PLACEBO

The terms placebo and placebo effect are quite difficult to define. Most commonsense definitions contain serious inconsistencies. For example, one commonly hears placebo defined as an "inert remedy"; but if a placebo were totally inert, there would be no point in giving it.

In Latin, placebo means "I shall please," but the effects of a placebo can be either positive or negative (the term nocebo, roughly meaning "I shall harm," is sometimes used to designate negative effects). Adolf Grünbaum emphasized that whether or not a remedy is a placebo is always relative to some biomedical theory. A sugar pill is a placebo for a migraine only because the biomedical theory agreed upon by all discussants denies any pharmacologic efficacy of small amounts of oral glucose in altering the pain of vascular headache.

Some find it useful to locate the species "placebo" under the genus "nonspecific therapy," by which they mean a therapy that strengthens the general resistance of the organism to disease of many sorts (as opposed to a therapy that removes the specific cause of a single disease or class of diseases). But the latter term may be as hard to define precisely as placebo is. Moreover, there may be an unspoken assumption that nonspecific therapies are synonymous with "therapies that operate through psychological rather than biological mechanisms." But this is clearly false; some psychological therapies may be very specific for certain diseases according to established psychiatric theories, and some biological therapies, notably diet and exercise, seem to be good candidates for "nonspecific" status.

For purposes of ethical analysis, placebo effect may be defined generally as the change in a patient's condition that results from the symbolic aspects of the encounter with a healer or with a healing setting, and not from the pharmacological or physiological properties of any remedy used. The term symbolic alludes not only to the psychological processes that occur within the patient but also to the social and cultural belief systems that form a background to the patient's thoughts and feelings and that give meaning to the healing process. A placebo, then, is a remedy administered either for purposes of eliciting the placebo effect or as a control in an experimental situation. Virtually any modality, including surgery and psychotherapy, can function as a placebo; the term is not confined to pills, capsules, or injections.

The practical goal of defining placebo effect as precisely as possible is to distinguish the changes it produces in the patient's condition from changes produced by other causes. In treatment, the two factors likely to be confused with placebo effects are the pharmacological or physiological effects of the therapy employed and the natural history of the illness. For example, if a patient with gastritis visits a physician, who recommends antacids, and the patient improves, the relief could have come from the pharmacological properties of the antacids, the natural tendency of gastritis to heal over time, the soothing symbolic effects of the physician consultation, or some combination of the three. The two-group design in a controlled experimental trial ("active" treatment versus placebo) allows the investigator to distinguish pharmacological or physiological effects from the placebo effects and the natural history of the illness. It does not allow a distinction to be made between natural history and placebo effects.

It is also helpful to distinguish a pure placebo, thought to have no pharmacological potency under any circumstances whatever, from an impure placebo, which has pharmacological potency under some circumstances. Common examples of impure placebos are vitamins administered to patients who have no documented deficiency and antibiotics administered to patients who have viral illnesses (which do not respond to antibiotics). In today's medical practice, impure placebos are probably used much more commonly than pure placebos.

Scientific Controversies

A number of works published in 2001 showed the controversy surrounding the science of the placebo effect. A careful meta-analysis of 114 randomized controlled trials concluded that the placebo effect does not exist in that context, and changes previously attributed to the placebo effect resulted from either natural history or random variation (Hróbjartsson and Gøtzsche). Other scientists reported further evidence that placebo effects in pain are mediated by endorphin release in the brain (Amanzio et al.) and that alteration in dopamine release in response to placebo therapy for Parkinson's disease can be detected by positron emission tomographic imaging of the brain (de la Fuente-Fernández et al.). The results of a conference on "The Science of the Placebo" sponsored by the U.S. National Institutes of Health (NIH) were published (Guess et al.), and the NIH announced that research programs would for the first time be devoted specifically to studying the mechanisms and extent of placebo effects. Readers were thus led to various conclusions: that the placebo effect is a myth; that scientists understand better how it works; and that further research into its mechanisms will be fruitful. The majority view appears to be that the "myth" dismissal is premature and that more study is needed.

