cientific publishing giant Elsevier recently came under fire after its director of scholarly communications, William Gunn, wrote a controversial tweet in response to a post that “people shouldn’t have to jump through hoops” to access important medical research.
Yes, everyone should have rainbows, unicorns, & puppies delivered to their doorstep by volunteers. Y'all keep wishing for that, I'll keep working on producing the best knowledge and distributing it as best we can.
— mrgunn (@mrgunn) August 13, 2018
While I disagree entirely with the harsh tone of Gunn’s tweet, the underlying rationale behind his comment should not be overlooked: to manage submissions, peer review, editing, and publication, quality scientific publishing requires infrastructure that costs money.
In this age of fast and free access to published information, many people use the internet to inform themselves about new and cutting-edge treatments for medical conditions. This is especially true with cancer and pediatric conditions, especially those that are life-changing or life-threatening.
As a physician-scientist myself, I’ve dedicated my career to improving health through research and to getting essential treatments to more people more quickly. I empathize with those who are suffering from conditions we have not yet found therapies or cures for, and I understand the frustrations of those who want fair and increased access to pertinent medical research.
Only a handful of models exist today for scientific publishing. One is the traditional subscription price model, which is the one often critiqued by those looking for increased access to research papers. This model places articles behind paywalls — often ones that come require a pricey journal subscription or a steep single-article fee. This isn’t unique to scientific publishing; newspapers, magazines, and online publications use a similar model to cover costs and generate revenue.
The issue that arises with scientific papers is that a subscription restricts access by the taxpayers who funded the work in the first place. In the U.S., this issue has been partially addressed: All publications funded at least in part by the National Institutes of Health must be submitted to the National Library of Medicine’s PubMed Central website, where they are freely available for download in a “plain” format. A number of scholarly publishers automatically send manuscripts to PubMed Central, but many do not, leaving a large gap, especially for work that was not funded by the U.S. government.
At the other end of the spectrum are preprint publications, which are essentially self-published papers. Researchers are able to post their work months or years ahead of formal publication in a journal, a practice that has been used by the engineering and mathematics communities for years. Today, clinical investigators are able to publish preprints on sites such as bioRxiv and PeerJ. These sites have low operating costs, and allow readers to access information free of charge.
The drawback is that preprints are not peer reviewed and, in this age of “truthiness,” accuracy is of the utmost importance — even more so when it comes to clinical research that can affect human lives. People who don’t work in the scientific and medical fields may not have a yardstick to judge the quality of a particular piece of research, or know if the conclusions follow from the evidence presented. So while preprints improve accessibility, there’s always the risk that they may contain serious flaws with far-reaching impacts.
This is not to say that rigorous peer review is perfect. Clinical trials are required to endure rigorous reviews both before and during the trial, as well as peer review before publication. But this has not prevented pharma and biotech companies from authoring research that was later found to be skewed. In one example, a health care company sponsored clinical studies over the course of a decade on the immunosuppressant tacrolimus, the results of which were published in high-quality, peer-reviewed journals. These studies found a lower incidence of drug-induced diabetes using a non-standard definition, a finding now recognized to be questionable.
Sitting somewhere in the middle of these models is open-access publishing. It combines the benefit of professional publishing and peer review with increased access to scientific research. Open-access journals are not print based, which reduces costs. They offset the loss of paid subscriptions with higher publication charges to researchers. The European Union, driven in part by its mandate to make all scientific articles freely available by 2020, is largely moving toward an open-access model. Research funders from France, the United Kingdom, the Netherlands, and eight other European nations have just unveiled a radical open-access initiative called Plan S. The 11 agencies propose mandating that the scientists they fund make resulting papers free to read immediately upon publication, with a liberal publishing license allowing anyone to download, translate, or reuse the work. Time will tell how European publishing companies and investigators respond to the proposed changes, but this is a major shift for European science.
So the ultimate question is this: How will the scientific community and the country as a whole pay for and support an infrastructure that ensures scientific rigor and accuracy while preserving accessibility? Even without a profit margin or print-based journals, the fact remains that it costs money to publish scientific research.
In my view, it seems inevitable that U.S. scientific publishing will follow Europe’s lead and continue to move towards an open-access model. To survive financially, however, publishing companies will need to adapt and innovate, finding new ways to reduce costs and generate revenue rather than relying on traditional subscription fees. This process has already begun, with publishers increasingly hosting online document preparation platforms to help researchers format their papers correctly, which reduces the amount of formatting on the publisher’s end. I also envision publishers expanding add-on services such as copy editing, graphic design, and social media advertising, all for an additional fee.
Publishers will also have to reevaluate their method for recruiting peer reviewers, which is a major effort. With decreased revenues, publishers may begin to require researchers to review a paper every time they submit one.
We may also see the use of limited embargos, where a paper is available for download immediately — but exclusively — on the publishing journal’s website for some period of time. This would increase access while also driving traffic to the publisher’s site.
Publishing companies aren’t the only ones that will need to change. Researchers, funding agencies, and the public at large will need to adapt as well. In the open-access model, investigators will be responsible for covering the costs associated with producing quality work. As publishing fees increase — they are presently upwards of $3,000 to $4,000 per manuscript — research funding agencies may need to incorporate additional funds for publishing into their grants.
That, of course, means taxpayers will need to support increased government spending in this arena. While most Americans say they favor government funding for medical and scientific research, there’s clearly still a notion that these advancements should come at minimal cost to the public.
In the end, it would serve us all to preserve the valued tradition of publishing quality scientific research while increasing access for those who need it most. We need to recognize and support the value of open access to science and research, and use our skills at innovating to keep costs down.
Martin Zand, M.D., is the co-director of the Center for Leading Innovation and Collaboration at the University of Rochester Medical Center; professor of medicine and public health sciences at the University of Rochester; co-principal investigator of the university’s Clinical and Translational Science Institute; and director of the Rochester Center for Health Informatics.