Abstract
Since its introduction in 1987, Benjamin Freedman’s principle of clinical equipoise has enjoyed widespread uptake in bioethics discourse. Recent years, however, have witnessed a growing consensus that the principle is fundamentally flawed. One of the most vocal critics has undoubtedly been Franklin Miller. In a 2008 paper, Steven Joffe and Miller build on this critical work, offering a new conception of clinical research ethics based on science, taking what they call a “scientific orientation” toward the ethics of clinical research. Though there is much to recommend Joffe and Miller’s scientifically oriented conception of clinical research ethics, I believe that both the critical and constructive projects suffer from the same basic mistake: inattention to context. The internal norms of science cannot be fully specified, let alone satisfied, independently of contextual (external) factors that only come into view when we are attentive to the particular context of that form of inquiry.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
At the outset of their article, the authors state their intention to offer a “full description of the normative structure of clinical research ethics” (emphasis mine). Later, in the course of articulating the internal norms appropriate to biomedical research, they begin by listing a set of (internal) norms that are clearly methodological or, broadly speaking, epistemological. As a result, it is unclear what the authors mean by “normative structure.” On the one hand, they may be exclusively concerned with elucidating the ethical norms appropriate to clinical research—in which case their list of methodological norms is out of place. On the other hand, they may be concerned with elucidating both the ethical and epistemological norms appropriate to clinical research, believing as I do that epistemological norms are partly constitutive of clinical research ethics. Given their purported intention to ground clinical research ethics in science, I am inclined to interpret their work in the latter sense and the following should be read with this in mind.
I hasten to point out that the objectivity of science does not depend on the independence of the contexts of discovery and justification. Rejecting their purported independence does not, in my view, commit one to a form of epistemic relativism.
In fact, Williams is concerned with knowledge more generally; that is, he is concerned with showing that knowledge is not a genuine theoretical kind, not just science. However, it is safe to say that Williams would extend the same argument to science itself. After all, he argues that the supposition that there are invariant epistemological constraints on justification misrepresents the way that justification and inquiry proceed both in common life and in science. For a similar discussion that is explicitly concerned with science, see Richard Rorty [14].
Of course, science is not a purely epistemic endeavour, but that is part of the point.
Given the interest-relative character of justification in Williams’s view, some readers will be tempted to suppose that this brand of contextualism amounts to a form of epistemic relativism. This would be a mistake. Williams offers an extended argument to this effect in “Why (Wittgensteinian) Contextualism is Not Relativism” [16].
I should point out that Ahrens’s analysis of these distinctions runs contrary to prominent contemporary formulations. Consider, for example, the three part definition of clinical research offered by the NIH Director’s Panel on Clinical Research in their 1997 report See [18, p. 1]. Prima facie, their definition is similar to Ahrens’s. However, the panel defines clinical research in terms of interaction with patients, and persists in contrasting clinical research with basic research in these terms. According to the panel, clinical research involves direct interaction with patients/human subjects (at the very least, clinical research is “research in which it is necessary to know the identity of the patients from whom the cells or tissues under study are derived”), whereas basic research does not. In this the panel is not alone. Many commentators define clinical research, and contrast it with basic research, in this way see, for example, [19–21]. Goldstein and Brown’s so called “handshake test” seems to be particularly popular: “[a]s a rule of thumb, if the investigator shakes hands with the patient in the course of the research, that scientist is performing POR [i.e., clinical research]” [19, p. 2806].
Of course, strictly speaking this category of clinical research is not restricted to studying the management of disease. Studies of management of disease are paradigmatic of this form of clinical research, but this category of research also concerns itself with managing well-being, conceived more broadly than disease, and with managing injury, disability, and the like, none of which can or should be categorized as diseases.
Note: “legitimate” modifies “rival,” not “evidence”.
Some might wonder where “preferred” came from here; others may take it to be obvious that APOR is concerned with the improvement of the management of disease and, thus, with “questions concerning the preferred modality of disease management.” In the interests of clarity, I will be explicit: I am reading improvement into the connotation of “new” as that term is used in Ahrens’s definition of APOR. It seems to me that, insofar as we are concerned with the management of disease, we are (and should be) concerned with hypotheses about interventions that promise to improve the management of disease, rather than hypotheses about interventions that are merely novel. Analogously, it is widely recognized that human subjects research is morally justified only if it promises to yield valuable results, i.e., results that hold out real promise for the improvement of clinical practice. From this perspective as well, newness is not sufficient for moral justification, unless “new” implies improved, as in “new and improved!”.
