Open access publication fees at the BMJ

From this week the BMJ will start asking authors of eligible research articles to pay a publication fee. All submitted research will continue to be judged entirely on its importance, originality, quality, and relevance, and not on authors’ ability to pay. Editors and reviewers will not know whether a fee is payable: the fee will be requested only after a study is accepted, and administrative staff will handle payments and associated correspondence. Moreover, all research in the BMJ will remain openly accessible, regardless of whether a fee has been paid for its publication.

The publication fee will be £2500 for each accepted research article. This fee will apply only when the funder of the research that is reported in the article has already pledged to pay for open access publication and when authors can claim the BMJ fee, in full, from their funder for that specific piece of research. Consideration of the paper is not related to whether authors can or cannot pay the fee. We will ask for the fee only once we have accepted a paper and we will send an invoice only once authors tell us (via openaccess.bmj{at}bmjgroup.com) they can claim the fee. Seeking and processing fees will not delay editing or publication.

We neither wish nor intend to deter anyone from submitting research to the BMJ. We appreciate that research—even in high income countries—is often unfunded or may be supported by a patchwork of different resources, and that even those funders who support open access will not always cover extra costs arising downstream from a primary study. So the options for all research authors will be either to pay the fee or not to pay it: the BMJ will not have a sliding scale of fees. Nor will we seek a fee for publishing other types of article, such as editorials, clinical reviews, analysis articles, or research methods and reporting articles.

We are introducing this policy as the next step in our efforts to ensure the sustainability of open access publication of research in the BMJ, and we are doing so in the spirit of experimentation. Many research funding organisations, sponsors, and universities now provide grants that cover journals’ fees for open access publication, and most other open access journals levy such fees to cover the costs of peer review, journal production, and online hosting and archiving. Because the BMJ is providing open access to research and funders are providing fees, it makes good sense for the journal to defray at least some of its publishing costs in this way.

The BMJ was one of the first scientific journals to make the full text of its articles freely available online, starting in 1998. From very early on our policy included immediate deposition of the full text in the National Library of Medicine’s archive PubMed Central, as well as allowing authors to retain copyright and to share and reuse (non-commercially) their articles. Although research has remained freely available to everyone, from 2005 access controls were gradually introduced for non-research content (although the whole journal remains freely accessible to institutions in 113 low and low-middle income countries through the BMJ Group’s membership of the HINARI initiative (http://group.bmj.com/products/journals/countries-with-free-access).1

Access controls were introduced to maintain subscription revenue as a contribution to the costs of commissioning, peer reviewing, editing, and publishing the BMJ’s editorials, education, comment, features, and news. Using subscriptions to non-research content to subsidise access to research distinguished the BMJ from the other open access journals that sprang up subsequently. Yet the BMJ was not formally recognised and listed as an open access journal until October 2008 (http://resources.bmj.com/bmj/about-bmj/policies/press-release-greater-openness-is-the-future2014the-bmj-and-open-access). Now the BMJ has a full set of open access credentials: a Creative Commons licence (http://resources.bmj.com/bmj/authors/editorial-policies/copyright), membership of the Open Access Scholarly Publishers Association, and—as from now in a limited way—authors’ fees.

Some publishers with an open access option have promised to titrate journal subscription rates2 and site licences for institutions3 against revenue from authors’ fees, once that income exceeds a certain threshold. The BMJ has no plans to do this, for several reasons. Firstly, our foray into charging authors’ fees is an experiment, and we aim to evaluate its success before going any further. Secondly, we do not expect authors’ fees for BMJ research articles to come close to covering the costs of peer review for submitted research, given that we publish around 300 research articles a year but receive more than 4000, and given that only a minority of published research is likely to come with open access funding. Thirdly, we aim to increase the amount of non-research content we publish, in response to reader feedback requesting more education, comment, and news alongside the BMJ’s peer reviewed research. Author fees cannot and should not be expected to cover the costs of creating this extensive and growing range of clinical and journalistic material, which will continue to be funded through subscriptions and advertising (http://resources.bmj.com/bmj/about-bmj).

Last but definitely not least, we are pleased to introduce a new sister journal to be launched this autumn, BMJ Open (www.bmjopen.bmj.com). This is an online open access journal that will provide rapid publication of research across all medical disciplines and therapeutic areas, and which will welcome both high and lower impact studies to ensure that well conducted studies find a home where they can be fully reported. We particularly hope that BMJ Open will attract work that deserves publication but does not quite meet the specific needs of the audiences of the BMJ and the BMJ Group’s specialist journals. All authors whose work is accepted by BMJ Open will be asked to pay an article processing charge. Fees will be waived for authors without the means to pay, and as with the BMJ, editors will be unaware of the payment status of articles when they are making their decisions on publication. All articles will be openly peer reviewed and, if accepted, will be published with reviewers’ comments and other background information. As well as publishing full research reports in the traditional way, BMJ Open will encourage transparency at all stages of the research dissemination process by publishing study protocols, pilot studies, and pre-protocols and by facilitating the sharing of raw data either as additional electronic material or through direct links to data repositories.

With the introduction of author fees for some of the open access research published in the BMJ, and with the launch of our new online journal BMJ Open, the BMJ Group continues its support for both open peer review and open access to peer reviewed research. We hope you will join us by sending us your work.