The placebo effect [response]
Online posting: Sept. 17, 1997
Published in print: Oct. 15, 1997 (CMAJ 1997;157:1019)
Re: Bioethics for clinicians: 10. Research ethics, by Dr.
Charles Weijer and associates, CMAJ
1997;157:1153-7 [full text / résumé]
In response to: S.M.H. Alibhai
Dr. Alibhai asks a good question: When are placebo controls
ethically acceptable and scientifically necessary in clinical
research? At the start of a clinical trial, a state of clinical
equipoise must exist; that is, there must be genuine disagreement
among expert practitioners as to the preferred treatment.[1] A
nonvalidated treatment may be compared to a placebo control if
(1) no standard therapy exists, (2) standard therapy exists but
it has been shown to be no better than placebo, (3) standard
therapy is placebo, (4) standard therapy is toxic and of marginal
benefit, or (5) validated treatment exists but is not available
because of cost or limited supplies.[2] Two special circumstances
deserve to be made explicit. Placebo controls are appropriately
used when the new, nonvalidated treatment is an "add-on" to
standard therapy (so that the comparison is standard therapy plus
new drug versus standard therapy plus placebo) or when the study
population is restricted to persons whose disease has not
responded to standard treatment and for whom no second-line
standard treatment exists.[3]
The ethical issue is no different for phase 1 and phase II
clinical trials. In my experience, the use of placebo controls is
unusual in such trials. These studies enrol small numbers of
patients and thus comparison with a control, placebo or
otherwise, is unlikely to have sufficient statistical power.
Phase I trials evaluate the toxicity and pharmacodynamics of new
drugs and most often involve healthy subjects. In such cases, the
risks to the subjects are justified by the importance of the
knowledge likely to be gained, even though no therapeutic benefit
is expected.[4] Because of the toxicity of anticancer agents,
phase 1 cancer studies typically involve patients for whom no
effective therapy exists. In either case, and assuming the design
of the study required the use of a control, a placebo control
would be ethically unproblematic. Phase II trials seek to
establish whether a nonvalidated treatment is efficacious for a
particular condition, so actual patients are enrolled. The use of
a placebo control (again, assuming that the study design
necessitates a control) would have to meet the conditions laid
out in the first paragraph.
All too often clinical trials performed as part of the drug-
approval process in Canada subject research participants to
placebos despite the existence of effective treatment. Not only
is this practice unethical, but it exposes researchers and
institutions to legal liability.[5] Researchers, research ethics
boards, pharmaceutical companies and the Health Protection Branch
of Health Canada must work together to ensure that new-drug
research meets the highest ethical and scientific standards.
Charles Weijer, MD, PhD
Bioethicist
Mount Sinai Hospital and
University of Toronto Joint Centre for Bioethics
Assistant Professor
Department of Medicine
University of Toronto
Associate Member
Samuel Lunenfeld Research Institute
Toronto, Ont.
cweijer@mtsinai.on.ca
References
- Freedman B. Equipoise and the ethics of clinical research.
N Engl J Med 1987;317:141-5.
- Freedman B. Placebo-controlled trials and the logic of
clinical purpose. IRB: Rev Hum Subj Res 1990;12(6):1-6.
- Freedman B, Weijer C, Glass KC. Placebo orthodoxy in clinical
research. I: Empirical and methodological myths. J Law Med
Ethics 1996;24:243-51.
- Freedman B, Fuks A, Weijer C. Demarcating research and
treatment: a systematic analysis for the analysis of the ethics
of research. Clin Res 1992;40:653-60.
- Freedman B, Glass KC, Weijer C. Placebo orthodoxy in clinical
research. II: Ethical, legal and regulatory myths. J Law Med
Ethics 1996;24:252-9.
 CMAJ email
CMAJ Readers' Forum
CMAJ home page
|
