Australian News

Australian news and media publication

New Australian research has called into question the ethical and fact-checking standards of the world’s most influential medical journals during the coronavirus pandemic.

The report, published today in the Medical Journal of Australia, found that one in five COVID-19 studies published by journals during the first five months of the pandemic had corrections issued after publication.

This compares with 7.4 percent, or roughly one in 13 studies, published at the same time last year.

This is equivalent to tripling the number of errors that go undetected during the generally rigorous review process prior to publication.

On the contrary, no such retractions were made in 2019.

The study’s lead author, Professor Michael Reade of the University of Queensland, told that there was widespread speculation about compromised standards of medical literature in the rush to find a cure for COVID-19, but this report was the first to quantify this.

The results of the Queensland researchers’ analysis were published today in the Medical Journal of Australia.

Novel coronavirus SARS-CoV-2 This transmission electron microscope image shows SARS-CoV-2, the virus that causes COVID-19, isolated from a patient in the US Virus particles are shown emerging from the surface of cells grown in the laboratory. Spikes on the outer edge of virus particles give coronaviruses their name, in the shape of a corona. Credit: NIAID-RML (AP / NIAID-RML)

Five medical journals considered most critical in informing global health policies and clinical practices were examined: The Lancet, New England Journal of Medicine, JAMA, The BMJ, and Annals of Internal Medicine.

The analysis compared 134 research articles published between January 1 and May 31 this year with 54 published during the same period last year.

The findings show that only five percent of the coronavirus trials were randomized controlled clinical trials, considered the “gold standard” of medical research, compared to 35 percent of the 2019 trials.

Professor Reade said this was not surprising, given the short time since the COVID-19 strain was discovered and the requirements for randomized controlled trials.

More unexpected, he says, was the drastically shortened time frame given to review, approve and publish the essays.

The Journal of the American Medical Association was the only publication that published their data on this.

In this journal, the average time from first shipment to publication date was reduced from 139 days to just 23 days.

“This suggests that they weren’t being as thorough and were letting go of things that they wouldn’t normally let go of,” said Professor Reade.

He noted that the number of corrections issued subsequently, usually within a week after the article was published, reinforced this notion.

It also appears that in many cases standard ethical guidelines were not followed.

About half of the COVID-19 studies did not explicitly state that consent was obtained from trial participants.

Several articles also stated that they were granted exemptions from the ethical review requirement due to the nature of the pandemic.

Professor Reade, who also chairs the Defense Force’s Military Medicine and Surgery unit, said it was more critical than ever that medical journals maintain their credibility.

“In the new information age, it’s great that people can spread information very quickly; you can post a document online and read these things very quickly,” he said.

“But the flip side of that is that by the time it gets to a magazine, if the magazines are going to add something to this process, they need to give the stamp of approval that it’s true.”

The Queensland investigators’ report made a series of recommendations on how to address the apparent flaws in the rigor of the investigation during times of crisis.

These include suggestions that journals establish a two-way review process for pandemic and non-pandemic research and the exchange of peer reviews between reviewers and journals.

Leave a Reply

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