Covid prompted new ways to publish research – it’s time to embrace them

Breaking from the status quo is never simple, and scholarly communication has proven to be particularly resistant to change over the years.

But the Covid pandemic changed many things about the world, one of which was the way biomedical research was shared. For one thing, the urgent need to develop a vaccine forced publishers to acknowledge that keeping relevant discoveries behind paywalls was unethical, and most of them opened up their corpuses for everyone to read.

In addition, publishers recognised that their own review processes must not be allowed to slow the dissemination of new discoveries, and they made clear that they would not discriminate, in their editorial decision-making, against articles that had previously been published on preprint servers. That led submissions to bioRxiv and medRxiv to increase at what one study called a “staggering rate” in 2020, swelling 10-fold at medRxiv alone in the first few months.

Yet this firehose of unreviewed papers left researchers with no way of knowing what was worth reading. Under the traditional scientific publishing system, there is no way to tell whether preprints that remain unpublished in journals have been reviewed and rejected, are still under review, or simply weren’t submitted.

Fortunately, Covid also saw a surge in the number of groups publicly evaluating preprints, under what became known as the Publish-Review-Curate (PRC) model of preprint publishing. What was radical about this was that all the reviewers’ comments – both positive and negative – were made public. What’s more, the reviewing groups were able to publicly evaluate the papers on numerous axes, such as the strength of the methods, the significance of the findings and the clarity of the reporting. Journals could provide none of this nuance since readers only saw the result of a binary decision: accept or reject. They had no way of knowing how the paper rated on each of those different axes.

eLife launched its own version of PRC in January 2023. It too focuses on publishing all the reviews, positive or negative, of the preprints it chooses to send for review, and all papers sent out for review are now published in the journal. Moreover, decisions to send out a paper for review are based not only on perceived quality but also on public interest, such as if the preprint is getting a lot of attention and public peer review would add useful context.

Since there is no accept/reject decision after peer review, authors can focus on improving their article to address any concerns raised by the reviewers without the fear of having the revised version rejected. You might think that they would lack the motivation to do so without the incentive of publication, but 96 per cent of papers are revised before the version of record is published.

eLife has also formalised the process of evaluation by using a common vocabulary to summarise the significance of the paper’s findings (on a scale ranging from landmark to useful) and the strength of the evidence (from exceptional to inadequate). This approach emphasises that a paper should be judged on its scientific content, rather than the name or impact factor of the journal it is published in. And it explains why eLife switched wholesale to this new model instead of launching a new journal to test it; meaningful reform of the scientific evaluation system would not have been possible if we had continued to run a prestigious journal.

Not everyone was convinced of our wisdom, of course. The notion that there were other ways of sharing research genuinely seemed to be a shock to many in the scientific community. The initial backlash was perhaps best captured in this very publication by virologist Paul Bienasz, who argued that the concept of “getting into” a highly selective journal is deeply baked into academic life.

Relatedly, last November, Clarivate made the decision not to assign eLife an impact factor on the grounds that our article-level assessment is at odds with such journal-level metrics because publication is “decoupled from validation by peer review”. But while Clarivate’s move has led to a fall in submissions from some parts of the world – most notably China, where we know the impact factor still holds considerable importance – many in the scientific community have recognised our new model as a solution to the slowness, waste, cost and inefficiency of traditional publishing. As a result, while publications are down about 70 per cent from China, they have only dropped by around 33 per cent from the US and 25 per cent from Europe, belying the widespread belief that a loss of impact factor is synonymous with a collapse in journal output. 

What this shows is that there is a willingness among researchers to embrace new models of publishing. Impact factors are not the only thing that matters to them, and a journal can operate without one as long as there are visible quality markers to replace its brand for assessment purposes: in our case, our robust public review process and eLife Assessments that summarise significance and rigour.

However, there is still a caution if, nevertheless, researchers perceive a risk to their success in evaluation processes. In our experience, this risk is more perceived than real, but while no institution has told us that they would consider eLife papers differently since the Clarivate decision, many researchers still believe they will.

That perhaps explains why, despite the persistence of many of the positive changes that were introduced during the pandemic years, we still haven’t seen the major leap that we at eLife believe the whole industry needs to take towards an open and transparent model of peer review and publication. That requires wider adoption of preprints by researchers, more vocal institutional support for alternative models of peer review and evaluation, and greater willingness among publishers to integrate them into their own processes.

We have shown that this is possible. But the required step change will not occur until all stakeholders recognise, as they must, that continuing to cling to the status quo is an affront to science.

Damian Pattinson is executive director at eLife.