By Sarah-Jane O'Connor 06/06/2019

A few years ago I was talking to a physicist at a work function when he told me about arXiv (I made him spell it out for me so I could look it up the next day).

You mean, people put their research papers up online for free, for people to read them before they’re peer reviewed and you don’t have to wait for the journal publication that will probably be locked behind a login? *Head explode emoji*

Now that I’m no longer at a university, I’m fortunate to have access to journal articles through work, but if I was a regular Jo Citizen much of the research I’m interested in would be locked away where I can’t access it. That’s not good for scientific literacy as a whole, nor would it be very easy to keep up to date on research for someone who was temporarily not affiliated with an institute, for instance between short-term contracts or during parental leave.

So there’s arXiv (pronounced archive), bioRxiv and just announced today, medRxiv – soon-to-be launched by Cold Spring Harbour Laboratory, Yale University and the BMJ. (Cold Spring created bioRxiv in 2013.) Which has got me thinking about the pros and potential cons of making more and more research available before peer review.

Media access to research

Allow me to take a detour to discuss a movement to improve reporting on health and medical research. I’ve written previously about research examining the role press releases play in hyping up health-related research. This week Nine’s main Australian mastheads – The Age, The Sydney Morning Herald, Brisbane Times and WA Today – have published new guidelines for medical research reporting.

These include:

  • a focus on peer-reviewed research published in reputable journals;
  • requiring full copies of research, not just press releases or summaries;
  • making it clear if ongoing research or that presented at conferences hasn’t been peer reviewed;
  • examining and disclosing conflicts of interest;
  • treating animal studies with caution, and;
  • avoiding language such as “safe”, “guaranteed” or “miracle”.

I do wonder if the impetus for the guidelines may have been spurred by a retracted press release from the University of Sydney, which The Sydney Morning Herald reported on last month. A study linking elderberries to reduced influenza symptoms was funded by a company that sells elderberry-based flu remedies – not surprising, really, but definitely something that is crucial to understanding the study and any implications.

While the funding was disclosed in the paper itself, it was left out of the university’s press release – fortunately this kind of bad behaviour is rare, but it is why we always encourage journalists to check the paper for funding conflicts. That’s hard to do if the paper isn’t readily available for journalists.

Many high-profile journals work hard to make their materials available for verified journalists under a media embargo. Springer Nature, the UK Royal Society, JAMA and the New England Journal of Medicine all spring to mind as journal publishers that have good media sites and aren’t too difficult for journalists to gain accreditation. Others, such as The BMJ, PLOS and The Lancet, simply email press releases to their media lists with links to download the study.

Where it’s harder for journalists is when an institute sends out a press release unrelated to anything that the journal itself might be promoting – obviously each journal only profiles a handful of articles from each issue, so there are many more where institutes send out releases perhaps months after the paper was published, which normally means any media access via the journal has expired. So when an institute neglects to mention crucial funding information, or key limitations to the study, in a press release it might mean it goes unchallenged.

Plan S and Plan U

Which brings us to open-access publishing. Accessing journal articles wouldn’t be a problem if everything was published somewhere an everyday person, or in this case, a journalist, could download and read it, paying special attention to any issues in methodology and funding.

That’s what Plan S aims to address: the push from several European research funding agencies calls for publications that result from publicly-funded grants to be published open access. It’s been delayed a year, now they’re aiming for funders to start implementing the initiative in 2021, and it’s yet to be seen how many funding agencies will get on board. Definitely a case of watch this space.

Then, this week in PLOS Biology researchers suggested another strategy that they’ve dubbed Plan U, which would have funding agencies mandate that research papers were posted on pre-print servers (the U stands for universal).

Some journals are grappling with how to deal with pre-prints. PLOS and Cold Spring announced last year an agreement that enables articles submitted to PLOS journals to be posted automatically to bioRxiv. PLOS has also introduced a policy that any work published as a pre-print will not have a media embargo placed on it. This is one of the thorny questions that journals need to figure out, as it’s difficult to tell journalists they can’t cover a story that has been freely-available for months, if not longer, on a pre-print server.

Meanwhile, Springer Nature has updated its policy encouraging pre-print sharing, which includes guidance similar to that from PLOS that encourages researchers to talk to journalists ahead of publication but to emphasise “that the study has not been peer reviewed and that the findings could change”. Additionally, Springer Nature suggests that journalists who cover pre-print work indicate that the study has yet to be peer reviewed.

It’s promising to see this kind of convergence from the journals and journalists, but I admit I’m a little wary. Pre-prints work as a marvellous way to allow for pre-publication peer review, but there are potential pitfalls if media start covering research at this early stage (as is already happening). Peer review has been the gold standard for so long, and though not without its own pitfalls it at least gives a mark of quality to published research. And while pre-print will ensure that journalists and the public can access a version of the research after it’s been published, it certainly reinforces the important work of ensuring journalists have the tools to assess the validity and rigour of a study. That’s why it’s so pleasing to see the Nine stable of newspapers putting their own guidelines in place – I hope others will follow.

0 Responses to “A new pre-print server and media access to research”

  • This topic cuts across several issues I’ve been very interested in recently.

    First is the pros/cons of pre-prints. It’s good to see Nine self-regulating in the area of pre-print reporting. Some people suggest pre-prints are a bad idea because journalists may report findings inappropriately before the paper is reviewed, but I think the onus is clearly on journalists to spell out the caveats, instead of it somehow being the fault of pre-print servers or uploaders.

    Second is public access to research. Easier media access to research is one of the reasons I’m a fan of green open access, or ‘self-archiving’ a copy of the reviewed manuscript (before the publisher formats and typesets it) onto an institutional repository or personal website. (Note this is not the same as uploading the final version to sites like ResearchGate). The major downside is that power still rests with the publishers who grant green oa but who could alter it or take it away in future.

    But of course easier access to research by itself won’t help much if the researchers step outside their data and make overblown claims, such was the case with the ‘insectageddon’ or ‘insect apocalypse’ hype a while back. Its easy for the public and researchers to point the finger at sensational reporting and journalists eager to harvest clicks, but all too often it is the fault of overambitious press teams who exaggerate press releases, or the researchers themselves who go above the results of their study. In my view this was especially the case with the recent Australian review that spawned headlines proclaiming the extinction of 40% of insect species in the next 100 years. The fact that biodiversity is declining is no reason to throw out the baby with the bath water and fearmonger.

    Lots of complicated issues, particularly public access to research, but hopefully we get closer to solutions soon.

  • “…I think the onus is clearly on journalists to spell out the caveats…”

    The onus may be on them, but they will not do it! They seem to have great difficulty getting science stuff right, even under the current system.