Trial Registration

The design of all studies should be publicly available

Declaration of Helsinki. Principle 16


Biomedical research is a public endeavor, intended primarily to produce scientific knowledge to be used to promote health and well-being. The cost of research is funded, directly or indirectly, by the public. The public's health is best promoted when research is conducted in a manner that permits continuing scrutiny and appraisal by lay and expert citizens.

Calls for trial registration have been made by journal editors (1), funding agencies (2), and others (3,4) seeking to improve the scientific and social value of clinical trials. Registers diminish publication bias by exposing under-reporting of trial results; prevent duplication; promote collaboration; and directly serve physicians and patients by making information about available trials readily accessible. A comprehensive summary of the reasons why trials should be registered has been published by Tonks (5), and is adapted and reproduced in the table below.


Why register trials?
To mitigate against publication bias - the underreporting of trials with disappointing, negative, or inconclusive results - which misleads researchers conducting systematic reviews and doctors making decisions based on published evidence
To prevent unnecessary duplication of research effort, while encouraging appropriate replication and confirmation of results
To alert researchers to gaps in the knowledge base
To foster international collaboration among researchers and stimulate recruitment to clinical trials, enhancing their chances of success
To provide reliable intelligence about ongoing trials that will help funding bodies target their money where it is most needed
To aid recruitment to trials by direct appeal to the public
To provide a searchable database of current research efforts
To improve accessibility and therefore credibility of research performed by the pharmaceutical industry
To satisfy public demand for unbiased evidence on the effectiveness of treatments, and to promote the public accountability of medical research in general


Journals are increasingly likely to require evidence of trial registration before a manuscript reporting trial results is considered eligible for publication. The Lancet has instituted a program (6) whereby it will review trial protocols before the inception of research in anticipation of eventual publication of the results in the journal. This consideration is contingent on trial registration.

Many trial registers exist. Some register trials in all areas of medicine, while others are restricted to speciality areas. To promote ease of access and efficiency in searching for relevant trials, cooperation between registers and centralization is necessary. Electronic links between registers, trial-related Web sites and online Systematic Reviews is necessary. A searchable meta-register of trials may provide essential details concerning a trial, with a link to a trial-related Web site wherein additional information of relevance to researchers and research ethics committees is provided.

At present, trial registration is generally voluntary. In the U.S., the Food and Drug Modernization Act of 1997 (7) mandates the submission of information to a Clinical Trials Database for drug studies involving "serious and life-threatening diseases," and the F.D.A. recently issued draft guidance on this requirement (8). In addition, the F.D.A. has proposed the mandatory submission of clinical trial information for research involving gene transfer and xenotransplantation (9). This is an important, but limited step, in the right direction. Research ethics committees should require evidence of trial registration for all controlled clinical trials, irrespective of the nature of the research, in service of the requirement for scientific and social value.

International Standard Randomised Controlled Trial Number

The concept of trial registration will work best with a centralized, coordinated Internet-based registration system. A large number of trial registers already exist, generally organized around specific disease entities. Such dispersal is best avoided in favor of a centralized database of trials, the latter facilitating efficient and reliable electronic searching by interested parties.

To facilitate the recognition of trial registration, each trial should acquire an International Standard Randomised Controlled Trial Number (ISRCTN). This number is assigned to trials registered at Current Controlled Trials, which maintains a searchable meta-register of trials. Each registered trial may be associated with an external link to a specialized trial register or a Web page wherein additional information about the trial is provided. The number will greatly facilitate tracking of clinical trials and associated reports worldwide, and may be electronically linked to repositories of Systematic Reviews concerning the clinical issue under investigation. Research Ethics Committees should support this effort to establish a centralized registration system given the international scope of performance of clinical trials.

References

References available online are hyperlinked

1. Horton R, Smith R. Time to register randomized trials. BMJ 1999; 319:865-866.

2. Medical Research Council (U.K.)

3. Chalmers I. Current Controlled trials: an opportunity to help improve the quality of clinical research. Curr Control Trials Cardiovasc Med 2000; 1:3-8.

4. Blunt J. Research ethics committees must ensure registration of research and accessibility of results BMJ 1998; 316:312

5. Tonks A. Registering clinical trials. BMJ 1999; 319:1565-1568.

6. McNamee, D. Review of clinical protocols at the Lancet. Lancet 2001; 357:1819-1822

7. FDA Modernization Act of 1997. CDER-Related Documents.

8. Guidance for Industry. Information Program on Clinical Trials for Serious or Life-Threatening Diseases: Establishment of a Data Bank. 2001.

9. Human gene therapy or xenotransplantation; data and information disclosure. Federal Register. January 18, 2001.