Ob-gyn loses PhD after committee finds he made up research

It was déjà vu last month when a university in Belgium stripped Egyptian physician Hatem Abu Hashim of his doctorate after he was found to have fabricated data in his thesis. 

Just weeks earlier, another Egyptian doctor, Ahmed Badawy, lost the PhD degree he had earned at a Dutch university in 2008. Abu Hashim and Badawy are both professors in the department of obstetrics and gynecology at Mansoura University in Egypt.

According to an investigation by the Vrije Universeit Brussel (VUB), which awarded Abu Hashim his PhD in 2013, the researcher was in “serious violation of scientific integrity” based on “overwhelming evidence of fabrication of statistical outcomes” and “clear lack of statistical proficiency.” 

Ben Mol of Monash University in Australia, a researcher turned data sleuth who alerted VUB and Utrecht University to problems with Abu Hashim and Badawy ‘s research in 2021 and 2020, respectively, told Retraction Watch by email, “The good news is obviously that there is a firm conclusion from both universities after a robust process independent of the complaint.” 

Mol also laid out his concerns in a study published with then-PhD student Esmée Bordewijk and others in 2020, as Retraction Watch reported that year. 

“Yes, it could have been a bit faster, but on the other hand we have this conversation because they took the right decision,” he added.

Abu Hashim’s PhD thesis is based on 11 randomized controlled trials, all of which have been published. Ostensibly, the studies were done at Mansoura University before Abu Hashim enrolled as an external PhD candidate at VUB. 

A report from the Flemish Commission for Scientific Integrity, which gave a second opinion on the VUB findings following a request from Abu Hashim, offers a “credible” scenario for how the 11 papers came about, suggesting:

that Abu Hashim had learned to write medical papers by reading others, that he made up all reported values and that he wrote more papers by adapting previous papers, copying results between articles and applying small alterations (+1 or -1 in some digits).

The commission agreed with VUB that “complete (or virtually complete) fabrication is the only reasonable explanation for the findings.” It also noted that “strikingly,” the researcher did not address any of the allegations against him:

To the contrary, his defence consists mainly of accusing those bringing forward the complaint of misconduct and questioning their work and methods.

Neither Abu Hashim nor Mansoura University responded to requests for comment.

The school, however, has known about Abu Hashim’s fraudulent research for a decade. In an internal investigation from 2014, then-head of department Nasser El Lakany and five other professors found that one of the researcher’s trials had never been done; six trials included an impossibly large number of women with polycystic ovary syndrome; and two reported 366 ovarian-drilling procedures while records were found to exist only for 94. The latter two groups of studies formed part of Abu Hashim’s PhD thesis.

“There is no excuse for the researcher’ [sic] misconduct (fabricating imaginary data and studies not done at all, or studies with doubtful cases not in records),” the Mansoura professors write, according to an English translation of the original Arabic report.   

In 2021, sleuth Nick Brown also began poring over the Egyptian researchers’ work after a Dutch journalist requested his opinion. 

“People don’t read papers. They read the abstract. They say, congratulations, great paper. And then they go back to what they were doing the rest of their day because reading a paper is quite hard,” Brown told Retraction Watch. “I’m not very good at statistics, but I can read a table and things jump out at me.”

Brown quickly realized that Badawy and Abu Hashim’s publications were littered with “fatal flaws.” Virtually all of the P-values were wrong. In some cases, they exceeded 1 – a mathematical impossibility. In others, vastly different values were given for identical statistical tests that by definition should have yielded the same results. 

“I assume the authors were just making up ‘likely-looking’ numbers in a hurry and didn’t realise that these needed to be identical,” Brown said in an email. “We often find that people who cheat are not very good at knowing what genuine numbers should look like.” 

Brown, who himself has an external PhD from a Dutch university, noted that institutions receive the same amount of money from the government whether a PhD candidate is external or internal:

“So someone comes along with some papers already done. They need to write a top and tail of a thesis. They’re probably not going to need a whole lot of supervision.  Exactly how many questions do you ask?” 

A spokesperson for Utrecht University told Retraction Watch by email:

We have asked ourselves the question how this could have happened. Why did the supervisor and the Doctoral Examination Committee not notice this? The articles that were the basis for the thesis, were published in peer reviewed journals. Only much later it came to light that the data underlying these articles had been compromised.

She added that the rules for external PhD candidates have been tightened since 2008, when Badawy obtained his degree (the changes are described here). 

Sam Jaspers, a VUB press officer, told us:

The Vrije Universiteit Brussel is updating its PhD regulations. External PhD students working with existing datasets created at a university other than the VUB and publications reviewed by scientific journals will soon (this spring) be fully audited by the VUB.

Meanwhile, Mol, whose work on various cases recently featured in The Economist, worries about all the fake studies that have not yet been retracted, and the impact they might have on patient care. 

“I cannot understand that … three years after our publication of the Bordewijk study, still half of the Badawy and Abu Hashim studies are out there even without an expression of concern,” he said. “What ideally should happen is that there should be a mechanism that all the journals and publishers bundle their investigation.”

Like Retraction Watch? You can make a tax-deductible contribution to support our work, follow us on Twitter, like us on Facebook, add us to your RSS reader, or subscribe to our daily digest. If you find a retraction that’s not in our database, you can let us know here. For comments or feedback, email us at team@retractionwatch.com.

13 thoughts on “Ob-gyn loses PhD after committee finds he made up research”

    1. It makes you wonder of the general value of a PhD from that university if the dissertation and papers are so inherently useless that no one ever reads them. Why fund stuff like that?

  1. Like the notion that one only ever reads the abstracts and shrugs off the details of the data is so completely foreign to me.

  2. Peer-review is a random procedure. Given the number of journals multiplied by the number of manuscripts divided by the number of good reviewers and the time they can allocate, the probability that both or all 3 referees are knowledgeable about all aspects of the manuscript’s subject is pretty slight. Especially in clinical research, eminent clinicians may not always be eminent statisticians. In a medical faculty, 1 or 2 statistical staff provide statistical service to 100 or so medical researchers. Some clinicians acquire a good expertise in methodology and statistics and also can detect flaws in design and analysis. So here you may have a reason that the number of people who can give a thorough review is pretty low. Even if Ukraine has the technically better army, they are overwhelmed at Bachmut by the sheer numbers.

  3. As a reviewer for many peer reviewed publications, I have observed that lazy and incompetent reviewing is par for the course. I have had editors intentionally ignore known critical flaws in papers that they still choose to publish. They need ink on paper. It’s money.

    1. Absolutely. So many open-access journals including PLoS One accept and publish junk even when I explicitly detail the reasons for rejection.

  4. To avoid all of this fabrications in the future it should be submission of the raw data as pre – assessment of the paper for publications as well as the editorial commette should include a statiscian as a member for statistical analysis of the results.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.