6
Conclusion and Next Steps
The Institute of Medicine (IOM) brought together many diverse entities to express their views on the content, purposes, and implementation of a clinical trial registry. A summary of these views has been attempted here, and there have been clear signs of convergence and momentum since the discussions that were convened by the IOM began.
Most of the possible content fields for a clinical trial registry are not being debated. The discussions at the workshop centered on the following five concepts:
-
Purpose. There was little or no controversy over whether a registry should track the status of individual clinical trials and provide a complete record of all trials to aid in doing systematic reviews of the evidence. There appeared not to be agreement on whether the same registry could or should be used for patient recruitment, whether the registry should include the results of the trials, and if it did include the results, when those results should be posted.
-
Which Trials to Include. There was little controversy over the desirability of expanding the information currently collected by ClinicalTrials.gov beyond serious or life-threatening diseases and conditions. There was no agreement, however, over the inclusion of exploratory trials in a registry. The line between exploratory and confirmatory trials is not a bright one, and although some participants thought these trials should be included for the sake of completeness, others did not, citing proprietary interest and the lack of applicability to policy and health care decisions. More work is needed to come to agreement on this.
-
Delayed Disclosure Mechanism. For proprietary reasons, speakers from industry argued for the delayed disclosure of certain fields (i.e., at project initiation, not disclosing the hypothesis statement, primary and secondary outcome measures, and projected year of trial completion) for a minority of trials. However, the workshop did not discuss implementation questions such as what is meant by “delay” (in terms of time or staging in the marketing process) or whether there should be objective criteria for when industry would invoke such a delay. More work is needed to reach agreement on whether delayed disclosure is an acceptable approach and, if so, to define processes for authorizing these delays. An option that might accompany delayed disclosure is to store the data in a non-public part of the registry.
-
Reporting Results of Completed Trials. The International Committee of Medical Journal Editors (ICMJE) does not call for results to be reported. The World Health Organization (WHO) calls for the reporting of results once a trial is completed. The Joint Position of the pharmaceutical industry calls for reporting of results for products that are brought to market in any country. As the interested parties continue to discuss and seek a resolution to the issue of reporting results, the other purposes of a clinical trial registry can be met.
-
Compliance. Incentives for complying and consequences for not complying with registry requirements would seem necessary for a registry to be comprehensive. Possible roles for Institutional Review Boards, the Food and Drug Administration, and others were discussed, but this is still a controversial topic.
Activities toward a comprehensive, publicly accessible clinical trial registry continued after the workshop. Depending on when a clinical trial started, it has to be registered before enrolling the first patient or (for trials initiated before July 1, 2005) before submitting an article to an ICMJE journal. The international pharmaceutical industry working with the National Library of Medicine (NLM) has begun registering all but exploratory clinical trials on ClinicalTrials.gov, and providing the results of clinical trials on marketed products on ClinicalStudyResults.gov, an Internet database launched in September 2004 by the Pharmaceutical Research and Manufacturers of America. These new efforts are being monitored closely on the pages of the medical journals. Zarin and col-
leagues at NLM (2005) noted that, between May and October 2005, there was a large increase (73 percent) in the number of clinical trial registrations on ClinicalTrials.gov, an increase that staff attribute to enactment of the ICMJE policy. They note, however, that completeness of registration information varies, indicating differing levels of discomfort with full disclosure, as expressed at the IOM workshop. In an accompanying editorial, Drazen and Wood (2005) note that a handful of commercial entities are still using meaningless entries for the Intervention Name field and for the Primary Outcome field. They point to the fact that several other pharmaceutical firms are in full compliance, “undercutting any argument that this failure reflects a commercial imperative.” Finally, several of the individuals involved in this IOM workshop have moved into advisory roles on the WHO project. WHO is in the process of establishing norms and standards for international clinical trial registration.