The Proceedings of the American Thoracic Society 2:2-3 (2005)
© 2005 The American Thoracic Society
The Unintended Consequences of Noble Intentions to Inform
Alan R. Leff, M.D., Editora,
Steven D. Shapiro, M.D., Editorb and
Edward Abraham, M.D., Editorc
a Proceedings of the American Thoracic Society
b American Journal of Respiratory Cell and Molecular Biology
c American Journal of Respiratory and Critical Care Medicine
Scientific discovery is a cornerstone of civilization. In the medical community, it is the basis upon which process is elucidated and innovative therapies are developed. It follows that a well informed public deserves access to this information, which is discovered at least in part through public funding of research grants. Disclosure of scientific data has become an issue of considerable focus in the broadcast and print media, and a central focus of the National Institutes of Health, which has mandated that all scientific information gathered in any part from grant award funds from the NIH be submitted for publication on the Institute's website, PubMed.
At first pass, this may suggest some spectacular advantages to the public. Every citizen with access to the internet will be able to access all the information at no charge at the earliest possible time. Unfortunately, the policy fails to consider some vital aspects of the structure of the system by which scientific information is developed. Over many decades, modern biomedical science has been developed in large part from meetings of privately funded, nonprofit scientific societies, such as the American Thoracic Society. Here, thousands of scientists meet, present papers, and engage in vigorous discussion of the merit of the work. Many of the papers presented at these meetings eventually are published in journals that deal with the topic of interest. The most important component of this process is peer interaction and review. The most prestigious journals are those with the most rigorous peer review. Hence, journal publication not only reviews the validity and importance of the work but "ranks" the importance of the findings by the reputation of the journal in which it is published. This offers the reader both the reassurance that other scientists have reviewed the work and some notion of its relative quality.
The current system is not perfect, but it is still fair to say that the quality of science emanating from American scientific journals is the best in world. The system has evolved to insure both ranking and quality control. Unfortunately, the process is not immediate. The peer review process takes some time. Careful review is essential. Poorly reviewed work, or work that the public cannot distinguish as borderline or even dubious in quality, poses more of a threat to scientific development than a benefit.
Recent reports in the media, both broadcast and print, have suggested that the process of publishing papers in scientific journals serves to deny the public access to novel therapies as they arise and leads to "duplicate" charges to the public for the dissemination of the material. These and other concerns deserve some in-depth consideration, as it is clear that mandatory publication in government-sponsored journals may defeat the very system that NIH has so long worked to establish, i.e. the dissemination of reliable scientific information to the public after rigorous peer review. It should be noted that funding novel research projects at NIH—the first limb of scientific discovery—takes a minimum of 9 months from time of submission of a grant application. It would follow that appropriate review and opportunity to vet the reports would also require a similarly rigorous approach. It is fair also to say that the NIH rarely, if ever, bears the full cost of funding the research that they sponsor. Some other suppositions that form the basis for the PubMed initiative are worthy of further consideration.
- 1. Supposition: The public is being denied ready access to valuable scientific information, which, if disseminated immediately upon completion of the work, would alleviate human suffering and possibly save human life. This often uttered (and printed) cause célèbre presumes circumstances that simply do not exist. Ideas for new therapies arising from NIH grants are theoretical constructs designed to inspire development of new therapies. Even identification of clear targets requires an average of 7 to 10 years to complete development for approved use in humans. There are few true "discoveries." There are technical advances—and the accompanying controversies, which need to be resolved—as scientists sort through the highly technical literature that is many steps away from the development of therapies that will benefit patients. In all of the discussions about this issue, no one has identified a therapy that did not reach public access because of the current system of publication review and the inherent delay in evaluating the validity of the findings. New potential therapies are announced daily through press releases and rapid pre-publication announcements. It also should be noted that nearly every major scientific journal is available to all investigators on institutional websites at no cost to the government or the reader. Lay readers have access to these sites in every state through their state universities and some of the larger public libraries.
- 2. Supposition: The public is paying twice for publication of scientific information—once to fund it and again to pay a private publisher to print it. This, of course, does not account for the cost to the NIH to publish the papers on the website. Currently, all journals mount the data on websites available to the general public within 6 to 12 months of publication at no cost to the NIH. Virtually all the societies publishing these papers are nonprofit organizations. Accordingly, NIH funds can be directed to the creative process (funding scientific research) rather than the publication process. It should be noted that fully two-thirds of all ATS publications (including basic science papers) are funded through private subscriptions by physicians/scientists, and industry (not paid for by the taxpayer), advertising, and society dues. NIH grants do not allow for purchase of journals or society memberships. Hence, there is a huge hidden cost in the movement of publication to PubMed, which would, at best, itself be a duplicative process for which most of the costs are already covered by nongovernment, nontaxpayer sources.
- 3. Suppostion: The peer review process will survive immediate publication on PubMed. The review process requires editors, numerous staff persons, computer specialists, office space, and, finally, societies to identify reviewers and classify the documents. Publication costs (borne mostly through nongovernmental subscriptions) support these societies AND their journals. Without the societies, the vetting process is not possible. If all information is accessed from PubMed at no cost, there is no infrastructure to support the peer review process nor to facilitate annual meetings at which the essential discussion and rediscussion of the new science occurs. The minimal delay in publication that PubMed avoids (to the clear benefit of no one) would serve only to undermine the existence of the very organizations that now are supported by nongovernmental funds and which insure the quality of the work. PubMed without peer review is a dangerous process. The potential for bad information or self-serving information related to product promotion veiled as basic science is indeed a threat.
- 4. Supposition: Immediate publication by PubMed will not destabilize the system. Much of this has already been discussed, but a true danger of the attrition of scientific societies, peer review prioritizing, abolition of venues for scientific discussion, and eventual unilateral publication by the federal government is a potential unintended consequence worthy of some policy analysis. The NIH grant review process has been designed to rely on peer review, so that worthy ideas can be funded on their merit. The risk of losing the independently organized peer review process for publication is that governmental policy on social issues may become perilously linked to scientific publication. Publication of science must remain apolitical.
The issues raised here are not unique to our society. The challenge of the information age must be met at each stage of the development of information technology. Nonetheless, it is important to frame issues that arise from these new technologies both in terms of their current reality and the downstream consequences of their implementation. At present, there is no identified beneficiary to the implementation of an immediate publication system by the federal government, and the underlying policy consequences of undermining the entire scientific communications and information exchange system—with no benefit to the public—is a genuine concern.
 |
FOOTNOTES
|
|---|
Conflict of Interest Statement: A.L. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; S.D.S. has participated in advisory boards for Boehringer Ingelheim, GlaxoSmithKline, Millennium, Pfizer, Wyeth, and ICOS, and his laboratory has performed research in collaboration with Pfizer, Arriva, ONO, and Taisho, for which no personal income was obtained; and E.A. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript.