I realize that this view of APOR differs from, for example, the view of APOR implicitly condoned by the FDA. The FDA does not require that novel interventions be compared against standard treatment (unless the standard is nothing). As numerous commentators have pointed out, however, comparisons of second generation interventions against placebo are, except in exceptional circumstances, epistemically irrelevant from the clinician’s perspective (i.e., if we are interested in the management of disease). Such trials, it seems safe to say, serve different interests (e.g., the interest in approving new interventions regardless of their merits relative to existing interventions). According to my view, such studies are not bona fide APOR.
I am assuming it will be readily conceded that there is such a thing as an accepted treatment.
Of course, it is very difficult to say what “all the evidence” means. Here I use the phrase to suggest something like “all the relevant evidence” or “evidence that is convincing to the community (and is in fact reliable).” By using this phrase I do not mean to suggest that justified belief in the veracity of a hypothesis requires certainty.
Obviously, “clinically relevant knowledge” needs further unpacking. This is an important project that requires addressing. Notice, however, that even as it stands, “clinically relevant knowledge” cannot be equivalent with “generalizeable knowledge” simpliciter. Though a trial comparing a second-generation intervention against placebo may well produce generalizeable knowledge simpliciter, it will not typically produce clinically relevant (generalizeable) knowledge.
Insofar as APOR is defined in terms of the improvement of the management of disease, a hypothesis counts as a bona fide hypothesis in APOR only if it involves the evaluation of an intervention that promises to improve the management of disease. However, this does not mean that all trials of second-generation interventions must explicitly involve comparison with the accepted first-generation treatment in order to count as APOR, though this is ideal; it does mean that all trials of second-generation interventions must involve, at the very least, an implicit comparison with the accepted first-generation treatment. Of course, the validity of implicit comparisons is suspect.
Some might wonder at this argument given that Ahrens seems to have managed to provide a definition of APOR, a definition that I have relied on here, without referring to the context of practice. Notice, however, that Ahrens’s definition refers to “incompletely tested new diagnostic or therapeutic technique[s]” (emphasis mine). Furthermore, he explicitly states that the latter are often tested by “comparison with an accepted one.” To my mind, both “new” and “comparison with an accepted one” refer to the context of practice, implicitly in the former instance, explicitly in the latter.
Some authors draw a sharp distinction between the criteria for justified belief and the criteria for reasonable action, arguing that only the latter are morally relevant in the clinical context [22]. This position seems both implausible and irresponsible to me. Given a commitment to evidence based practice, the criteria for reasonable action and justified belief cannot and should not come apart. Of course, this does not necessarily mean that individual physicians will recommend the same treatment to the same patient given the available evidence. It does mean, however, that recommendations based on unjustified beliefs are unreasonable.
A recent article by Emily L Evans and Alex John London is an exception [23]. In this article, the authors make a similar point, arguing, “[clinical] equipoise can be used…as a model that the research community can use in determining how to advance the state of knowledge so as to improve the future standard of care” [23, p. 446]. Notice that, on Evans and London’s view, equipoise is playing an epistemic role here. My view is very similar except that, whereas Evans and London state that the research community can use equipoise in this way, I insist that they must (if they propose to conduct bona fide APOR).
To be clear, clinical equipoise is not a sufficient condition for the moral permissibility of trial enrolment but a necessary condition. Numerous other conditions must also be met (e.g., informed consent).
To be clear, I am assuming that “professional disagreement” implies that the beliefs of both parties involved in the disagreement are justified on the basis of good (i.e., reliable) but incomplete evidence. Notice, furthermore, that from an epistemological (and, I would argue, ethical) point of view, this modified version of the principle is preferable precisely because it does not turn on the actual beliefs of the community of expert practitioners. Rather, it turns on the evidence which, given a commitment to evidence based practice, it surely should.
Again, I hasten to emphasize that clinical equipoise is not a sufficient condition for the moral permissibility of enrolment. Furthermore, the ethical component of clinical equipoise is also partly constitutive of APOR (not merely regulative): the epistemological and the ethical are two sides of the same constitutive principle (i.e., the principle of clinical equipoise). After all, the epistemic conflict characteristic of a state of clinical equipoise is normatively salient only given a prior ethical commitment to evidence based practice (i.e., a particular view of the relationship between clinical research and clinical practice). Much more needs to be said about this in future work.
References
Freedman, Benjamin. 1987. Equipoise and the ethics of clinical research. The New England Journal of Medicine 317 (3): 141–145.
