Just Say Yes to Drug Trial Information

The road to increased access to medical research might not be easy, but it is the route we must take.

Several different research-access stories were in the news in September. The National Institutes of Health (NIH) issued a proposed rule that would require that all scholarly papers based on NIH-funded research be made available for free on a government Web site six months after they are published. The Pharmaceutical Research and Manufacturing Association (PhRMA) announced that it would establish a free Web site where drug companies could voluntarily publish summaries of the results of drug trials. On September 8, a group of editors of major medical journals issued a statement in which they said that they will accept articles on clinical trials only if information about the trials is announced publicly before they begin. All these events are related to the fundamental issue of how much free information is the public entitled to, but they differ significantly in the particulars.

In the first case, NIH is paying for the research with tax dollars, and NIH director Elias Zerhouni is responding to pressure from patient advocacy groups and others, who argue that because they paid for the research, they deserve to see the results. Resistance to the NIH proposal comes from the scholarly publishing industry, which earns its money by selling subscriptions to the journals that publish the research papers. They claim that the articles are their intellectual property and that NIH has no right to give it away. They maintain that the entire medical publishing enterprise could come crashing down if the need to subscribe to the journals is eliminated.

In the case of PhRMA, the companies fund the clinical trials, and they feel that it is up to them whether or not to share the results. To the surprise of no one, they eagerly publish papers in scholarly journals when the results cast a favorable light on their products and are decidedly more circumspect about research that finds that the product is ineffective or inferior to a competitor’s product. Because posting results on the proposed Web site is completely voluntary, we can expect companies to remain selective about the information they make available. The pharmaceutical industry did not become the most profitable in the United States by publicizing its failures.

The goal of the editors is to put pressure on the health care industry to release results of all trials, not just those that they find favorable for their products. They also believe that it is in the public interest for doctors and patients to know what trials are being conducted and when results can be expected.

The NIH case is the trickiest. Scholarly journals are an essential component of the research enterprise. They review articles to ensure that they are worthy of publication, they edit manuscripts carefully to ensure accuracy, they publish corrections when necessary after publication. Many of these journals are published by private companies that earn a profit for their efforts, but many are published by nonprofit professional societies, which use the revenue to support efforts they consider beneficial to the scientific discipline or profession. By and large, these societies perform a valuable function in promoting the value of science, educating the public, and attracting students to study science. Neither the for-profit nor the nonprofit publishers are doing anything objectionable, and no one would argue that they do not perform a valuable service.

The problem arises when we focus on the narrow goal of protecting the way the service is currently being provided rather than the more important goal of sharing information in a timely and cost-effective way. The goal is to communicate research results to other researchers and to the public, not to maintain the profitability of scholarly publishers or to support the work of professional societies. Research communication should not be held hostage to these secondary goals. Although they will not like the analogy, the publishers are like the fishers discussed in the two articles on ocean policy in this issue. Fishers perform a vital function by catching fish and making them available to consumers. But policymakers have to give priority to the health of the oceans and the general public interest. Their goal is not to guarantee that a certain group of people can earn a living from fishing. Likewise, policymakers have no reason to ensure that a particular group of people, companies, and organizations be allowed to control scholarly publishing.

The goal is to move toward a socially optimal way of providing fish to consumers and research results to the public. That might mean changes to these industries. That’s life. Of course, when government action changes the economic landscape, it has an obligation to pay attention to those hurt and to ease their transition.

The American Medical Publishers Association wrote to Zerhouni on August 23 in anticipation of the new NIH policy to protest NIH’s failure to solicit sufficient advice from the publishing industry. NIH has actually sought input from medical publishers, and there is a 60-day public comment period before the new guidelines are implemented. To be effective and to maintain quality, the proposed depository of articles must be developed in consultation with the publishing industry. In addition, the industry has the right to squeal if government policy hurts them financially. But there should be no doubt that the goal is to move forward, not to preserve the status quo. The more rapid the change, the greater will be the impact on the industry. But this is not a reason to slow change. It could be a reason to provide recompense to the publishers.

As the Council of the National Academy of Sciences noted in its statement supporting the NIH policy, it is important that the government Web site provide the published version of the articles, not the original submissions from the researchers: “This will ensure that only one version of a paper is extant, and that the public has access to the version of record.” The Council notes that whatever system for publishing research develops, journal publishers should have enough income to provide the vital services of peer review and archiving. The Council also urges NIH to take its responsibility seriously by formally pledging to maintain the depository in perpetuity and to accept all papers published in qualified, refereed biomedical journals.

The purpose of the PhRMA announcement is to appear to be responding to the demand for more information. Posting results from all the companies on one Web site will make it easier to find information, but the public also has an interest in seeing that all results are posted. Reps. Henry Waxman (D-Calif.) and Edward Markey (D-Mass.) have already said that they will introduce legislation to require that information about all trials be placed on a government Web site. At a September 9 hearing of the House Energy and Commerce Subcommittee on Oversight and Investigation, committee members asked Food and Drug Administration representatives why they did not make public all the information that companies submitted to them in their applications for drug approval. The FDA officials responded that they were legally prevented from doing so, but the committee asked for more information about the rules. It might be that FDA already has the authority to make more information public. If not, Congress should continue to work to make this information easily available.

The momentum for increasing open access to research information is clearly gaining strength, and justifiably so. Policymakers need to be alert to the effects that their policies will have on stakeholders such as publishers and the health care industry, but they should keep their eyes on the prize of increased access to information. They should do what they can to ease the transition for those hurt by change, but they should not let the narrow interests of a few stand in the way of the larger public interest.

Cite this Article

Finneran, Kevin. “Just Say Yes to Drug Trial Information.” Issues in Science and Technology 21, no. 1 (Fall 2004).

Vol. XXI, No. 1, Fall 2004