Opening the lid on open access

BMJ 2008; 336 doi: (Published 27 March 2008) Cite this as: BMJ 2008;336:688
  1. Susan Mayor
  1. 1London

    As the National Institutes of Health make open access compulsory for research they have funded, Susan Mayor looks at the policies of other funders

    The chief public funding body for medical research in the United States, the National Institutes of Health (NIH), is introducing an open access policy from next week. All papers resulting from research that it has funded will have to be made freely available to the public no later than one year after they have been published.

    This is the latest policy from key research funders to promote open access to research findings (table). It is based on the argument that the public should have free access to results from research that it has funded, and researchers should have free access to papers they have written or reviewed rather than have to pay subscriptions or single access fees to journals. Open access publishing also makes research freely available to help advance research around the world.

    Open access policies

    View this table:

    There are two main publishing models for open access. Researchers can publish their findings in a journal that offers an open access option, such as journals published by BioMed Central and the Public Library of Science, by contributing towards the costs associated with publication. Alternatively, they can submit their research to a journal that charges readers to access papers provided that the research is placed in a free online repository after a certain time. The BMJ uses a third way, however, charging neither authors nor readers for full and immediate access to research papers, funding this from the journal’s overall revenue.

    The NIH policy uses the term “public access” rather than the more common “open access.” It makes it mandatory for researchers to make electronic versions of any peer reviewed papers they publish that are based on NIH funded research available on the National Library of Medicine’s PubMed Central, the free NIH digital archive of full text, peer reviewed journal articles, no later than one year after publication. This replaces a previous voluntary policy, which had a poor compliance rate, of less than 10%. The change follows a law passed by Congress at the end of December 2007 that requires research funded by the NIH to be openly accessible (BMJ 2007;335:906; doi: 10.1136/bmj.39384.638241.DB).

    Policies mandating open access to publicly funded research have been in place in Europe for some time. The Wellcome Foundation, a large UK based charity that funds medical research, introduced a policy in October 2006 that requires electronic copies of research papers accepted for publication in peer reviewed journals that they supported wholly or in part by its funding to be made available through PubMed Central and UK PubMed Central “as soon as possible and in any event within six months of the journal publisher’s official date of final publication.”

    The UK Medical Research Council (MRC) has a similar policy, also introduced in October 2006, that requires researchers to make papers freely available within six months of publication.

    In its latest guidelines, the scientific council of the European Research Council (ERC) reduced the time from one year to six months by which peer reviewed publications from council funded research projects should be available on open access. In an indication that the timescale may be shortened further, the guidelines noted that “the ERC is keenly aware of the desirability to shorten the period between publication and open access beyond the currently accepted standard of six months” (BMJ 2008;336:176-7; doi: 10.1136/bmj.39468.592280.DB).

    Universities are also starting to require their staff members to make their papers freely available. In February Harvard University’s faculty of arts and sciences, which does not include the medical school, although a university spokesman said it is working on an open access policy, adopted a policy that will put faculty members’ papers in an open access repository hosted by the university. Details are yet to be decided as to how quickly faculty members would have to place papers in the repository.

    One potential problem for researchers is that they have to find the costs of publishing in an open access journal if they choose this route. The NIH has said that it will reimburse publication costs. In the United Kingdom, the Wellcome Trust and the MRC provide grant holders with additional funding to cover open access charges, providing these have been included in grant proposals and that the costs fall within the period of the grant.

    Another problem is enforcing open access policies. “They are not easy to monitor in practice,” acknowledges Tony Peatfield, head of corporate governance at the MRC. He notes that a recent survey of all units funded by the MRC shows that more than 90% had an open access policy in place. However, there is no mechanism for recording publications by MRC grant holders working in other institutions. In the United States, the NIH will ask research to include evidence of submissions in grant applications and reports as part of monitoring its public access policy.

    Publishers of journals are having to introduce new mechanisms to support open access. Peter Ashman, publishing director of the BMJ and BMJ Journals, said, “We remain committed to supporting open access and to fulfilling the criteria set out in the NIH directive.” He explained that the company introduced a hybrid option for all the BMJ Group’s specialist journals in 2006. This allows authors to pay a subsidised fee to ensure that their article is freely available from publication.

    Mr Ashman also commented on the fact that all original research articles published in the BMJ are freely available from publication. “I like to think of this as ‘pure open access’ as we don’t charge subscribers or authors for this service.” He added, “Other publishers are having to make adjustments to their publishing policies to accommodate the demands of the NIH and research communities. They will need to make similar commitments to offer hybrid open access policies in order to continue to attract the high quality research funded by NIH.”

    It is still relatively early days to determine the effects of open access policies. Mr Peatfield reported that some researchers funded by the MRC have found that journals without open access arrangements have introduced mechanisms to facilitate this when necessary. Research so far has indicated that researchers may gain greater exposure for their work because studies have shown that open access articles are cited more often than non-open access articles from the same journal (; BMJ 2005;330:1128; doi: 10.1136/bmj.38422.611736.E0).


    View Abstract