Options to Access the Peer-Reviewed Literature

At present, there are 3 major ways to access the medical literature. Subscription-based journals still predominate, published by a wide range of commercial and not-for-profit publishers, including many medical societies that derive needed income from these endeavors to support their mission. The most prestigious journals covering large disciplines are published by this mechanism, but there are also some journals with very restricted readership. Individuals purchase a print subscription and most libraries have digital subscriptions; some libraries continue with print subscriptions as well, to a select (and shrinking) number of publications.

A second source to access peer review literature are the open access journals that turn the subscription-based model around and, instead of relying on subscription revenues from libraries or individuals, charge a fee to authors or funders before an article is published. For some open access journals these fees provide the sole mechanism to support the journal’s production costs. Elsevier has about 40 journals that are total open access journals. There are several variations of open access journals or articles that will be addressed below.

A third method of access is through repositories organized by many universities or in other established locations online, the most prominent being the National Institutes of Health (NIH)-sponsored PubMed Central. Repositories are not publishers but rather provide access to some version of the manuscript, either before or at some point after they have been published, usually subject to an embargo period. Universities in many countries have established their own repositories, but the rates at which published papers have been deposited in them so far have been very disappointing. Also, since these are only preprints or postprints the final published article in the journal might be slightly different from the article deposited in the repository, since the repository does not usually contain the journal-added formatted and copyedited version.

At present approximately 14% of the publications in medicine are available online free of charge from the onset of publication, and another 8% are available on a delayed embargo path to open access. PubMed Central attracts 420 000 visitors per day, of which only a quarter originate from a university log-on; this confirms the high demand for scientific information outside academia.

Background of the Open Access Initiative

Readers of a certain age will remember the days of print-only publications and before PubMed, when clinicians and researchers had to explore the books of Index Medicus to hopefully find articles on a specific topic, usually finding only a title, or pay a fee to a reference librarian to perform this search. Most articles were obtained by copying machines, with the tight binding of some journals precluding copying. Further possible references were noted within the articles and then the library process was repeated. As a result, many articles of interest were simply never located. Fast-forward to today’s Internet search tools and the delivery of electronic tables of contents. Today you can search PubMed and PubMed Central from home, and most academics use their institutional software licensed from publishers to perform searches, frequently from home. Most articles can be downloaded as a PDF file or, using the HTML version online, there are ready links to the full text of the other articles cited using the CrossRef, a collaborative reference linking service developed by publishers. The main task for readers or researchers today is to focus and manage the search terms, since there is usually an overabundance of articles.

This giant leap in the access to peer review literature took place for 2 essential reasons, the first the result of the traditional publishers’ long history of the production, dissemination, and archiving of the huge volume of peer-reviewed scholarly articles and the second being the arrival of the Internet, permitting easy dissemination of this information. After decades of a successful publication process, the traditional subscription model of publishing research is now at a crucial and evolving stage. In this traditional system, the universities, research organizations, or clinical practices conduct the basic or clinical research; foundations or funding agencies provide funds to conduct research or authors simply provide their own resources; authors write research articles with no compensation for publication; publishers accept manuscripts on condition of copyright transfer and then facilitate the editorial process of peer review and manage the production and distribution processes needed for disseminating the articles; and libraries use institutional funds to purchase, organize, and preserve this scientific record and make it available for current and future researchers, or the individual clinicians purchase subscriptions to the journals. Even researchers need to pay or have a personal or institutional subscription to obtain access to their own work. This system is mature but complex, interdependent, and perhaps now recognized by some as inequitable or unacceptable.

The new technology of the Internet has the potential to bestow worldwide electronic distribution of the peer-reviewed journal literature to all scientists, scholars, teachers, students, and public. Hence the movement for open access by users of the peer review literature, by librarians, and by the public was born and has now developed, although not gracefully. Traditional journals have only cautiously accepted digital technology to further enhance free access to journal content. Even though each of these processes can now be accomplished in a digital format alone, there are still significant costs to libraries and subscribers. Other features have emerged from the publishers, such as bundled journal digital subscriptions and enhanced content with new services derived from technology (eg, reference linking). Libraries initially had to have both print and online versions of journals, and pay for both. The bundled subscriptions offered were frequently inadequate, since many of the desired journals required additional fees. Although paper and postage costs were eliminated with digital subscriptions, they were replaced by costs to the publisher associated with online submission-and-review systems and hosting platforms. So, the digital availability did not appreciably reduce the cost of publication initially, and these cost issues have strained the previously cordial relationship between publishers and academic librarians. Libraries’ acquisition budgets for online access have not increased in parallel with the doubling of the budget of the National Institutes of Health (NIH) between 1997 and 2003 and of medical research worldwide. In 2009, the STM (science-technical-medical) publishing market published nearly 1.5 million articles from over 25 000 unique titles, from over 2000 different publishers. One solution is that libraries simply accept the reality of the explosion of medical information and either accommodate their budgets (unrealistic) or freely acknowledge their inability to provide all the literature to their constituents. Alternatively, coupling growth in research output to publication funding is a consideration, so that budgets and research outputs stay in tandem. But this only assists research-oriented institutions and alternatives are limited, especially in the United States.

