Open Access to Research for the Developing World

Kofi Annan, then secretary-general of the United Nations, noted in 2002 that “[A] wide consensus has emerged on the potential of information and communications technologies (ICT) to promote economic growth, combat poverty, and facilitate the integration of developing countries into the global economy. Seizing the opportunities of the digital revolution is one of the most pressing challenges we face.”

The intervening five years have seen a rapid expansion in the reach of digital technology to encompass much of the developing world. Top-down efforts such as the One Laptop per Child initiative, now commencing production of its sub-$200 laptop, represent one approach. Arguably more significant, however, is the change that is being driven from within developing countries. In the area of mobile telephony, for example, Africa has generally been neglected as a marketplace by the major international telecom companies, but this has not prevented domestic mobile phone companies from adding subscribers at a spectacular rate. The number of mobile phones in Africa has doubled in the past two years, and there are now more than 200 million mobile phone users on the continent: 10 times the number of landlines. Although cellular modems are not the ideal way to connect to the Internet, this is nevertheless an enormous leap in access.

As a result of these trends, developing countries are now more connected than ever before, and the digital infrastructure that now exists has the potential to transform access to knowledge. The primary obstacles are no longer technological but are related to issues of content licensing, distribution, and access control.

Access to knowledge is clearly a fundamental requirement for development. It is difficult to see how the following United Nations Millennium Development Goals can be effectively achieved without ensuring that developing countries have access to the latest relevant scientific and medical knowledge:

• Reduce child mortality
• Improve maternal health
• Combat HIV/AIDS, malaria, and other diseases
• Ensure environmental sustainability
• Develop a global partnership for development

In particular, a global partnership aimed at addressing development issues would be hugely facilitated by a “knowledge commons” to ensure that developing and developed country researchers are not operating in isolation from one another.

Solutions to problems in the developing world depend on full and effective collaboration between those working in the developed and developing worlds. Leaving low-income countries to fend entirely for themselves in the face of problems that can be addressed using current scientific knowledge is not an ethically or morally acceptable choice. But nor is “parachuting in” solutions that have been developed entirely in the developed world without reference to local knowledge. The history of development cooperation is full of examples of attempted solutions imposed by the North on the South that lack the participatory elements crucial for socioeconomic acceptance and uptake.

In combination with appropriate local skills and expertise, online access to the latest research can help low-income countries not only deal with practical priorities in areas such as public health and agriculture but also provide a vital starting point to developing their own research capacity. M. S. Swaminathan, a key participant in India’s Green Revolution and now an active proponent of the role of access to knowledge in development, warns that “Many developing countries remain poor largely because they let the Industrial Revolution pass them by. They can ill afford to miss the information technology revolution.”

The importance of ensuring that developing countries have access to the latest medical research was recognized by the World Health Organization in 2000, and this led to the HINARI initiative, a partnership with research publishers that provides free or low-cost online access to medical journals for researchers working in the poorest countries. The HINARI model has spurred similar initiatives in agricultural research (AGORA) and environmental research (OARE).

Undoubtedly, initiatives such as HINARI have significantly improved access to research in developing countries. However, they offer only a partial solution. They represent a discretionary concession by publishers, allowing certain countries limited access to some content for a period of time. However, the publisher typically still retains exclusive rights over that content and determines how it may be used. For example, it is typically not allowable (without special permission) for research distributed under such schemes to be reprinted (for example, to allow distribution to sites without Internet access) nor for derived works (such as educational material) to be created and distributed. Initiatives such as HINARI also fail to address the access problems in countries with large economies such as Brazil, India, China, and South Africa. Although these countries have low per-capita incomes, they are nevertheless generally excluded from initiatives such as HINARI.

In addition, actual access has not met promises. A recent study published in BMC Health Services Research noted that users in African countries reported problems in accessing journal content through HINARI due to the technical requirements for login and authentication. Furthermore, because HINARI focuses on providing access at the institutional level, it does not fully address the access needs of practitioners, journalists, policymakers, and others who may not be affiliated with major institutes.

Research librarians in all countries are familiar with the increasing tendency of users to rely on Google and other Internet search engines for discovering information. Any system that aims to improve access to knowledge for developing countries must take this into account. Thus, accessibility should not depend on articles being accessed via a special portal or proxy server, or via complex authorization schemes. The simplest and most reliable way to ensure that knowledge is available where and when it is needed is to avoid access barriers altogether through a universal open-access model.

