New Open-access Policy Under Discussion

The idea is simple: make peer-reviewed research articles freely available. Such public access helps fulfill a university’s most valued role, creating new knowledge for the benefit of society.

For nearly a year, researchers funded by the National Institutes of Health have been required to post their latest peer-reviewed manuscripts at the public NIH digital archive, PubMed Central, so taxpayers can access the products of their tax dollars.

Many of those scientists are still learning what it takes to comply with the NIH public access policy. The key new tasks include negotiating conforming copyright-transfer agreements with the journals, depositing the accepted peer-reviewed manuscripts in PubMed Central if the journal does not submit the published paper, and noting the public access citations in reports back to NIH. Some researchers also wonder how they can make their work publicly accessible even if it is not funded by the NIH.

These issues could be solved by a new open-access policy under discussion in the Harvard medical community. A team at the Countway Library has developed a two-pronged strategy to help scientists smoothly manage the latest changes in scholarly publishing and further expand the open-access model at Harvard.

For the last year, librarian Scott Lapinski has been holding drop-in brown-bag information sessions and by-request department tutorials on the Quad and at affiliated institutions to walk through the steps of complying with the new public access publishing rules.

A longer-term solution is an HMS-wide open-access policy and repository to streamline NIH-funded article deposits and to showcase the range of scholarly contributions by medical, public health and dental faculties, said Alexa McCray, co-director of the HMS Center for Biomedical Informatics at the Countway and HMS associate professor of medicine at Beth Israel Deaconess Medical Center.

“The creation of new knowledge to benefit society is at the heart of a research university,” McCray said. “This is definitely true in biomedical research. The value lies in effective dissemination. Restricted access is at odds with that imperative.”

A voluntary online repository called HMScholar already exists at the Countway website for those authors who have retained the necessary copyright terms (https://www.countway.harvard.edu/menuNavigation/libraryServices/countwayOpenAccess.html).

Under an open-access policy, the system would automatically make the NIH-required submissions to PubMed Central and enable the University to track NIH compliance better. “We would take care of everything,” McCray said.

The policy would simplify the copyright negotiations, allowing authors to retain copyright, granting nonexclusive publishing rights to the University and journal, and providing immediate full access to published papers. Open access facilitates scientific collaborations and enhances education, said McCray, who is in discussion about the policy with key HMS and affiliate groups. A waiver of open access would also be available.

The policy would be similar to those adopted last February by the Harvard Faculty of Arts and Sciences and in June by the Harvard Law School. The online collection would be integrated with the new University-wide open-access institutional repository DASH (Digital Access to Scholarship at Harvard) in Cambridge, said Amy Brand at the Harvard Office for Scholarly Communication. Plans call for DASH to use the Countway mechanism for deposits to PubMed Central.

Managing the Details

In the meantime, NIH-funded researchers on campus and at affiliated hospitals must comply one-by-one with the public access policy. A similar policy by the influential private research organization Howard Hughes Medical Institute affects the 30 Hughes investigators in the Harvard medical community.

Interest in Lapinski’s briefings stepped up this fall when the NIH began sending “Dear Principal Investigator” letters to remind grantees to post and cite their work on PubMed Central in progress reports or else risk suspension or termination of their grants, said Lapinski, digital resources and services librarian at the Countway.

“The NIH is serious about this,” said Tiffany Blackman, an administrator in the Medical School’s Sponsored Programs Administration. Blackman teams with Lapinski for Quad-based briefings and with host librarians at affiliated hospitals. The policy applies to any author on a paper using “any NIH funding, from 1 percent to 100 percent of the work,” she said. It also applies to “any source—salary, supplies, or equipment.”

Though researchers can publish anywhere they choose, one of the easiest ways to comply with the NIH policy is to publish in a journal that automatically submits its content to PubMed Central. NIH allows publishers to impose up to a 12-month embargo on public viewing, but the researcher still receives the crucial PMC citation.

For all other journals, there are almost as many variations as there are titles.

“A lot of publishers have some way of ‘cooperating,’ although they don’t always make it easy for authors,” Lapinski said. “Sometimes they submit the paper [to PubMed Central] automatically on the author’s behalf. Sometimes they expect the author to do the work. Sometimes they will put in the accepted peer-reviewed manuscript [the minimum required by NIH]. It’s confusing at this stage. There’s no one-size-fits-all policy.” Some publishers also try to charge fees to submit a final published version, he said.

Scholarly publishers are pushing back in other ways. In early February, Rep. John Conyers (D-MI) reintroduced a bill that would effectively reverse the NIH Public Access Policy and make it impossible for other federal agencies to put similar policies into place.

Watch Out for Copyright

To comply with NIH policy, the most important step for researchers is the copyright-transfer agreement, which usually happens at acceptance, but can occur earlier. As a term and condition of an NIH award (and for HHMI investigators), authors must insure they do not sign something that prevents their manuscript from being published in PubMed Central.

“It’s an eye-opener to most researchers that they as creators have control over the copyright,” Lapinski said. “All a publisher needs is a nonexclusive license to publish what you wrote. It’s a misconception that you have to sign away whatever the publisher tells you.”

A fill-in-the-blanks addendum conforming to NIH policy is available to Harvard researchers online (http://countway.harvard.edu/publicaccess).

In testimony to Congress in September, NIH director Elias Zerhouni reported that well over half of NIH-funded articles were being submitted to PubMed Central since the mandatory policy kicked in on April 7, 2008, compared with less than 8 percent in the previous two years under the voluntary policy.

Web update: By the end of 2008, about half of all NIH-funded papers subject to the policy were being posted to PubMed Central by publishers or individual researchers through the four NIH-approved methods (http://publicaccess.nih.gov/index.htm). The rates appear to be increasing, said Neil Thakur, special assistant to the NIH deputy director for extramural research. The PMC website contains more than 1.75 million articles, including the NIH public-access collection, and is used by about 400,000 people a day to retrieve about 600,000 papers a day. Thakur and his colleagues are crunching the numbers and expect to have a more detailed report out in April.

For more information on complying with NIH copyright policy or to schedule a briefing, e-mail Scott Lapinski at paul_lapinski@hms.harvard.edu or see http://countway.harvard.edu/publicaccess. For more information about the HMS open-access policy under discussion, contact Alexa McCray at at alexa_mccray@hms.harvard.edu.