Paul S. Mueller, MD, MPH; Victor M. Montori, MD, MSc; Dirk Bassler, MD, MSc; Barbara A. Koenig, PhD; Gordon H. Guyatt, MD, MSc
Potential Financial Conflicts of Interest: Consultancies: B.A. Koenig (Affymetrix Corporation).
Requests for Single Reprints: Paul S. Mueller, MD, MPH, Program in Professionalism and Bioethics, Department of Medicine, Mayo Clinic College of Medicine, 200 First Street SW, Rochester, MN 55905; e-mail, firstname.lastname@example.org.
Current Author Addresses: Drs. Mueller, Montori, and Koenig: Mayo Clinic College of Medicine, 200 First Street SW, Rochester, MN 55905.
Dr. Bassler: University Children's Hospital, Department of Neonatology, Calwerstrasse 7, 72076 Tuebingen, Germany.
Dr. Guyatt: McMaster University, 1200 Main Street West, Room 2C12, Hamilton, L8N 3Z5 Ontario, Canada.
Stopping randomized trials early because of an apparent benefit is becoming more common. To protect and promote the interests of trial participants, investigators may feel obligated to stop a trial early because of the apparent benefit of a study treatment (compared with placebo or other treatment). There are, however, serious ethical problems with doing so. Truncated trials systematically overestimate treatment effects; in cases where the number of accrued outcome events is small, the overestimation may be very large. Generating seriously inflated estimates of treatment effect violates the ethical research requirement of scientific validity. Subsequent use of inflated estimates to inform clinical decision making and practice guidelines violates the ethical requirements of social value and a favorable risk–benefit ratio. Researchers should ensure that a large number of outcome events accrues before stopping a trial and then continue recruitment to assess whether positive trends continue. This can balance the need to protect research participants with the ethical requirements of scientific validity, social value, and a favorable risk–benefit ratio.
Table. Ethical Violations Resulting from Stopping a Trial Early for Apparent Benefit
Istanbul University, Cerrahpasa Medical Faculty, Istanbul, Turkey
June 24, 2007
Transparency in the informed consent process
We read with interest the opposing views of Drs. Mueller et al and Goodman (1, 2) about ending randomized control trials (RCT) "before term". We suggest there is yet another issue about terminating RCTs early, not discussed by either of our colleagues. We propose the information about what the patient will exactly know or knew about trial duration at the time he/she signed the related informed consent at the very start should be transparent to the reviewers of grants, journal submissions, to the readers if the trial report is accepted for publication and, in fact, to the general public. Such transparency, we reason, not only is simply fair to the trial patient, both seeking remedy and helping science as he/she enters a drug trial, but also is an essential starting point whenever we decide to discuss the related ethical and practical issues of a RCT in the optimal way. The transparency we propose also is also desirable when the issue is not early termination but, just the reverse, undue prolongation of RCTs. The interested reader might find the related debate (3, 4) about one such RCT explanatory. This was where we had brought up the point that we as readers, were not optimally informed about the actual intended duration of the RCT in question when it commenced and we had concerns about what the trial patients knew at the beginning and at various time points. It is important here to note that one potential scientific problem in both the extended and the early terminated RCT is an, what some might call an overzealous, attempt of proving one's hypothesis "“ that the drug understudy is working - and shying away from the time honored method of good science, falsification (5). We propose a simple solution to these concerns. Clinical trial registration was initiated in 2005 and has been welcome. We now suggest that effective from a certain point in time in the near future, all manuscripts about clinical trials submitted and published, perhaps even all studies needing informed consent, should submit their informed consents to these registries where the peer reviewers, interested readers, in fact interested anybody, can see what the initial informed consent form (or any amendments thereof) related to that trial actually reads. We envisage our proposal, should it find general support, will surely be a further step in the right direction for transparency which we all seek.
Hasan Yazici, MD 1
Yusuf Yazici, MD 2
1 Istanbul University Cerrahpasa Medical Faculty, Istanbul, Turkey
2 New York University, Hospital for Joint Diseases, New York, New York, USA
Corresponding author: Yusuf Yazici, MD 246 East 20th Street, New York, NY 10003 email@example.com
1- Mueller PS, Montori VM, Bassler D, Koenig BA, Guyatt GH. Ethical issues in stopping randomized trials early because of apparent benefit. Ann Intern Med 2007; 146: 878-81
2- Goodman SN. Stopping at nothing? Some dilemmas of data monitoring in clinical trials. Ann Intern Med 2007; 146: 882-7
3- Yazici Y, Yazici H. Trial of etanercept and methotrexate with radiographic and patient outcomes two-year clinical and radiographic results: Comment on the article by van der Heijde et al. Arthritis Rheum. 2006; 54: 3061-2
4- Van der Heijde D, Klareskog L. Reply to letter by Yazici and Yazici commenting on the two-year report on the trial of etanercept and methotrexate with radiographic and patient outcomes. Arthritis Rheum. 2007.;56:1031-2.
5- Yazici H. Use and Abuse of the Controlled Clinical Trial. Bull NYU Hosp Jt Dis 2007; 65:132-4
Mueller PS, Montori VM, Bassler D, et al. Ethical Issues in Stopping Randomized Trials Early Because of Apparent Benefit. Ann Intern Med. 2007;146:878–881. doi: https://doi.org/10.7326/0003-4819-146-12-200706190-00009
Download citation file:
Published: Ann Intern Med. 2007;146(12):878-881.
Results provided by:
Copyright © 2020 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use