A more contentious and volatile aspect of the open access issue is that, as Nosek and Bar-Anan describe, many feel that the commercial publishers have exploited the researchers, reviewers, and editors who provide the content and perform peer review mostly on a voluntary basis, surrender copyright of the work to the publisher (who fails to publish the work quickly), and then turn around and charge the authors’ universities (inflated) fees to access the content either in print or online. So the open access movement was driven in part by annoyance and frustrations at the real or perceived (excess) profit of commercial publishers who seem to be exploiting the academic community and the government or commercially funded research. In 2011, the biggest publisher in the STM field, Elsevier, made £768 million ($1.2 billion) on revenues of £2.1 billion, representing a profit of 37%. Other publishers likewise have large profit margins.

Open access advocates further opine that publishers, through their editorial boards, have limited, unnecessarily filtered, and excessively classified the work produced by researchers. They argue that publisher activity makes the peer review literature more orderly and placed in context but that this disturbs or restrains the scientific spirit of exploration and knowledge. Some ardent proponents of the open access movement pose the doctrine that all scientific research, experiences, or opinions should be made available open access and that time should be the ultimate judge of the value of the content (if chaos doesn’t intervene). Since the objective of science is to build a shared body of scientific knowledge through trial methods and the medium of the Internet can permit easy transmission of that knowledge, the Internet should be fully utilized so that others can evaluate, challenge, enhance, and provide supplementary innovation.

This open access movement is especially vocal with regard to research that was funded by taxpayers. For some researchers, scientific research, especially with funding support, is inherently accompanied by the moral imperative to circulate the results widely and, in the digital age, this requires exploring the new publishing technologies and the economic models that permit access of research for all. By extension, in the opinion of the open access idealists, “open access” means not only free availability of the article on the public Internet (called gratis open access) but also the ability of the readers to reuse the data in their own software for additional research without any barriers except for proper acknowledgment of the original authors (called libre open access when these additional rights are available). Obviously, not all endorse this full definition of open access, but rather some believe open access to published articles (gratis open access) and access to the background or raw data (libre open access) are separate topics that should be decided by individual journals or publishers or by researchers. Admittedly, many of these liberal concepts of the open access proponents run counter to some of the principles of entrepreneurship and patent rights.

Background of the Open Access Initiative

Readers of a certain age will remember the days of print-only publications and before PubMed, when clinicians and researchers had to explore the books of Index Medicus to hopefully find articles on a specific topic, usually finding only a title, or pay a fee to a reference librarian to perform this search. Most articles were obtained by copying machines, with the tight binding of some journals precluding copying. Further possible references were noted within the articles and then the library process was repeated. As a result, many articles of interest were simply never located. Fast-forward to today’s Internet search tools and the delivery of electronic tables of contents. Today you can search PubMed and PubMed Central from home, and most academics use their institutional software licensed from publishers to perform searches, frequently from home. Most articles can be downloaded as a PDF file or, using the HTML version online, there are ready links to the full text of the other articles cited using the CrossRef, a collaborative reference linking service developed by publishers. The main task for readers or researchers today is to focus and manage the search terms, since there is usually an overabundance of articles.

This giant leap in the access to peer review literature took place for 2 essential reasons, the first the result of the traditional publishers’ long history of the production, dissemination, and archiving of the huge volume of peer-reviewed scholarly articles and the second being the arrival of the Internet, permitting easy dissemination of this information. After decades of a successful publication process, the traditional subscription model of publishing research is now at a crucial and evolving stage. In this traditional system, the universities, research organizations, or clinical practices conduct the basic or clinical research; foundations or funding agencies provide funds to conduct research or authors simply provide their own resources; authors write research articles with no compensation for publication; publishers accept manuscripts on condition of copyright transfer and then facilitate the editorial process of peer review and manage the production and distribution processes needed for disseminating the articles; and libraries use institutional funds to purchase, organize, and preserve this scientific record and make it available for current and future researchers, or the individual clinicians purchase subscriptions to the journals. Even researchers need to pay or have a personal or institutional subscription to obtain access to their own work. This system is mature but complex, interdependent, and perhaps now recognized by some as inequitable or unacceptable.

