The fourth in an occasional series of articles on changes in scientific publishing.
--
With two ill children facing blindness and early death from a rare genetic disorder, Sharon Terry needed information -- information her doctor could not provide her.
Information on the disease was available, much of it paid for by federal research dollars, but it was out of reach in expensive, specialized medical journals. The journals had become so costly, closely held or difficult to access that the Terrys -- like many others desperately researching their family's health problems -- ended up using other peoples' passwords and getting students to obtain the material for them.
"Here I had two kids suddenly with a very serious disease," Terry told United Press International. "I found no clinician that understood this disease and then on top of that I couldn't get access to information that had been already gathered and published by federal tax dollars. I couldn't get that information."
To ease access for patient's families and other seeking the latest research, the National Institutes of Health proposed a new policy late Friday that would mandate delayed free Web access to any journal paper based on NIH-funded research. To do this the new policy would require authors to submit a copy of their final edited journal manuscript to the National Library of Medicine. Six months after its original publication, NLM would make the full text of the paper available through its PubMed database.
Such an approach would constitute a sea change for scientific publishing, where most material is copyrighted by the journals and made available only through subscription. Both print and electronic subscriptions, generally bought by libraries, typically run $100 to $500 per year. Some can be as high as $15,000 to $20,000 annually with electronic access. Contracts can be restrictive and may not, for example, allow access by "walk in" visitors such as the Terrys.
The delay is designed to protect publishers who are adamant that posting the articles will undermine the financial health of the journals. Unless they have exclusive rights to control the material, several publishers told UPI, the scientific societies that depend on journals for revenue and as a membership benefit will suffer. Some societies are indeed concerned that even delayed free access will kill off their journals, the main benefit they offer members. Others, however, already are posting articles on the Internet for free.
Dr. Elias Zerhouni, director of the National Institutes of Health, told a Washington meeting of patient advocates and society representatives Aug. 31 that his agency would be open to feedback and would work out operational details. One negotiable area might be the six-month delay. That might be enough of a delay for weekly or monthly journals, but it could cause unintentional damage to the subscription base of quarterly journals.
The public can comment on the draft of the policy for the next 60 days. It is likely some sort of policy will go forward, however, driven by the rationale that the taxpayer is entitled to access, as well as the NIH's desire to manage its portfolio of research better.
"When we do clinical trials, we would like to be able to link that to the paper that was published," Zerhouni told UPI. "The way to do that is to have an archive, an NIH research archive, and frankly we don't have that. ... It would allow us to also do peer review more effectively, do analysis of our portfolio more effectively so we can see what the entire production of agency is. This would be a great tool from the standpoint of NIH," he added.
"If we have this collection of the NIH papers, then we can mine it to look at patterns and evaluate (if) we are really getting the effect that we want with new kinds if funding programs in terms of encouraging new directions in biomedical research," explained David Lipman, director of the National Center for Biotechnology Information. "Are we being as effective as we can possibly be?"
The NIH's desire to use the database for grant management would probably preclude one solution proposed by publishers and some open access advocates: having authors publish their papers on their own or their institution's Web sites. Search engines might make the papers generally accessible, but it is unlikely that NIH could use such scattershot postings to do its grant evaluations.
The House has weighed in on the issue approving language, mandating that NIH research be published in the PubMed database, starting in fiscal year 2005, which begins Oct. 1. The Senate already is under pressure by the publishers not to support the move.
Even if NIH does go forward -- and it may be able to do so on its own initiative -- the action will not solve the problem. The NIH may be the largest single funding source for biomedical research in the United States, but the majority is still funded by other organizations.
It is possible, however, other organizations would follow the lead of NIH and mandate similar policies. During interviews with UPI, it was clear the publishers thought such a shift was quite possible.
The new NIH repository policy could have international implications, because Japan, the United Kingdom and France are considering setting up national archives as well. Those archives could be shared with a U.S archive, the NIH said in its draft proposal.
Pat Furlong, executive director of Parent Project Muscular Dystrophy, just wants the papers posted on the Internet. Her son, a teenager with MD, died because he was given an inhalational anesthesia in the hospital.
"The physician did not have the access to the information he needed for my son to survive," Furlong told the meeting.
People in towns far away from major medical centers, like her town of Middletown, Ohio, need to have these papers on the Web to be able to access them. That could make a difference to other families, she told UPI.
--
E-mail [email protected]
