2 August 2019

BMJ joins the medical pre-prints game

KnowCents Research

Ever so cautiously, the medical research world is starting to embrace pre-print archives, with The British Medical Journal launching a pre-print server, called medRxiv, in June.

The website, medRxiv (pronounced “MedArchives”), shares medical research papers that have not yet been peer reviewed. There’s no paywall, and anyone can access the papers.

The website was launched in partnership with Yale University and Cold Spring Harbor Laboratory in New York, which also runs bioRxiv.

“A number of people and research funders are promoting pre-prints as a way to speed up the research enterprise,” Dr Theodora Bloom (PhD), the executive editor of the BMJ, told the World Conference of Science Journalists in Lausanne, Switzerland last month.

“And that’s really where I come at it. I used to be a researcher. I’m all about trying to make researchers communicate with each other better.

“And I rather like the notion that as journalists you are eavesdropping on that conversation, because that’s exactly how I view it as well.”

Pre-prints were helpful to researchers because they allowed them to crowd-source informal feedback on their paper before submitting to a journal, she said.

“And we all know peer review is flawed. So, more feedback is likely to improve their papers more.”

Posting unpublished research to online archives was also way for researchers to “Get a date stamp: ‘This is when I did this bit of work and shared it with the world’,.” she said.

“And there has been so much worry about who is going to scoop whom within the research enterprise, but I think that’s sort of helping,” Dr Bloom said.

Around two-thirds of pre-print papers are peer reviewed and published within a year or two, according to Dr Bloom.

Pre-prints could also be invaluable during a public health crisis, she said.

“Things like the Ebola and Zika crises have made people think that the three or four months that it can take even for a fast-tracked research paper to be published in a high-profile journal is too long when there are emergencies,” she said.

The BMJ had put some basic safeguards in place to ensure that papers weren’t published on the website if they could pose a threat to public health, she said.

Papers submitted to medRxiv aren’t posted until they are checked by volunteer researchers and a professional medical editor.

The website also prominently displays a caution – in red – that pre-prints are “preliminary reports of work that have not been peer-reviewed” and “should not be relied on to guide clinical practice or health-related behaviour and should not be reported in news media as established information”.

However, many in the science communication industry have expressed concern about pre-prints for medical research.

Speaking at the conference, Tom Sheldon, senior press manager at the Science Media Centre in the UK, said journalists were likely to use medical research pre-prints as a source of stories and that this could affect public health if the research turned out to be inaccurate.

The BMJ’s initiative follows in the footsteps of physicists and mathematicians who have been sharing unpublished pre-prints on arXiv since 1991, and biologists who started posting to bioRxiv in 2013.

These archives have grown in popularity, with bioRxiv boasting just under 1,500 submissions per month last year. arXiv is now publishing more than 10,000 papers per month.

The BMJ actually launched its first pre-print server two decades ago, but there proved to only a limited appetite for the service at that time, Dr Bloom said.

“We first started a pre-print server in 1999 and we closed it in 2005 with a total of about 18 pre-prints having been submitted over that time,” she said. “And I think that’s pretty much the experience that others had at that time.”

The Lancet also launched a pre-print service in 2018 in collaboration SSRN.