I continue to be amazed that despite the appearance of the internet, which some have compared with the invention of fire, our methods for disseminating scientific studies are essentially the same as they were 50 years ago. We still have journals, and, although papers have electronic versions, those papers are indistinguishable from those of 50 years ago. I’m constantly searching for the “disruptive innovation” that will change everything.
By definition those stuck in the old paradigm (that’s all of us) can’t imagine the new paradigm. We will be slow to spot the significance of a development that could change everything, but I wonder if the appearance of PLoS [Public Library of Science] ONE could be a gamechanger. It appeared a couple of years ago, and a signal that it might be highly significant is that it has already been copied many times, with BMJ Open being the latest copy.
The idea behind PLoS ONE and copies is that they will publish any scientific study where the conclusions are supported by the methods and results. The editors and reviewers do not attempt to decide what is “original” and “important,” which they do badly anyway. So long as authors stick to what they did and what they found and avoid excessive conclusions, the curse of medical papers, then almost everything can be published. For example, most traditional journals would not publish a survey with only a 10% response rate, but so long as the authors say that it was 10%, give their results, and don’t attempt to draw conclusions on the whole population then the paper can be published.
PLoS ONE is now receiving almost 2000 studies a month, and the trend is steadily upwards. If the copies of PLoS ONE start to receive as many studies, then the supply of studies to traditional journals may begin to dry up. Publishers will then be caught in a dilemma. They have stuck with traditional journals because they are so profitable, but they are attracted to copies of PLoS ONE because they can be very lucrative.
One reason that authors stick with traditional journals is because academic credit comes with publishing in them. But another innovation, “article level metrics,” may change that. These metrics give in close to real time details of accesses to the article, downloads, citations in several databases, mentions in blogs, and other data. Detailed data are then available on each article, and it is much more scientific to use these as a measure of academic achievement than to use the “impact factor” of the journal in which people publish—because the impact of articles does not correlate to the impact of the journal because the impact factor of the journal is driven by a few highly cited articles.
Authors might also chose to publish in traditional journals because they can reach a large audience, but many traditional journals do not have a large audience. Plus we know that most readers of medical journals do not read the scientific articles. In fact it’s crazy to send an almost random selection of scientific studies to an audience that doesn’t read them. And we know that publication is a small step in dissemination. Much better, I suggest to authors, to have your study published quickly in something like PLoS ONE or BMJ Open—and then put a lot of energy into dissemination, Tweeting like mad, circulating the URL on listserves, and most importantly directing your study towards those for whom it is most relevant.
Long ago Ian Roberts and I imagined a world in which studies would not be published in journals but rather in databases. The job of journals would not be to spend resources peer reviewing and circulating studies to people who don’t read them but rather to pick out the few studies that matter and present critical appraisals of them to the right audience. Perhaps the proliferation of copies of PLoS ONE will bring that vision closer.
Competing interest: RS is a member of the board of the Public Library of Science. He is not paid. He was the editor of the BMJ until 2004 and is director of the United Health Group’s chronic disease initiative.