Published online 7 April 2009 | Nature | doi:10.1038/458690a
Corrected online: 15 April 2009


Open-access policy flourishes at NIH

Researchers, institutions and publishers have complied with the mandate, but it still has its opponents.

One year on, advocates of free public access to scientific literature are calling a law that requires researchers at the US National Institutes of Health (NIH) to make their manuscripts publicly available at the PubMed Central repository a success. At the same time, the measure continues to be challenged by a senior congressman and some publishers.

Click for larger image

Since the legal requirement that NIH-funded researchers make their manuscripts publicly available after acceptance for journal publication came into effect last April, the number of articles being approved by their authors for processing by the repository has more than tripled. In March 2009, 6,425 such original articles were approved by their authors for processing; a year earlier, the number was 1,852 (see graph). The articles become available through PubMed Central no more than a year after their journal publication. Author compliance "has been dramatically altered" by converting an "anaemic" voluntary policy into law, says former NIH director Harold Varmus, now president of the Memorial Sloan Kettering Cancer Center in New York and a keen supporter of open-access initiatives.

But the policy still has opponents. "This so-called 'open access' policy was not subject to open hearings, open debate or open amendment in Congress," John Conyers (Democrat, Michigan), chairman of the House Judiciary Committee, wrote on The Huffington Post website last month (he declined to be interviewed for this article). In February, Conyers re-introduced a bill from the last congressional session that would amend US copyright law to forbid the NIH making funding conditional on manuscripts being publicly accessible. However, congressional observers say that the bill has little chance of going anywhere this year.

Open-access policies have caught on around the world in recent years. Britain's Wellcome Trust, the Italian National Institute of Health (ISS), the European Research Council and many others have implemented similar mandates, including all seven UK research councils. Disease groups such as the high-profile US foundation Autism Speaks have done the same, as have the Massachusetts Institute of Technology and the Faculty of Arts andSciences at Harvard University.

The reason that the policy has succeeded at the NIH "is that there has been cooperation — whether they wanted to [comply] or not — by grantees, by extramural staff, by the universities and by publishers", says David Lipman, director of the NIH's National Center for Biotechnology Information in Bethesda, Maryland. He believes that the new requirement is at least partly responsible for the increased use of PubMed Central; there were approximately 550,000 articles downloaded from the site in March 2008, and 680,000 last month.

John ConyersCongressman John Conyers has tabled a bill to challenge the NIH open-access requirement.M. Wilson/Getty Images

Much of the increase in author participation has occurred since the autumn, when the NIH began e-mailing reminders about the new requirement to investigators submitting grant proposals or progress reports that cited agency-funded papers lacking PubMed Central identifiers. "We expect every paper falling under the NIH public-access policy to be posted to PubMed Central, and we are reviewing every NIH award to make sure that happens," says Neil Thakur, who oversees the policy for the NIH's Office of Extramural Research.

However, opposition to the law persists among some publishers. "What is being done by this policy is imposing a specific model of publication that we think the government has no business imposing," says Allan Adler, legal counsel for the Association of American Publishers (AAP) in Washington DC. (Nature is a member of the AAP but cooperates with and supports the NIH on open access.)

Because of the delay of up to a year on NIH uploads to PubMed Central, it is difficult at this stage to gauge the impact of the policies on societies and other publishers that rely heavily on subscriptions for revenue. The recession is driving down endowments and with them library budgets, leading independently to some degree of subscription cancellations.

Heather Joseph, executive director of the Scholarly Publishing and Academic Resources Coalition (SPARC), an advocacy group in Washington DC with an open-access agenda, says that the consortium's members are reporting large-scale cuts in journal subscriptions. "But they are across disciplines, completely due to the economic meltdown and not the NIH policy," she adds.

But Martin Frank, executive director of the American Physiological Society in Bethesda, says that "in an environment where access is readily available whether after 12 months or 6 months or immediately, the subscription model starts wobbling". Frank predicts that, as subscription revenues tank, publishers will be forced to levy stiff fees on authors for publishing. He notes that the open-access journal group the Public Library of Science has boosted its publication fees for PLoS Medicine and PLoS Biology from US$1,500 in 2006 to $2,950 today; the author fees for four of its other five journals have risen to $2,300.


And last week, the formerly open-access Journal of Visualized Experiments (JoVE) abruptly switched to a subscription model, charging individuals $99 per month and larger institutions up to $2,400 per year. "While we do support open access — we think it's a great idea — we simply cannot survive with the open-access model, at least now," says Moshe Pritsker, JoVE 's editor-in-chief and co-founder.

Scientists seem to be coping with the changes, although there are some grumblings. "I don't have any complaints about the procedures. They are pretty user-friendly," says Carol Mason, a neuroscientist at Columbia University in New York. When she published a paper in the Journal of Neuroscience last month, the journal submitted the paper to PubMed Central for her. After receiving an e-mail from the NIH, it took her "a couple of minutes" to click through a step-by-step guide and log into the processing system, and another 15 minutes to review the attached PDF of her paper before hitting 'approve'. 


There is a Correction associated with this article (_Nature_ *458,* 819; 2009). The number of articles downloaded from PubMed Central represent the average usage on a typical weekday, not a monthly total as we state.


If you find something abusive or inappropriate or which does not otherwise comply with our Terms or Community Guidelines, please select the relevant 'Report this comment' link.

Comments on this thread are vetted after posting.

  • #60670

    The NIH Public Access Policy addresses the public’s growing need for high-quality health information and promotes accelerated scientific advancement in the biomedical sciences.i

Commenting is now closed.