NIH tries to find a happy medium in online debate

By Patricia Daukantas

GCN Staff

The National Institutes of Health has revised its plans for a comprehensive online biomedical publication'seeming to accede, at least halfway, to the objections of some of the nation's pre-eminent scientific journals.

Dr. Harold E. Varmus, NIH's director, originally proposed that the agency set up a database of full-text biomedical research articles [GCN, July 12, Page 1]. Because NIH is also a major source of research funds, critics raised the possibility of a conflict of interest because the agency also would control the content of the massive article repository.

The revised plan would make outside groups, such as scientific societies and commercial journal publishers, responsible for providing article texts to the NIH site. The database, whose tentative name has been changed from E-biomed to PubMed Central, now would encompass information on all life sciences, from botany to agricultural research.

One aspect of Varmus' proposal remains controversial: the division of PubMed Central into two archives with different criteria for their content.

One section would house articles that have gone through traditional peer review by fellow scientists. The other section would hold reports that received less rigorous scrutiny.

The impetus to add plant and agricultural research to the database came from researchers, said Judith Axler Turner, special assistant for the project at NIH. 'There's a lot of overlap in molecular biology between plant and animal and human research,' she said.

NIH is starting to work with publishers of existing journals to supervise the non-peer-reviewed section of PubMed Central, Turner said, but she would not name the groups involved.

From the boss

PubMed Central would likely cost NIH $2 million to $3 million per year, Turner said. It is scheduled to go live in January. She said the PubMed announcement on the Web last month came straight from Varmus, a Nobel Prize winner who has headed NIH since 1993.

'This is something he cares about a lot, that he gave a lot of thought to,' Turner said.

David Lipman, director of the National Center for Biotechnology Information within NIH's National Library of Medicine, is in charge of the implementation details, Turner said.

Between the time Varmus first floated the E-biomed proposal and the most recent announcement, critics charged that NIH would encourage scientists to bypass print journals in favor of Web publishing, threatening those publications' existence.

Dr. Richard Glass, a psychiatrist serving as interim co-editor of the Journal of the American Medical Association, expressed concern last week about the potential economic effect of PubMed Central.

JAMA's main source of revenue is advertising targeted at physicians, Glass said. 'You have to have some sort of advertising to support the infrastructure and the personnel who coordinate the review process,' he said. Unlike advertising-free scientific journals, JAMA does not assess page charges to authors for publishing their findings.

Last week, JAMA and 10 other journals published by the AMA began making full-text articles available online. Glass said the articles will be free only for a trial period of about six months.

He said JAMA has serious concerns about making preliminary research material available to physicians and the public via the non-peer-reviewed section of PubMed Central.

The Massachusetts Medical Society, publisher of the New England Journal of Medicine, has written a letter to Varmus outlining its concerns about the proposed Web repository. The letter, approved by the society's publications committee but not yet sent, refers to the NIH director's earlier plans, not to PubMed Central.

Funds, credibility

The society's letter expresses concern that the repository could divert funds from individual research grants during lean budget years and that releasing unreviewed clinical trial data could 'undermine the quality of readily available scientific data and analysis that have a direct impact on human health and safety.'

In addition, the letter notes, Congress might set limits on the types of reports posted on an NIH Web site, as it has limited use of federal funds in several politically controversial fields of biomedical research.

'NIH should avoid duplicating these efforts [of scientific societies] and should instead focus on facilitating access to published material and other scientific information,' the Massachusetts Medical Society said.

Given the amount of life science research already available online, PubMed Central may duplicate some efforts, said Mary Waltham, an independent publishing consultant in Princeton, N.J. Waltham recently stepped down as president of journal publisher Nature America Inc. of New York.

As an alternative, Waltham cited PubScience, a new database of journal abstracts that was expected to come online late last week under the sponsorship of the Energy Department's Office of Science and the Government Printing Office. PubScience contains abstracts plus links to articles that remain on publishers' Web sites.

Marvin Stodolsky, a biologist with Energy's Human Genome Program, said he had not studied the new NIH proposal in detail but said that he believes the overall concept of Web publishing is important. He said he was speaking as an individual, not a department representative.

'I think it's a satisfactory evolutionary step midway between complete freedom and legitimate concerns of the professional societies,' Stodolsky said of PubMed Central.

Scientists outside the United States do not have access to current information because scholarly journals are too expensive, Stodolsky said. Putting biological articles online will be 'empowering for the world,' he said.

He dismissed the squabble over the lack of peer review of some PubMed postings. He noted that articles posted via an existing physics preprint server on a Web site run by Los Alamos National Laboratory are not edited. The submissions are of high quality, Stodolsky said.

'Scientists don't want to make fools of themselves' he said.