Navigating across medicine's electronic landscape, stopping at places with Pub or Central in their names
BMJ 2001; 323 doi: https://doi.org/10.1136/bmj.323.7321.1120 (Published 10 November 2001) Cite this as: BMJ 2001;323:1120
Attempts to use the internet to free up access to the world's biomedical literature have resulted in several similarly named initiatives emerging over the past two years. PubMed Central, BioMed Central, and the Public Library of Science have joined the slightly older PubMed, which has a different function but a name similar enough to add to the confusion.
The debates around free access are some of the most important that have taken place in the three centuries of scientific publishing.1 Yet confusion about who wants to do what to whom is hampering this debate. This attempt to dispel some of the confusion was up to date at the time of writing, but given the speed at which medicine's electronic landscape is changing, it is likely to date fast.
Summary points
Several initiatives have recently emerged to provide free access to biomedical literature though the internet
Traditional publishers have been reluctant to join these initiatives because of fears about lost subscriptions
PubMed Central's “decentralised model” could be the trigger for greater publisher participation
The advent of free electronic journals—paid for by authors' charges—could profoundly change the publishing landscape
PubMed
PubMed provides access via the world wide web to over 11 million Medline citations dating from the mid-1960s to the present. It covers 4300 journals devoted to medicine, nursing, dentistry, veterinary medicine, healthcare systems, and the preclinical sciences. PubMed also provides access to life science journals that are not indexed by Medline but have submitted their full text to PubMed Central (see below).
As well as providing access to these abstracts and citations, PubMed links to more than 2000 websites that provide full text articles (figure). Access to these full text articles usually entails registration, subscription fees, or some other form of payment, although bmj.com is free.
In the four years since PubMed was launched as Medline's web interface, it has become the starting point for literature searches for most medical researchers. Over one million searches are conducted through PubMed each working day. A service provided by the US National Library of Medicine, PubMed is paid for by the US taxpayers.
Decisions for authors
In terms of exposure, publishing original biomedical research in a journal not indexed in PubMed is akin to sealing a manuscript in a bottle and launching it on the tide. For authors, it's publish in a journal indexed by PubMed or perish.
Decisions for publishers
Research journals that want to be taken seriously need to appear in PubMed. The main route in—acceptance by Medline—is difficult. Only 15-20% of applications to Medline are successful, and the process can begin only a year or two after publication of a new journal. The fact that publishers can leapfrog this hurdle by agreeing to submit the full text of articles to PubMed Central has fuelled publishers' grievances against PubMed Central. Should Medline level the playing field by being less exclusive? Arguments about constraints on space in the paper world do not apply online.
PubMed Central
As its name suggests, PubMed Central is an extension of PubMed and is similarly funded by the US taxpayer through the National Institutes of Health. Whereas PubMed provides free access to bibliographic citations and abstracts, PubMed Central provides free access to the full text of peer reviewed articles.
Suggested in 1999 by Harold Varmus, when head of the US National Institutes of Health, PubMed Central became operational within a year. In its first iteration, the full text of original research articles had to be made available from PubMed Central some time after publication. Fearing lost subscriptions, most publishers sat on their hands and hoped it would go away.
Given this negative reception, PubMed Central proposed an additional “decentralised model” earlier this year. Although publishers must still send the full text to PubMed Central—to allow for sophisticated searching—PubMed Central could route users back to publishers' sites to access the full text. Publishers have to guarantee free access to this material within a year of publication.2
Screenshot of abstract of a BMJ article on PubMed, showing its Link Out facility. In this example, the blue button links directly to the full text of the article on bmj.com
Few journals are currently available on PubMed Central. These are BMJ, Bulletin of the Medical Library Association, Molecular Biology of the Cell, Plant Physiology, Proceedings of the National Academy of Sciences of the United States of America, and 55 journals published by BioMed Central (see below). Another dozen have pledged to participate. None is yet available in the new decentralised format, but 17 journals have either signed up or indicated that they will participate (Ed Sequeira, personal communication).
Decision for authors
Most researchers want to maximise exposure to their work and the credit that it brings them. However, in a world where ever more material is read by ever fewer people, the standing of the journal in which a study appears is widely used as a proxy for the quality of the study itself. (Research assessment exercises have led the way here.)
Although PubMed Central clearly satisfies researchers' aspiration for exposure, the non-participation of most high status journals threatens the credit that could accrue to authors. Authors therefore have a difficult decision: by publishing in journals that have signed up with PubMed Central they might increase the chance that future researchers would have free access to all scientific literature—but at the cost of damaging their own careers.
Decisions for publishers
Publishers' initial response to PubMed Central was a staunch defence of the status quo. Researchers may conduct and write up the research as well as peer review and edit articles, but publishers expected to sell the results of their labours back to them and their librarians at a substantial profit.
The first iteration of PubMed Central apparently threatened subscription revenue: if readers could find the full text of original articles on PubMed Central then neither they nor their librarians would need to subscribe. Yet this first iteration placed no limit on the time between journal publication and its subsequent appearance on PubMed Central. The gap could have been as long as two years—hardly likely to affect subscriptions.