Temple, Robert, and Susan S. Ellenberg. 2000. Placebo-controlled trials and active-control trials in the evaluation of new treatments, part I: Ethical and scientific issues. Annals of Internal Medicine 133 (6): 263–455.
The National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. 1978. The Belmont Report. Bethesda, MD: US Department of Health, Education, and Welfare.
Joffe, Steven, and Franklin G. Miller. 2008. Bench to bedside: Mapping the moral terrain of clinical research. The Hastings Center Report 38 (2): 30–42.
Miller, Franklin G., and Howard Brody. 2003. A critique of clinical equipoise: Therapeutic misconception in the ethics of clinical trials. The Hastings Center Report 33 (3): 19–28.
Anderson, James A. 2007. Clinical research in context: The ethics and epistemology of clinical science. PhD diss., Dalhousie University.
Bernard, Claude. 1927. An introduction to the study of experimental medicine. New York: MacMillan.
Giere, Ronald N. 1997. Understanding scientific reasoning, 4th ed. Harcourt, Brace, Jovanovich: Fort Worth.
Jansen, Lynn A. 2008. Doctor vs. scientist? The Hastings Center Report 38 (2): 3.
Reichenbach, Hans. 1938. Experience and prediction: An analysis of the foundations and the structure of knowledge. Chicago, IL: University of Chicago Press.
Longino, Helen. 1990. Science as social knowledge: Values and objectivity in scientific inquiry. Princeton, NJ: Princeton University Press.
Campbell, Richmond. 1998. Illusions of paradox: A feminist epistemology naturalized. Lanham, MD: Rowman and Littlefield.
Williams, Michael. 1996. Unnatural doubts: Epistemological realism and the basis of scepticism. Princeton, NJ: Princeton University Press.
Rorty, Richard. 1988. Is natural science a natural kind? Construction and constraint. Notre Dame, Indiana: University of Notre Dame Press.
Williams, Michael. 2001. Problems of knowledge: A critical introduction to epistemology. New York: Oxford University Press.
Williams, Michael. 2007. Why (Wittgensteinian) contextualism is not relativism. Episteme: A Journal of Social Epistemology 4 (1): 93–114.
Ahrens, Edward H. 1992. The crisis in clinical research: Overcoming institutional obstacles. New York: Oxford University Press.
NIH Director’s Panel on Clinical Research. 1997. Executive summary: NIH director’s panel on clinical research. Bethesda, MD: US Department of Health and Human Services.
Goldstein, Joseph L., and Michael S. Brown. 1997. The clinical investigator: Bewitched, bothered, and bewildered—But still beloved. Journal of Clinical Investigation 99 (12): 2803–2812.
Schechter, Alan N. 1998. The crisis in clinical research: Endangering the half-century National Institutes of Health consensus. JAMA 280 (16): 1440–1442.
Rees, Jonathan. 2004. The fundamentals of clinical discovery. Perspectives in Biology and Medicine 47 (4): 597–607.
Hellman, Deborah. 2002. Evidence, belief, and action: The failure of equipoise to resolve the ethical tension in the randomized clinical trial. Journal of Law, Medicine and Ethics 30: 375–380.
Evans, Emily L., and Alex J. London. 2006. Equipoise and the criteria for reasonable action. Journal of Law, Medicine and Ethics 34 (2): 441–450.
Searle, John. 1969. Speech acts: An essay in the philosophy of language. London: Cambridge University Press.
Rawls, John. 1955. Two Concepts of Rules. The Philosophical Review 64 (1): 3–32.
Wittgenstein, Ludwig. 1969. On certainty. Ed. G.E.M. Anscombe and G.H. von Wright (trans: Paul, Denis and Anscombe, G.E.M.). New York: Harper.
Miller, Paul, and Charles Weijer. 2003. Rehabilitating equipoise. Kennedy Institute of Ethics Journal 13 (2): 93–118.
Acknowledgements
This paper was completed with the support of a grant from the Canadian Institutes of Health Research: “Advancing the Ethics of Clinical Trials: Enhancing Participant Protection and Scientific Rigour.” I would also like to thank Susan Sherwin, Richmond Campbell, Michael Hymers and Gordon McOuat, as well as the members of the Clinical Trials Research Group at McGill University, for their helpful comments on earlier drafts of this paper.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Anderson, J.A. Contextualizing clinical research: the epistemological role of clinical equipoise. Theor Med Bioeth 30, 269–288 (2009). https://doi.org/10.1007/s11017-009-9104-6
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11017-009-9104-6