In the past decade, there has been a revolution in how we think about licensing digital information. Open licenses, which explicitly allow redistribution, reuse, and the creation of derivative works, have proved to be an extremely effective way to maximize the value to the community of digital resources.

The Linux open-source computer operating system and MIT’s Open CourseWare initiative are notable examples, as is Wikipedia, the free and open encyclopedia. In just a few years, Wikipedia has grown to become an invaluable knowledge resource, including more than 2 million articles in English and 5 million more articles in 200 other languages. Despite being a small nonprofit organization, Wikipedia is now one of the world’s top 10 most-accessed Web sites, demonstrating what can be achieved by bottom-up, openly licensed efforts. Wikipedia’s model of open contribution and distributed editorial control means that it is best seen as a complement to, rather than a replacement for, authoritative information sources. However, the massive use of Wikipedia, and the richness of its content, demonstrate a global thirst for knowledge that has not been addressed by traditional alternatives and provide encouragement for those seeking to create open models for the publication of peer-reviewed, editorially filtered content.

To facilitate the exchange of openly licensed resources, the organization Creative Commons has created a set of standard licenses, designed to be human- and computer-readable, that are now used by publishers and content creators to flag open content that is available for reuse. An increasing number of open-access scientific and medical journal publishers, including BioMed Central, Public Library of Science, and Hindawi, are embracing the use of open licenses for research publication. Researchers, after all, want their published research to be as widely read and cited as possible and for others to build on the results of that research. The same is true of the funders who pay the underlying costs of the research.

The Budapest Open Access Initiative declaration, drafted in 2002 by a group of publishers and other members of the research community, noted that open access to published scientific research had the potential to be an “unprecedented public good,” analogous to other open resources such as the human genome sequence and the global positioning system. In all cases, the openness of the resource not only ensures wide usage, but also stimulates innovation by allowing anyone to develop new value-added services that make use of these resources.

Alongside the growth of open-access journals, another important trend in expanding access to published research has been the development of open digital repositories, operated by research institutions and funders, which provide open access to deposited copies of research articles, including those published in traditional subscription-only journals. In response to pressure from funders, most conventional publishers now allow at least the author’s manuscript version of published articles to be made openly available in this way, once an embargo period of 6 or 12 months has elapsed. Although delayed access to a nonauthoritative version is not ideal, it is better than no access at all and represents a stepping stone toward full and immediate open access.

The most well-known open digital repository is the U.S. National Institutes of Health’s (NIH) PubMed Central. This has recently been joined by UK PubMed Central, a mirror repository operated by a consortium of UK biomedical funders that includes the Wellcome Trust and the Medical Research Council. As of October 2007, NIH only requests deposit from authors, whereas the UK funders require their grantees to deposit research. Recent legislative initiatives to strengthen the NIH policy into a requirement have received overwhelming bipartisan support in both houses of Congress, despite strenuous opposition and lobbying from traditional publishers keen to preserve the status quo.

Traditional publishers sometimes suggest that open-access deposit policies such as that proposed by NIH threaten to undermine peer-reviewed research publishing, but the growing success of peer-reviewed journals operating on a fully open-access model provides a powerful counterargument to these claims. Open-access journals generally cover their costs through publication fees (typically paid by the author’s funder or institution) instead of charging for subscriptions. Most open-access journals waive publication charges for authors from developing countries. Many open-access journals based in developing countries receive central institutional support, in which case there may be no need to charge fees to either authors or readers.

An unfortunate consequence of the traditional publishing system has been that in order to seek maximum global visibility and publicity for their work, leading researchers working in developing countries have until recently had no choice but to publish in developed-country journals, even though this has meant that readers in their own country would not have easy access to published results. However, the open-access publishing model is already improving this situation. D. K. Sadu, editor of the Indian Journal of Medical Sciences, published by the Indian open-access publisher MedKnow, refers to the challenge faced by subscription-based journals in developing countries as the “circle of limited accessibility.” These journals typically have limited circulation, which leads to poor visibility and readership. As a result, the journals receive limited recognition and few citations, and this means they attract few authors, few subscriptions, and low circulation—thus closing the circle. The transition of MedKnow’s journals, since 2001, to an open-access model has established a “circle of accessibility.” Manuscript submissions have multiplied severalfold, and the rate of citation of previously published articles from Med-Know’s archive, now easily accessible for the first time, is reported to have increased fivefold. In a short time, open access has thus allowed domestic journals in low-income countries to acquire an international reputation and audience.

