously. “These concerns have already been raised to the journal, which investigated accordingly,” he told us.
The dataset and research have also been investigated by St George’s NHS Trust, Odibo added. “The institutional investigation found no evidence of fabrication or other research misconduct. COPE reviewed the journal’s investigation process and determined that it was appropriate. As is standard, COPE has advised the journal to abide by the outcome of institutional investigation,” he said.
St George’s University Hospital NHS Foundations Trust said in an emailed statement to Retraction Watch: “We have formally investigated these claims, as has the journal, and no misconduct was found in either investigation. Research plays an integral part in what we do and we are committed to upholding high standards to the quality of our research.”
We contacted the University of Cairo through a comment form on its website but did not get a response.
Like Retraction Watch? You can make a tax-deductible contribution to support our work, subscribe to our free daily digest or paid weekly update, follow us on Twitter, like us on Facebook, or add us to your RSS reader. If you find a retraction that’s not in The Retraction Watch Database, you can let us know here. For comments or feedback, email us at [email protected].
I’m incredibly puzzled here. How could the institution look at this impossible data and possibly conclude there was no evidence of fabrication or misconduct? What is their explanation?
And if COPE’s standard is for the journal to abide by the findings of the institution, when we know institutions routinely protect bad actors, then where are we supposed to get accountability when institutions are apparently willing to uphold research lies even in life and death areas like maternal surgery?
I always find this behavior upsetting, but as a woman in America, where reproductive freedom has been sharply curtailed, and where c-sections, maternal mortality, and infant mortality far exceed that of other developed nations, this particular story just chills me to the bone. How are we supposed to have any hope of getting adequate medical care if no one in charge even cares enough to stop fabricated research from becoming part of body of evidence that will form the basis for standards of care? When will we stop abandoning pregnant women to the interests of selfish men?
Is there a funder to appeal to? I can understand why journals and scientific institutions want to look the other way on research fraud, but why aren’t funders taking these issues more seriously? If I was handing out millions in highly competitive grants, I would be incensed to know someone was essentially stealing that money and undermining the actual body of scientific evidence at the same time.
> If I was handing out millions in highly competitive grants, I would
> be incensed to know someone was essentially stealing that money
Or maybe not, if the millions you are handing out are not actually yours. In case, you may desperately need to keep those rich donators happy so that they can donate again if you want to keep your job. And things like “fraud”, “fabrication” or “misconduct” are unlikely to help there.
Or maybe I am just too cynical.
I’m not well versed in publications, but shouldn’t a notice about a potential conflict of interest been noted about one of the authors being an editor of the journal?
Notices are about otherwise hidden conflicts of interest and this one is in the open, so (at least usually) not redundantly mentioned. Reputable journals have procedures to keep the files of manuscripts by one of their editors completely hidden from those editors, too, so if handled well and in a large enough journal is not per se a problem
“Conflicts of interest arise when those who owe conscientious duties to others appear to have personal interests that might tempt them to subordinate those duties to their self-interest.”
Being editor of the journal does not constitute a conflict of interest for the author as such. It would be a conflict of interest in their role as editor, obviously, if they also participated the editorial process connected with the article. This frequently occurs, sometimes with disastrous results, in otherwise reputable journals, regardless of their stated policies. But it is usually noticed only when manifest garbage results.
The “special issue” is the most common venue for systematic abuse of the system in this respect. (There is little oversight.)
Articles written by editors of the publication should be clearly labeled as editorials, not as regular articles. It’s not as if there’s only one journal on a given topic. So expecting editors of journals to submit their own papers to journals other than the one they edit doesn’t seem like too much to ask.
The PubPeer entry is an interesting read as well. Just a quote:
When we drew the authors’ attention to the above, two withdrew. Their names appear on the preprint and, on the PubPeer citation above, but not in the version of record. We learned that one of them had flagged concerns during analysis and write-up.
The corresponding author, Dr Kamel, has another recent paper with a problematic dataset. Details here. https://pubpeer.com/publications/439A70E37294136EFA816F15CD73DB
In that case the problematic dataset is publicly available, as supplementary information, alongside the electronic version of the paper in BMC Pregnancy & Childbirth. Again, I’ve written to the journal. Although BMC Pregnancy & Childbirth is not an obviously predatory journal, it has been slow to process problems, and recently faced a mass-resignation of editorial board members. Details here. https://retractionwatch.com/2024/03/06/editorial-board-members-resign-from-obstetrics-journal-to-protest-handling-of-allegations/
Undeclared conflicts of interest constitute research misconduct, as highlighted in this study (DOI: 10.1080/08989621.2014.958218). This applies to authors, editors, and data sleuths. In a recent publication questioning data integrity by a single author (DOI: 10.1016/j.jogoh.2024.102794), neither Dr. Mol nor Dr. Thornton (coauthors of the publication) declared their conflict of interest with the author who raised concerns about 16 of Dr. Mol’s RCTs, as detailed in a publicly available preprint (https://www.researchsquare.com/article/sr 2823793/v3). Any reader would know that there is undeclared conflicts of intrest by Mol in his article which is likely to affect the reviwers, editors and readers’ ability to give unbiased opinion on the article thwy wrote questioning data integrity by a single author (DOI: 10.1016/j.jogoh.2024.102794). Research integrity rules should be applied consistently, with conflicts of interest declared transparently.