The new technology of the Internet has the potential to bestow worldwide electronic distribution of the peer-reviewed journal literature to all scientists, scholars, teachers, students, and public. Hence the movement for open access by users of the peer review literature, by librarians, and by the public was born and has now developed, although not gracefully. Traditional journals have only cautiously accepted digital technology to further enhance free access to journal content. Even though each of these processes can now be accomplished in a digital format alone, there are still significant costs to libraries and subscribers. Other features have emerged from the publishers, such as bundled journal digital subscriptions and enhanced content with new services derived from technology (eg, reference linking). Libraries initially had to have both print and online versions of journals, and pay for both. The bundled subscriptions offered were frequently inadequate, since many of the desired journals required additional fees. Although paper and postage costs were eliminated with digital subscriptions, they were replaced by costs to the publisher associated with online submission-and-review systems and hosting platforms. So, the digital availability did not appreciably reduce the cost of publication initially, and these cost issues have strained the previously cordial relationship between publishers and academic librarians. Libraries’ acquisition budgets for online access have not increased in parallel with the doubling of the budget of the National Institutes of Health (NIH) between 1997 and 2003 and of medical research worldwide. In 2009, the STM (science-technical-medical) publishing market published nearly 1.5 million articles from over 25 000 unique titles, from over 2000 different publishers. One solution is that libraries simply accept the reality of the explosion of medical information and either accommodate their budgets (unrealistic) or freely acknowledge their inability to provide all the literature to their constituents. Alternatively, coupling growth in research output to publication funding is a consideration, so that budgets and research outputs stay in tandem. But this only assists research-oriented institutions and alternatives are limited, especially in the United States.

A more contentious and volatile aspect of the open access issue is that, as Nosek and Bar-Anan describe, many feel that the commercial publishers have exploited the researchers, reviewers, and editors who provide the content and perform peer review mostly on a voluntary basis, surrender copyright of the work to the publisher (who fails to publish the work quickly), and then turn around and charge the authors’ universities (inflated) fees to access the content either in print or online. So the open access movement was driven in part by annoyance and frustrations at the real or perceived (excess) profit of commercial publishers who seem to be exploiting the academic community and the government or commercially funded research. In 2011, the biggest publisher in the STM field, Elsevier, made £768 million ($1.2 billion) on revenues of £2.1 billion, representing a profit of 37%. Other publishers likewise have large profit margins.

Open access advocates further opine that publishers, through their editorial boards, have limited, unnecessarily filtered, and excessively classified the work produced by researchers. They argue that publisher activity makes the peer review literature more orderly and placed in context but that this disturbs or restrains the scientific spirit of exploration and knowledge. Some ardent proponents of the open access movement pose the doctrine that all scientific research, experiences, or opinions should be made available open access and that time should be the ultimate judge of the value of the content (if chaos doesn’t intervene). Since the objective of science is to build a shared body of scientific knowledge through trial methods and the medium of the Internet can permit easy transmission of that knowledge, the Internet should be fully utilized so that others can evaluate, challenge, enhance, and provide supplementary innovation.

This open access movement is especially vocal with regard to research that was funded by taxpayers. For some researchers, scientific research, especially with funding support, is inherently accompanied by the moral imperative to circulate the results widely and, in the digital age, this requires exploring the new publishing technologies and the economic models that permit access of research for all. By extension, in the opinion of the open access idealists, “open access” means not only free availability of the article on the public Internet (called gratis open access) but also the ability of the readers to reuse the data in their own software for additional research without any barriers except for proper acknowledgment of the original authors (called libre open access when these additional rights are available). Obviously, not all endorse this full definition of open access, but rather some believe open access to published articles (gratis open access) and access to the background or raw data (libre open access) are separate topics that should be decided by individual journals or publishers or by researchers. Admittedly, many of these liberal concepts of the open access proponents run counter to some of the principles of entrepreneurship and patent rights.

