The study was retracted years ago. How did it get into guidelines for pregnant women?

Save articles for later

Add articles to your saved list and come back to them any time.

For her first two pregnancies, Gurwinder* lived in a state of pre-maternal bliss. “You’re just so happy, telling your family you’re having a baby,” the Melbourne-based nurse said.

Her third pregnancy, twins, was different – especially after she was told she had a shortened cervix, which increases the risk of a pre-term birth. “It was very overwhelming,” she said.

How did a retracted Egyptian study end up in Australian clinical guidelines?Credit: iStock

After diagnosis, Gurwinder was offered the hormone progesterone to cut the risk of the baby coming early – and thankfully delivered two healthy babies at full term.

On the face of it, that advice fits with guidelines issued by the Royal Australian and New Zealand College of Obstetricians and Gynaecologists, which note evidence progesterone can cut pre-term birth risk for pregnancies like Gurwinder’s.

But tracing the data behind the guidelines takes you on a strange global journey – in particular, to Egypt and Iran, where thousands of studies are churned out – and raises serious questions about the increasingly obscure scientific foundations our health recommendations are resting on.

The college’s guidelines cite an influential 2017 meta-analysis authored by a global team of researchers and funded by the US government.

The meta-analysis’s key finding is largely based on an underlying study published in the Archives of Gynecology and Obstetrics by researchers based in Egypt in 2016.

In 2021, the journal pulled the paper from the scientific record.

“The authors did not obtain approval from a research ethics committee before conducting this interventional randomised control trial,” a retraction note on the study now said. “Concerns about the data reported in the article are under investigation by the University of Mansoura.” The authors did not agree with the decision to retract the paper, the note adds.

When the retracted Egyptian paper was removed from the American meta-analysis, the results suddenly swung to suggest that progesterone had no discernible effect on preventing preterm births in pregnancies such as Gurwinder’s.

The winding saga is evidence of a dangerous new problem emerging in science, said Professor Ben Mol, head of the Evidence-based Women’s Health Care Research Group at Monash University. Mol first raised concerns about the paper in 2017.

The health sector relies on meta-analyses, which combine smaller underlying studies, to generate high-quality evidence that informs the way doctors treat disease.

But this can also obscure the provenance and quality of those underlying studies. The risk is similar to the way junk mortgages were pooled together into “safe” assets, precipitating the Global Financial Crisis.

“Nobody actually thinks about where a study comes from, how the authors did it,” said Mol. “And patients pay the price.”

Professor Ben Mol is head of the Evidence-based Women’s Health Care Research Group at Monash University, and first raised concerns about the paper in 2017.Credit: Wayne Taylor

A college spokeswoman said the institution is confident that its guidance is correct and informed by robust data, pointing to a different meta-analysis, but said it would be updating the guidelines and was looking at reforms to its review process to help it pick up similar issues.

Roberto Romero, chief of the Pregnancy Research Branch at the National Institutes of Health, told The Age an updated meta-analysis excluding the Egyptian study still favoured progesterone. “To the best of our knowledge, Mansoura University, which oversaw [the Egyptian study], has not concluded there is evidence of falsification or fabrication,” he said.

The Cochrane Collaboration is working to wean out the influence of bias or industry funding from science, and its 7500-plus Cochrane reviews are considered gold-standard evidence.

When Professor Zarko Alfirevic, a women’s health researcher based at the University of Liverpool, applied a test for trustworthiness to 18 reviews, 25 per cent of underlying studies were removed. In a third of cases, the reviews’ conclusions needed to be rewritten.

“Our results are just another wake-up call to the whole medical academic and scientific community, not just Cochrane,” he said.

There is no system for alerting authors of meta-reviews when concerns are raised about an underlying study, meaning they can be unaware of emerging problems.

Another study on progesterone by researchers based in Egypt was included in a systematic review published in May. Journal editors in February slapped an “expression of concern” on the study over worries about the data.

One author of the systematic review told The Age he only became aware of the concerns in the past few days.

“All authors of the paper agree that fraudulent research is not acceptable, and we are working with the editors of the journal of our paper to address this issue,” he said.

Mol is concerned a significant proportion of women’s health research may be based on untrustworthy studies. Common red flags include studies that rapidly recruit hundreds of people, discover very large effects and have tables filled with round numbers or multiples of 10.

This is driven by several countries, including Iran and Egypt, dramatically increasing the amount of research produced – much of it fake, Mol said.

In general, wealthier countries tend to produce more scientific papers. But Egypt and Iran buck the trend, producing more women’s health studies per dollar of GDP than any other nation.

Over the past 20 years, the number of women’s health studies published by Iran has jumped 2160 per cent, overtaking Switzerland, Denmark and South Korea.

Iranian and Egyptian researchers have the same incentives as their international peers: their careers are advanced by publishing as many papers as possible.

But Iran is under economic sanctions and Egypt is sliding towards dictatorship. Large, well-conducted trials are very difficult to do there, Mol said, leaving fraud as the best viable strategy.

Iran and Egypt hold the top two places in the proportion of scientific papers retracted; Stanford researchers in 2019 accused Iran of “systemic academic corruption”.

“The interests of the patient are not at the table,” Mol said. “No one involved has an incentive to discover these things.” There has always been fraud in science, he said, “but not at the scale we see now”.

The authors of the retracted Egyptian study, and the meta-analysis that incorporated it, did not respond to requests for comment. The University of Mansoura also did not reply.

*Gurwinder did not want her surname published.

Get the day’s breaking news, entertainment ideas and a long read to enjoy. Sign up to receive our Evening Edition newsletter.

Most Viewed in National

From our partners

Source: Read Full Article