Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
Ethics committees must require trial registration
- Andrew Jull (4 July 2005)
-
Publication bias, selective dissemination and trial registries
- James P. McCormack, Peter S. Loewen, Peter J. Jewesson (15 July 2005)
|
|
|||
|
Andrew Jull, Research Fellow Clinical Trials Research Unit, School of Population Health, University of Auckland, New Zealand
Send response to journal:
|
Sir The paper by Decullier and colleagues provides fresh impetus for the imperative that all clinical trials should be registered. Although 38% of studies overall have been published in the 10 years since approval was sought, only 26% (80/303) of the randomised controlled trials presented for ethical approval have been published (excluding those still ongoing or that failed to start, table 2, p21). Efforts to base treatment decisions on summaries of evidence are grounded in the completeness of the available evidence. If the failure to publish trial evidence observed by Decullier et al is widely replicated then many summaries and thus treatment decisions may be suspect. In addition to the International Committee of Medical Editors requiring prospective public domain registration of trials, ethics committees world-wide should require evidence of trial registration before granting ethical approval. The net must tighten so there is little chance of evidence not being employed to support clinical decision-making. Competing interests: None declared |
|||
|
|
|||
|
James P. McCormack, Professor University of British Columbia 2146 East Mall Vancouver BC Canada V6T 1W5, Peter S. Loewen, Peter J. Jewesson
Send response to journal:
|
Dear Editors: We applaud Decullier et al for their insightful evaluation of some factors influencing publication of health-related research projects. (1) In contrast to the approach taken by others who have used national registries (2,3) or sponsoring agency records, (4) these authors have examined research activity outcomes from the perspective of research ethics approval. Clinical trial registration is not universally required and sponsor databases are necessarily limited in scope. However, research committee approval is required of all research involving human subjects and includes protocols regardless of funding status and origin. We strongly agree with this approach and feel that research ethics approval represents the earliest convergence in the birth of human-subjects research projects, making them an ideal perspective from which to study the subsequent events in their life cycle. Failure to disseminate results is considered research misconduct and is an urgent scientific and ethical concern. (5,6) As confirmed by the authors, selective dissemination of research results in publication bias, typically skewing the literature toward reports with positive findings. However, in their discussion, the authors appear to conflate the issue of clinical trial registration with that of selective dissemination, implying that registration will remedy this situation. While the registration models promoted by the International Council of Medical Journal Editors, the Canadian Institutes of Health Research, European regulatory agencies and others increase transparency, they are not designed to ensure registered protocols result in publication; instead they are designed to ensure publications are only spawned from registered protocols. Preliminary results from a study we are undertaking at a major Canadian medical-doctoral university indicate that while meticulous financial records of research activity are kept, mechanisms for identifying and tracking the dissemination status of human-subjects research projects are virtually nonexistent. We hope to build on the work of Decullier et al. by characterizing the spectrum of dissemination (e.g. from local presentation to global distribution) that is relevant to clinicians in contemporary practice, to identify trends over time in modes of dissemination (e.g. oral, written and electronic), and to clarify what mechanisms are required to increase rates of dissemination as a means of enhancing the value, integrity, and the public’s trust in clinical research. James McCormack, PharmD, ACPR
Peter Loewen, PharmD, ACPR, FCSHP
Peter Jewesson, PhD, ACPR, FCSHP
Division of Clinical Pharmacy, Faculty of Pharmaceutical Sciences, University of British Columbia, Vancouver, BC, Canada References 1. Decullier E, Lheritier V, Chapuis F. Fate of biomedical research protocols and publication bias in France: retrospective cohort study. Bmj 2005; Jul 2;331(7507):19. 2. Pich J, Carne X, Arnaiz JA, Gomez B, Trilla A, Rodes J. Role of a research ethics committee in follow-up and publication of results. Lancet 2003; Mar 22;361(9362):1015-6. 3. Bardy AH. Bias in reporting clinical trials. Br J Clin Pharmacol 1998; Aug;46 (2):147-50. 4. Ioannidis JP. Effect of the statistical significance of results on the time to completion and publication of randomized efficacy trials. Jama 1998; Jan 28;279(4):281-6. 5. Chalmers I. Underreporting research is scientific misconduct. Jama 1990; Mar 9;263(10):1405-. What is research misconduct? London: BMJ Publishing Group; 2000. 6. Tri-Council Policy Statement: Ethical Conduct For Research Involving Humans. Health Canada Panel on Research Ethics; 1998 (with 2000, 2002 updates). Competing interests: None declared |
|||