Intended for healthcare professionals


Paying for

BMJ 2003; 327 doi: (Published 31 July 2003) Cite this as: BMJ 2003;327:241

This article has a correction. Please see:

  1. Tony Delamothe, web editor,
  2. Richard Smith, editor
  2. BMJ

    From 2005, some users will have to pay for some content

    Almost 10 years after it began, the BMJ ‘s experiment of allowing free access to everything on its website will come to an end. The BMJ Publishing Group board has decided that, from January 2005, visitors to should pay for access. The resulting revenue should not only defray the website's current costs but also allow us to fund further developments.

    Exactly which content will be behind access controls, for how long, and for whom has yet to be decided. We can, however, assure BMA members (including student members) and users from the World Bank's list of 120 low and lower middle income countries1 that access will remain free to them. We had hoped to extend this dispensation to medical students everywhere, but the difficulty of verifying the credentials of an estimated 14 000 medical student visitors each week makes this unlikely. Access to student will remain free.

    The model we are currently finalising for is likely to make all content free for a week or two after publication. Most of it will then be behind access controls for a year or more. Content that we intend keeping free throughout this period includes abstracts of articles, rapid responses, and the Editor's Choice column. All of BMJ Careers (Career Focus, recruitment and of course advertisements, and career services) will remain free.

    Our intention is to continue making the full text versions of our original research articles freely available on and PubMed Central from the day of publication. The abridged versions of these articles (“paper short”) are likely to be behind access controls.

    The board's decision to introduce access controls was precipitated by anxiety over falling library subscriptions to the paper journal. In common with many scientific journals, numbers of subscribers have been falling steadily over the past decade, but the BMJ ‘s rate of decline has increased recently. Our current total is 9% lower than the same time last year, whereas the publishing group's 26 specialist journals, 25 of which have access controls, have experienced falls of only 4%. Access controls have given these journals the possibility of selling electronic subscriptions—an opportunity lost to because of its free status.

    The fall in BMJ subscription revenue is likely to accelerate as increasing numbers of libraries abandon paper subscriptions in favour of electronic ones. A recent survey of large American academic libraries suggests that between 50% and 80% of journal subscriptions will be solely electronic within five years (Morna Conway, personal communication). So long as access to the electronic BMJ remains free, we will have nothing to replace the revenue lost from cancelled paper subscriptions.

    Although figure 1 shows that subscriptions account for only 12% of the BMJ ‘s total revenue, our problem is that our two other main sources of revenue are also under threat. Job advertisements, which account for 61% of our revenue, are threatened by an NHS job website for England (sites for jobs in Scotland and Wales already exist). And spending by UK pharmaceutical companies is down 8% over last year, with fewer new drugs to promote.2

    Although it's unclear whether the subscription model for scientific journals will survive in the long term, it still has a few years of life left in it.3 Like any business, the BMJ is obliged to preserve as many different revenue streams as possible. For example, had we been solely reliant on the pharmaceutical industry for financial support—like many publications—our recent theme issue on the need to disentangle doctors from the drug industry4 might never have happened.

    Nevertheless, the end of free access is likely to affect the sudden explosion in exposure to the BMJ. The number of computers accessing recently reached 132 000 per week, equivalent to about 185 000 individual users per week, or nearly twice the number of people who receive the paper journal (fig 2). Nearly two thirds of these users live outside the United Kingdom, with half rarely or never seeing a copy of the paper BMJ—a legacy of our previous policy, strongly endorsed by BMA members,5 to make the full electronic journal freely available to doctors everywhere.

    Fig 2
    Fig 2

    Weekly circulation of BMJ, 1840-2003

    While the journal's international influence has undoubtedly increased, it's much harder to quantify this than to quantify how much costs (about £4.60 per BMA member last year.) One proxy for influence, at least within the academic community, is the impact factor. The BMJ ‘s impact factor has risen much faster recently than the average for the group's specialist journals, which have remained behind access controls (fig 3).

    Fig 3
    Fig 3

    Impact factors, 1990-2002

    We expect to charge individuals between £10 and £20 ($16 to $33; €14 to €28) a year for an online subscription; charges to libraries will vary according to the number of people they serve. For users accessing the website via a library that subscribes, access will feel free. Non-BMA members whose libraries do not yet subscribe to the BMJ have nearly 18 months to persuade them to do so, or to save up for a personal subscription.


    • Competing interests TD and RS both work for the BMJ Publishing Group Ltd. Although the group's profits are likely to rise with the introduction of access controls to, their salaries will not be affected.


    View Abstract