At the biochemical and cellular level, placebo effects may induce organ changes via the release of catecholamines, endorphins, or immunoactive cells; all three have been shown to be very sensitive to a patient's psychological or emotional state. At the social and psychological level, one must identify aspects of the setting or of the human interaction that cause the patient to perceive the situation as a healing one, thereby releasing whatever biochemically active substances might be involved. It appears safe to claim that a positive change in the patient's health status is most likely to occur when at least three things happen: the patient receives a satisfying explanation of the illness and treatment, the patient feels cared for and supported, and the patient feels an enhanced sense of mastery and control over symptoms.

During the 1990s and early 2000s, the ethics of placebo-controlled trials has been both challenged (Rothman and Michels) and defended (Miller and Brody). Besides the ethical questions concerning whether it is permissible to deprive research subjects of an effective standard treatment, some researchers have questioned how much scientific benefit is added by the use of a placebo control as opposed to an active-treatment control (Freedman, Weijer, and Glass). Systematic reviews have claimed that at least for selected conditions, such as depression, studies conducted without placebo controls might be scientifically suboptimal (Walsh et al.).

Ethical Issues

In the traditional use of placebos, a pharmacologically inert pill might be administered to a patient under circumstances that encourage the belief that a powerful drug is being given. Many patients—the average of one-third is often cited, though this conceals a wide variation among different settings—will experience some degree of positive response (White, Tursky, and Schwartz). This traditional use is ethically questionable because the patient is deceived. Therefore, an ethical analysis of placebo use might proceed with two questions. First, is deception necessary to produce the patient benefit promised by the placebo effect? Second, are there nondeceptive uses of placebos?

If one wishes to use placebo effects for the benefit of patients, one can simply work to enhance those aspects of the patient encounter that have been scientifically correlated with symptom improvement. One can show care, offer explanations, and enhance perceived mastery and control in many ways that require no deception whatever. Because, in the traditional use of placebos, the deception is justified by appeal to patients' benefit (Rawlinson), it is important to see that in almost all patient encounters, a nondeceptive alternative can produce the same result. Moreover, Sissela Bok argued, in her 1978 book, Lying, that the defender of the deception entailed in the traditional use of placebos makes two miscalculations: ignoring possible short-term harm(e.g., missing a diagnosis of serious disease because a placebo has temporarily relieved the patient's complaints) and failing to see how apparently trivial acts build up into collectively undesirable practices (e.g., overreliance on medication).

One may conclude that the traditional use of placebos in therapy can be justified only by very unusual circumstances (in which the use of a dummy pill is the only way to encourage the desired psychological state, for instance). By contrast, because reassuring patients and offering explanations and emotional support are part and parcel of good clinical care, one may argue that a physician has a positive ethical duty to try to enhance the placebo effect in every patient encounter (Connelly).

Counterarguments in defense of the traditional use focus on the claim that the deception is apparent rather than real (Spiro). It might be argued, for example, that if one gives the patient a placebo and says, "There, this will make you feel a lot better," one has not really lied. The increasing scientific interest in the placebo effect has triggered a resurgence of interest in administering placebos to patients, and some have claimed that placebo administration can be combined with respect for patients' rights and with appropriate informed consent (Brown). Perhaps the best reply to these counterarguments was put forth in a 1903 article by Richard C. Cabot: "A true impression, not certain words literally true," (Cabot, p. 345) is what the physician is obligated to promote in the patient. Most efforts at "informed consent" for placebo therapy still seem to rely on some element of equivocation, assuming that if the patient fully understood the pharmacologically inert nature of the remedy, no meaningful placebo effect would result.

Placebos may be employed in ways that do not entail deception and may therefore be fully licit. When placebos are used in controlled studies, it is generally possible to obtain a fully informed consent. It is also possible to use placebos in the therapy of individual patients in a way that avoids deception. One formal procedure for doing so has been termed the "N of 1 Trial," because it is basically a double-blind, controlled research trial performed on a single, informed subject (Guyatt et al.).

howard brody (1995)

revised by author

SEE ALSO: AIDS: Healthcare and Research Issues; Healing; Informed Consent; Pharmaceutics, Issues in Prescribing; Research Methodology; Research, Unethical

BIBLIOGRAPHY

Amanzio, Martina; Pollo, Antonella; Maggi, Guiliano; and Benedetti, Fabrizio. 2001. "Response Variability to Analgesics: A Role for Non-specific Activation of Endogenous Opioids." Pain 90(3): 205–215.