Even before PubMed Central was mooted, many journals were giving free access to their original research articles one or two years after publication.3 But PubMed Central was certainly a spur to such activity, with the New England Journal of Medicine admitting as much when it announced its decision to give free access to its original research articles six months after publication.4
From the user's point of view, the combination of free archives available from journal websites and PubMed's Link Out facility (figure) looks very much like PubMed Central's decentralised model. Yet publishers seem as reluctant to embrace this model as they were the original model. This is despite the advantage of having a journal archive maintained by PubMed Central, thus addressing librarians' main concern about electronic journals—that publishers can't be depended on to maintain electronic archives.
BioMed Central
BioMed Central is an independent publishing house that aims to provide immediate free access to peer reviewed biomedical literature. It publishes all articles deemed scientifically sound simultaneously on its own websites and through PubMed Central.
BioMed Central provides a quick route on to PubMed Central for authors disenchanted with traditional journals' slow peer review process and their practice of charging users for the finished product. It does this in two ways—through its own electronic journals and by giving groups of scientists the ability to set up electronic journals under their own editorial control. Most of BioMed Central's large stable of electronic journals began publishing within the past 18 months; all of them are indexed by PubMed, even those that are yet to publish an article. To pay for this service, BioMed Central is considering introducing authors' charges of $500 (£330) per accepted article from next year.5
Decisions for authors
Publishing in a BioMed Central journal gives interested readers free access to the full text article as soon as it is published—a potential increase in exposure compared with publishing in more restrictive journals. But like all new entrants, BioMed Central's journals have yet to establish themselves as prestigious destinations for authors, for whom impact factors loom large. Understandably, BioMed Central is “working to develop additional methods for judging the relative importance of particular pieces of research.”6
Decisions for publishers
Publishers need to decide what to do in response to the BioMed Central “experiment.” Should they ignore it, emulate it, or buy the company—if they believe it is several years ahead of the curve?
Public Library of Science
Sharing some of the same architects as PubMed Central, the Public Library of Science is a pressure group best known for its open letter (box), which has attracted more than 28 000 signatories from 172 countries. The last paragraph of its letter is crucial. It states that signatories will publish in, edit or review for, and personally subscribe to only those journals that have agreed to grant unrestricted free distributions rights to all original research within six months of publication. What many signatories and publishers may not have grasped is that “unrestricted free distribution rights” means exactly what it says: six months after publication anyone should be able to reuse any material, for any purpose—including commercial ones.
Public Library of Science open letter
We support the establishment of an online public library that would provide the full contents of the published record of research and scholarly discourse in medicine and the life sciences in a freely accessible, fully searchable, interlinked form. Establishment of this public library would vastly increase the accessibility and utility of the scientific literature, enhance scientific productivity, and catalyze integration of the disparate communities of knowledge and ideas in biomedical sciences.
We recognize that the publishers of our scientific journals have a legitimate right to a fair financial return for their role in scientific communication. We believe, however, that the permanent, archival record of scientific research and ideas should neither be owned nor controlled by publishers, but should belong to the public, and should be freely available through an international online public library.
To encourage the publishers of our journals to support this endeavor, we pledge that, beginning in September, 2001, we will publish in, edit or review for, and personally subscribe to, only those scholarly and scientific journals that have agreed to grant unrestricted free distribution rights to any and all original research reports that they have published, through PubMed Central and similar online public resources, within 6 months of their initial publication date.
By the library's September deadline, no scientific journal had committed itself to free distribution rights after six months. Even BioMed Central, whose journals head the Public Library of Science's list of those that have come closest to adopting its policy, retains exclusive rights to commercial reuse of material in its journals.6
Decisions for authors
As signatories would have had no journals to publish in, edit or review for, or personally subscribe to from September, the Public Library of Science recommended that signatories should “make every effort to publish our work in, and give our full support to, journals that have adopted the policy proposed in the open letter.” (Policy, here, apparently does not extend to “unrestricted free distribution rights.”) But resistance from most publishers “has made it clear that if we really want to change the publication of scientific research, we must do the publishing ourselves.” So the library will launch its own journals. Author charges of about $300 per published article are expected to cover the costs of peer review, editorial oversight, and publication. Unrestricted rights to access, distribution, and use of all articles will be assigned to the public domain.
Decisions for publishers
Although many journals are now freeing up access to their articles soon after publication, they are still guarding their commercial rights beyond that. Compliance with the demands of the Public Library of Science could mean losing revenue from reprints of articles more than six months old, for which pharmaceutical companies pay medical journals substantial amounts of money. Should they forgo this income to comply with these demands? Or will the arrival of the Public Library of Science's own journals mean action is already too late?
Conclusion
The current practice of scientific publishing results from the balance of power among authors, funders of research, publishers, academia, and the alliances between them. As the great disrupter, the web seems likely to confound the status quo, although exactly how is hard to predict. The wheel is still in spin.
Key players
PubMed
www.ncbi.nlm.nih.gov/entrez/query.fcgi
PubMed Central
BioMed Central
Public Library of Science