Hostname: page-component-7c8c6479df-p566r Total loading time: 0 Render date: 2024-03-27T00:58:53.597Z Has data issue: false hasContentIssue false

Evidence, Belief, and Action: The Failure of Equipoise to Resolve the Ethical Tension in the Randomized Clinical Trial

Published online by Cambridge University Press:  01 January 2021

Extract

Clinical research employing the randomized clinical trial has, traditionally, been understood to pose an ethical dilemma. On the one hand, each patient ought to get the treatment that best meets her needs, as judged by the patient in consultation with her doctor. On the other hand, the method most helpful to advancing our understanding about what treatments are indeed best able to meet patient needs is the randomized trial, which necessitates that each patient's care is decided not by physician judgment or patient choice but instead by random assignment. The tension can be described as a conflict between the interests of individual patients who are sick today, and the interests of the group of people who will become sick in the future and would benefit from advances in medical understanding. How one ought to balance these important and often competing interests is an important ethical question that resists easy resolution.

Type
Article
Copyright
Copyright © American Society of Law, Medicine and Ethics 2002

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

World Medical Association, Declaration of Helsinki IV, 41st World Medical Assembly, Hong Kong, September, 1989. The fifth basic principle provides that “[c]oncern for the interests of the subject must always prevail over the interests of science and society.” See also Kass, L.R., Toward A More Natural Science: Biology And Human Affairs (New York: Free Press, 1985): at 196;.Google Scholar
Angell, M., “Patient Preferences in Randomized Clinical Trials,” N. Engl. J. Med., 310 (1984): 1385–87.Google Scholar
See, e.g., Freedman, B., “Equipoise and the Ethics of Clinical Research,” N. Engl. J. Med., 318 (1987): 141–55.Google Scholar
Baum, M. Zilkha, K. Houghton, J., “Ethics of Clinical Research: Lessons for the Future,” British Medical Journal, 299 (1989): 251–53.Google Scholar
See Hellman, S. Hellman, D., “Of Mice But Not Med: Problems of the Randomized Clinical Trial,” N. Engl.]. Med., 324 (1991): 1585–89; Hellman, S., “The Patient and the Public Good,” Nature Medicine, 1 (1995): 400–02.CrossRefGoogle Scholar
See Royall, R., Statistical Evidence: A Likelihood Paradigm (London: Chapman & Hall, 1997).Google Scholar
Lempert, R., “After the DNA Wars: Skirmishing with NRC II,” Jurimetrics, 439 (1997): 456–57.Google Scholar
“Frequentist” statistical method, which is the dominant method used in evaluating the results of clinical medical research, focuses on the frequency of outcomes. As the current method is used, the researcher calculates the probability that the observed result would occur due to chance alone if the null hypothesis (the hypothesis that there is no difference between the treatments being evaluated) were true.Google Scholar
Goodman, S.N., “Toward Evidence-Based Medical Statistics 1: The P Value Fallacy,” Annals of Internal Medicine, 130 (1999): 9951004.Google Scholar
The combination of the Fisher and Neyman/Pearson methods that is standard practice today is widely recognized as theoretically incompatible. See, e.g., Goodman, , supra note 9, at 997 (claiming that the Fisher and Neyman/Pearson methods “were incompatible but have become so intertwined that they are mistakenly regarded as part of a single, coherent approach to statistical inference”);.Google Scholar
Gill, J., “The Insignificance of Null Hypothesis Significance Testing,” Medical Research Quarterly, 52 1999): 647–74, at 649–50 (describing the “current null hypothesis significance test as a synthesis of two highly influential but incompatible schools of thought in modern statistics”).Google Scholar
Goodman, , supra note 9, at 998. Goodman explains that academic physicians routinely interpret a surprising study result with a p-value of 0.05 to mean that there is a 95 percent or greater chance that the null hypothesis is incorrect. As Goodman explains however, this is incorrect. As the p-value is calculated on the assumption that the null hypothesis is true, it “cannot, therefore, be a direct measure of the probability that the null hypothesis is false.” Goodman sees this misunderstanding as “reinforc[ing] the mistaken notion that the data alone can tell us the probability that a hypothesis is true” (emphasis added).Google Scholar
Id. at 996–97. Goodman explains well this familiar tradeoff between the virtues of inductive and deductive reasoning. He describes this as follows: “Under the assumption that two treatments are the same (that is, the hypothesis of no difference in efficacy is true), it is easy to calculate deductively the frequency of all possible outcomes that we could observe in a study. But once we observe a particular outcome, as in the result of a clinical trial, it is not easy to answer the more important inductive question, ‘How likely is it that the treatments are equivalent?’”.Google Scholar
Richard Royall makes precisely this point, see Royall, R., “Ethics and Statistics in Randomized Clinical Trial,” Statistical Science, 6 (1980): 5288, at 56;CrossRefGoogle Scholar
see also Hellman, S. Hellman, D., supra note 4.Google Scholar
Federal regulations allow the use of “promising new drugs” to treat patients with a “serious or immediately life-threatening disease” if “no comparable or satisfactory alternative drug or other therapy is available.” 21 C.F.R. §; 312.34.Google Scholar
Nelson, D. Weiss, R., “Penn Researchers Sued in Gene Therapy Death; Teen's Parents Also Name Ethicist as Defendant,” The Washington Post, Sept. 19, 2000,): at A3.Google Scholar
See Palca, J., “AIDS Drug Trials Enter New Age,” Science, 246 (1989): 1921; Emanual, E.J. Patterson, WB., “Ethics of Randomized Clinical Trials,” Journal of Clinical Oncology, 16 (1998): 365–71.Google Scholar
See, e.g., Gehan, E.A. Frei, E.J., “Non-randomized Controls in Cancer Clinical Trials,” N. Engl. J. Med., 198 (1974): 198203; Byar, D.P. et al. , “Design Considerations for AIDS Trials,” N. Engl. J. Med., 323 (1990): 1343–48;.Google Scholar
Concato, J. Shah, N. Horwitz, R.L., “Randomized, Controlled Trials, Observational Studies, and the Hierarchy of Research Designs,” N. Engl. J. Med., 342 (2000): 1887–92.CrossRefGoogle Scholar
See, e.g., Declaration of Helsinki, supra note 1.Google Scholar
In fact the National Institutes of Health funded trials of a shortened course of AZT which was given to pregnant women in developing countries in the attempt to reduce mother to infant transmission of HIV provoked controversy precisely because researchers defended giving some participants placebos on the grounds had the trials not taken place these women and their infants would receive no treatment. See Hellman, D., “Trials on Trial,” Report from the Institute for Philosophy and Public Policy, 18 (1998): 1318 (evaluating the ethical permissibility of placebo-controlled trials of the efficacy of short course AZT therapies for reducing mother to infant transmissions of HIV in developing countries).Google Scholar