Open Access History and Definitions

The Declaration of the Budapest Open Access Initiative (2002), the Bethesda Statement on Open Access Publishing (2003), and the Berlin Declaration on Open Access to Knowledge in the Sciences and Humanities (2003) were the documents that spurred the open access movement with the theme that peer-reviewed research articles, submitted by researchers with no compensation, should be made available online, free, and with the smallest possible number of usage restrictions. As described by Laakso and associates, the movement has been through 3 stages now, including the pioneering phase (1993-1999), when open access was largely confined to online repositories created and maintained by individual scientists or institutions; the innovation phase (2000-2004), when new business models for running open access operations were started; and now the consolidation phase (since 2005), when there has been a dramatic percentage growth in the number of open access journals and articles along with a developing infrastructure.

In reality, open access to the peer-reviewed literature has already been accomplished to some degree, and the forums have now progressed to debate the experiments and the structure that open access needs in order to maintain a sustainable financial framework while preserving the quality of research publications. Publishing costs money, so there have to be reliable revenue pathways to maintain these costs, especially in medical science. Against the background of this open access movement is the lingering concern that the invaluable scientific publishing system may not survive the crisis and/or that quantity will prevail over scientific quality—and that we might lose some of the literature that presently exists as the shake-out phase accelerates.

There is now the Directory of Open Access Journals (DOAJ), the primary indexing service for open access journals that also provides long-term archiving possibilities for these open access journals via an agreement with the Dutch National Library. Presently it lists 8733 open access journals from 122 countries and almost one million articles available open access. Many open access journals are highly regarded according to conventional measures of distinction (eg, impact factor). Emerging business models for open access journals or articles include publication fees paid by authors once an article has been accepted for publication (article processing charge or APC), direct support from research grants, and contributions from research institutions willing to contribute financially to publication systems for more openly accessible articles. Most of the online-only journals are open access journals with all their contained articles freely available. A print journal can also have an online open access component but is more likely to have a subscription pay model, although some open access articles can be made available, again with the article processing charge paid by the author or another sponsor, such as a funding body. This is the hybrid publishing model that this Journal is initiating and that is presented in the accompanying editorial.

The author charge fee was pioneered by the open access publisher BioMed Central (BMC), which has a stable of open access journals in most disciplines, including BMC Ophthalmology; it was funded by venture capital and was later purchased by Springer Publishers in 2008, now with article processing fees paid for by government funding in the United Kingdom (UK). The Public Library of Science (PLoS) was able to establish a cadre of high-quality open access journals (as measured by high impact factor) using a substantial start-up grant; author or sponsors pay an article processing fee. Open access publishing has gained traction over the past 10 years because of the success of the PLoS and BMC families of journals. The annual volume of articles published in open access journals has increased from 20 702 in 2000 to 340 130 in 2011—accounting for 17% of all articles published in 2011. Simultaneously, or in response, the general digitization of established print journals was rapid, and several society journals started to use the services of HighWire Press as well as other publishers. In 2004 mainstream publishers started experimenting with the hybrid open access model for their subscription journals, as announced for the Journal in this issue.

As defined in the accompanying editorial, open access can refer to a journal or to an article. “Gold open access” refers to an article that is available to the public immediately on publication because an article processing charge has been paid; BioMed Central and the series of PLoS are examples of gold open access journals. The importance of the article processing fee business model for funding open access publishing has grown rapidly. The Wellcome Trust in the UK is an example of a fund that sponsors the payment of article processing fees. The majority of Elsevier’s 1200 proprietary titles are now prepared for the hybrid open access option, permitting gold open access articles within an otherwise subscription journal. Most funding agencies are encouraging or requiring their grantees to publish in gold open access journals or hybrid journals, allowing the authors to pay the fees with money from their research grants or with funds allocated by the agency.

“Green open access” means the article is publicly available by self-archiving of the author’s work in a personal or institutional website or by the publisher’s depositing it in a central repository such as PubMed Central, either voluntarily or under various mandatory agreements from funding bodies; it might be the actual published manuscript or a preprint (unedited) version of an accepted manuscript and may be subject to an embargo period. PubMed Central is a repository of peer-reviewed and published manuscripts designed to provide the public with access to the research supported by their tax dollars or to articles made open access by the publisher. PubMed Central, however, diverts $4 million from the NIH budget each year to assist this archival process, and many universities who have their own institutional repositories spend an average of $150 000 a year from their library budgets to support their own archives. Some might argue that this might not be the best use of NIH or institutions’ monies.