Other organizations working on open-access initiatives in developing countries report similar success. Bioline, a joint Brazilian/Canadian project that helps journals from 24 low-income countries make articles freely available online, reports that the annual number of downloads of full text articles from its Web site increased from just 27,000 in 2000 to 2.5 million in 2006. This traffic, which comes from developing and developed countries all over the globe, emphasizes the ability of open access to bring researchers in different countries into a single connected community.

The Directory of Open Access Journals (DOAJ), based at Lund University in Sweden, tracks the rapid increase in the number of journals offering immediate open access to all research articles. Of the 2,700 journals listed in the DOAJ, a substantial fraction come from developing countries, including 222 journals from Brazil and 87 based in India.

The International Network for the Availabilty of Scientific Publications (INASP) is a nonprofit organization that takes a two-pronged approach to increasing access to scientific research in developing countries. Like HINARI, it negotiates to provide developing countries with low- or zero-cost access to existing journals where possible. In addition, INASP works with domestic journals in developing countries to enhance their online presence and accessibility. INASP assists journals in the setup and use of the Public Knowledge Project’s Open Journal System (OJS), which provides an attractive starting point for groups in developing countries wishing to operate open-access journals. For example, the African Journals Online project uses OJS to operate 285 journals.

This appeal of the open-access publishing model in research areas highly relevant to developing countries is strikingly demonstrated by the example of Malaria Journal, an open-access journal launched in 2002 that is already a global leader in its field. Starting with 19 articles in its first year, the journal has grown dramatically to 170 articles in 2007. What led malaria researchers to embrace this new online journal so rapidly? It is clear from author feedback that the journal’s policy of universal access, allowing not only researchers in all countries but also journalists, nongovernmental organizations, public health authorities, and educational institutions to have access, has played a central role. Currently, around 8% of the new articles appearing in PubMed are immediately and freely available online. In the field of malaria, that figure is closer to 20%, demonstrating the level of enthusiasm for open access in this field.

Although malaria kills millions of people each year, it did not attract the interest of traditional publishers because most of the deaths are in the developing world where the market for scholarly journals is very small. This meant that even as philanthropic foundations and governments were channeling increasing funds toward research of relevance to developing countries and global health issues, the traditional publishing model lacked effective means to communicate the results of that research to those to whom it was most relevant. The success of Malaria Journal indicates that there was significant demand for information about the disease even though there was limited money to pay for it. And the journal has achieved high quality as well as popularity; its 2006 impact factor 2.75 is the highest is the highest of any journal in Thomson Scientific’s Tropical Medicine category. As Pascalina Chanda of Zambia’s Malaria Control Centre has noted,

“Previously . . . it was almost impossible to know the latest in malaria research unless you read an abstract or an institution got some hard copies which always arrived a month or more after publication. [Open access] helps in providing the much-needed information on topical issues and one can learn from diverse methods and geographical settings and be able to participate in the global debate on health issues and also provide quality policy information. It also enables us from the developing world to publish our research findings and share the information with other researchers globally.”

The success of Malaria Journal demonstrates that open access can make high-quality research accessible to a global audience and can provide a platform for collaboration between researchers in the developed and developing worlds. The challenge now is to extend this success to other areas and to fully exploit the potential of open access to research to aid international development efforts.

To do this, several actions are necessary:

• Subscription-only journals that publish research of relevance to developing countries should eliminate the barriers that still prevent many of those in developing countries from having access to that research.
• Funders whose focus is global health should ensure that as a condition of funding, grant recipients are required to make the results of their research universally accessible. The policy of the Wellcome Trust, a significant player in research on global health issues, is exemplary in this area.
• Research institutions should ensure that their authors are not discouraged by structural financial disincentives from making their research openly accessible. Institutions currently support subscription journal publishing through central library budgets. If open-access publishing is to compete on a level playing field, similar central support must be available to cover the cost of open-access publication. Many institutions are now setting up central open-access funds, supported by indirect costs received from funders. This is a promising model as it provides a scalable and sustainable basis for open access to the results of research.
• Researchers working in fields of relevance to developing countries should investigate the many options for publishing their research in a way that guarantees fully universal access.
• Last, those involved in international development efforts must consider how best to work with local communities to make effective use of the additional sources of knowledge that are now becoming available. Traditionally, the success of a research journal has been measured in terms of the number of citations that are generated. But in the case of medical research, for example, the goal is not simply to stimulate further research but to generate positive public health outcomes. We need to develop the means to measure and enhance the real impact of medical research in the developing world.