Bok, Sissela. 1978. Lying: Moral Choice in Public and Private Life. New York: Pantheon.

Brody, Howard. 1982. "The Lie That Heals: The Ethics of Giving Placebos." Annals of Internal Medicine 97(1): 112–118.

Brody, Howard. 2000. "The Placebo Response: Recent Research and Implications for Family Medicine." Journal of Family Practice 49(7): 649–654.

Brown, Walter A. 1998. "Harnessing the Placebo Effect." Hospital Practice 33(7): 107–116.

Cabot, Richard C. 1903. "The Use of Truth and Falsehood in Medicine: An Experimental Study." American Medicine 5(9): 344–349. Reprinted in Ethics in Medicine: Historical Perspectives and Contemporary Concerns, ed. Stanley J. Reiser, Arthur J. Dyck, and William J. Curran. Cambridge, MA: MIT Press, 1977.

Connelly, Robert J. 1991. "Nursing Responsibility for the Placebo Effect." Journal of Medicine and Philosophy 16(3): 325–341.

de la Fuente-Fernández, Raúl; Ruth, Thomas J.; Sossi, Vesna; et al. 2001. "Expectation and Dopamine Release: Mechanism of the Placebo Effect in Parkinson's Disease." Science 293(5532): 1164–1165.

Freedman, Benjamin; Weijer, Charles; and Glass, Kathleen Cranley. 1996. "Placebo Orthodoxy in Clinical Research, I: Empirical and Methodological Myths." Journal of Law, Medicine, and Ethics 24(3): 243–251.

Grünbaum, Adolf. 1989. "The Placebo Concept in Medicine and Psychiatry." In Non-specific Aspects of Treatment, ed. Michael Shepherd and Normal Sartorius. Bern, Switzerland: Hans Huber.

Guess, Harry A.; Kleinman, Arthur; Kusek, John W.; and Engel, Linda L., eds. 2001. The Science of the Placebo: Toward an Interdisciplinary Research Agenda. London: BMJ Books.

Guyatt, Gordon; Sackett, David; Taylor, D. Wayne; et al. 1986. "Determining Optimal Therapy: Randomized Trials in Individual Patients." New England Journal of Medicine 314(14): 889–892.

Hróbjartsson, Asbjørn, and Gøtzsche, Peter C. 2001. "Is the Placebo Powerless? An Analysis of Clinical Trials Comparing Placebo with No Treatment." New England Journal of Medicine 344(21): 1594–1602.

Miller, Franklin G., and Brody, Howard. 2002. "What Makes Placebo-Controlled Trials Unethical?" American Journal of Bioethics 2(2): 3–9.

Rawlinson, Mary C. 1985. "Truth-Telling and Paternalism in the Clinic: Philosophical Reflections on the Use of Placebos in Medical Practice." In Placebo: Theory, Research, and Mechanisms, ed. Leonard White, Bernard Tursky, and Gary E. Schwartz. New York: Guilford Press.

Rothman, Kenneth J., and Michels, Karin B. 1994. "The Continued Unethical Use of Placebo Controls." New England Journal of Medicine 331(6): 394–398.

Spiro, Howard M. 1986. Doctors, Patients, and Placebos. New Haven, CT: Yale University Press.

Walsh, B. Timothy; Seidman, Stuart N.; Sysko, Robyn; and Gould, Madelyn. 2002. "Placebo Response in Studies of Major Depression: Variable, Substantial, and Growing." Journal of the American Medical Association 287(14): 1840–1847.

White, Leonard; Tursky, Bernard; and Schwartz, Gary E., eds. 1985. Placebo: Theory, Research, and Mechanisms. New York: Guilford Press.