Placebo Effect

When patients improve after receiving treatments that contain no active ingredient — sugar pills, saline injections, sham surgeries — the improvement is attributed to the placebo effect. The phenomenon has haunted medicine since at least the eighteenth century, when blind trials first demonstrated that patients could be cured by treatments that should not have worked. It sits at the center of an unresolved argument: is the placebo merely a nuisance variable to be controlled for in clinical trials, or evidence that the relationship between healer and patient is itself a therapeutic force? The answer matters to every tradition of medicine, orthodox and alternative alike.

The Discovery of Medical Suggestion

The observation that the mind works upon the body is old. Robert Burton noted in The Anatomy of Melancholy (1621) that “the mind most effectually works upon the body, producing by his passions and perturbations miraculous alterations, as melancholy, despair, cruel diseases, and sometimes death itself.” (Haller, 2014)

The first systematic attempt to separate genuine therapeutic effects from the influence of expectation came in 1789, when French royal commissions investigated Franz Anton Mesmer’s animal magnetism. The commissioners, including Benjamin Franklin, used blind trials — patients were exposed to “magnetized” and unmagnetized objects without knowing which was which. They concluded that the presumed cures could be better explained as the product of imagination. (Haller, 2014) Mesmer’s treatments represented, in the retrospective assessment of R. Barker Bausell, “demonstrations of how susceptible both patients and practitioners are to allowing their expectations (and the suggestions of others) to cloud their judgment.” (Haller, 2014)

The electrotherapy routinely used on “nervous” patients in nineteenth-century private practice operated by a similar mechanism. The ritual and the attention of the physician were themselves therapeutic, regardless of any electrical effect. (Shorter, 1997)

Numerical Method and the Decline of Heroic Medicine

Pierre Charles Alexandre Louis’s methode numerique in the early nineteenth century helped dismantle rationalistic beliefs embedded in the practice of bleeding and the heroic use of calomel and tartar emetic. As systematic methods of investigation demonstrated that these therapies performed no better than expectant management, medicine entered a period of therapeutic nihilism — the uncomfortable recognition that most of its traditional arsenal might be doing nothing at all. (Haller, 2014)

Arthur K. Shapiro later pushed this argument to its logical extreme: of the 4,875 different remedies and 16,842 prescriptions used by healers over recorded history, only a handful were actually effective. The minerals, herbs, and animal excretions used to purge, blister, sweat, and induce vomiting in patients were, in Shapiro’s hypothesis, nothing more than surrogates for the placebo effect. (Haller, 2014)

Beecher and the Powerful Placebo

Henry K. Beecher’s 1955 article “The Powerful Placebo” in the Journal of the American Medical Association announced that in clinical trials, the effects of the placebo often exceeded those of a pharmaceutical drug for symptoms of pain, headache, and nausea. (Haller, 2014) Beecher’s thesis hinted at the medicalization of phenomena that had thrived for generations under terms like verbal affirmation, visualization, suggestion, relaxation, and positive thinking. (Haller, 2014)

In 1961, Beecher strengthened his case with “Surgery as Placebo.” Double-blind sham surgery trials for bilateral internal mammary artery ligation had shown that the procedure had no greater effect on angina pectoris than a sham operation. Fourteen subjects received only a skin incision; twenty-one received actual ligation. The outcomes were indistinguishable. (Haller, 2014)

The Rise of the Randomized Controlled Trial

The methodological response to the placebo problem was the randomized controlled trial. Randomization was first formulated by Ronald Fisher for agricultural experimentation and applied to medicine by Austin Bradford Hill, whose 1937 Principles of Medical Statistics became the standard for medical research. (Haller, 2014) The first properly randomized clinical trial — testing streptomycin for tuberculosis in 1948 — used random-number allocation rather than strict alternation not because it better abolished selection bias, but because it was easier to conceal from investigators and patients. (Haller, 2014)

The RCT was designed to isolate the “specific” effect of a treatment from the “nonspecific” effects of expectation, attention, and the therapeutic relationship. But this very design made the placebo effect visible as a measurable phenomenon in its own right, not merely a source of noise. Haller observes that the placebo has since served as “both mediator and judge” between conventional and unconventional medicine, challenging both in unforeseen ways. (Haller, 2014)

Industry and the Corruption of the Trial

The pharmaceutical industry’s commercialization of the RCT introduced new problems. Drug companies created privatized contract research organizations, skewed published research toward results favorable to sponsors, supported ghost authorship, and — in the cases of companies like Lilly and Pfizer — even attempted to exclude placebos altogether from their drug trials. (Haller, 2014) The tool designed to protect patients from ineffective treatments was itself vulnerable to institutional capture.

The Placebo and Psychotherapy

Jerome Frank argued that “the placebo may be as effective as psychotherapy because the placebo condition contains the necessary, and possibly the sufficient, ingredient for much of the beneficial effect of all forms of psychotherapy.” (Haller, 2014) As late as 1997, the effectiveness of more than 90 percent of psychotherapy’s more than four hundred modalities had not proven superior to placebo. (Haller, 2014) This raised the uncomfortable question of whether much of psychotherapy was simply a variation of the New Thought, mind-cure, and positive-thinking traditions that had existed outside medicine for over a century.

William James had characterized the American New Thought healing tradition as “healthy-mindedness” — “the only decidedly original contribution of the American people to the philosophy of life” — because, pragmatically speaking, these nonreductionist healing therapies sometimes “worked.” (Haller, 2014)

The Dualism Problem

Haller identifies Cartesian dualism as the source of the conceptual impasse. Descartes’s separation of mind and body gave birth to two polar traditions in Western medicine: one grounded in reductionist biochemical processes, the other explaining healing in terms of the soul’s effect on the body. (Haller, 2014) The placebo effect sits exactly on this fault line. If healing can be produced by expectation, suggestion, and the therapeutic relationship — none of which are reducible to biochemistry — then medicine built exclusively on reductionist models is missing something. But if placebo effects are ultimately neurochemical events (endorphin release, dopamine pathways), they are just another mechanism to be studied within the reductionist framework.

The question remains open: is the placebo effect a bridge spanning psychological and neurobiological processes, or proof that one of these frameworks is inadequate? (Haller, 2014)

Etymology and Early Medical Usage

The word itself has a history. It appeared in George Motherby’s New Medical Dictionary in 1785, defined as a “commonplace method or medicine” used by practitioners who had nothing in their medical bag to address a disease but who intended to satisfy their patients’ “imaginations, beliefs, and expectations while nature took its course.” (Haller, 2014) The definition is already clinical: a placebo was a practitioner’s technique for managing patient expectation while doing nothing specific — an acknowledgment, baked into the vocabulary from the beginning, that expectation was itself a therapeutic variable worth managing.

Endorphins and the Neurobiological Turn

The discovery of endogenous opioids in the 1970s opened a new chapter. When John Levin, Newton Gordon, and Howard Fields reported in 1978 that some forms of placebo analgesia were initiated by endogenous opioids, a scientific debate erupted over whether the placebo was truly as “inert” as its definition suggested.(Haller, 2014) The emotional impact of suggestion upon the hypothalamus, Samuel Perry and George Heidrich argued, generates endorphin release — a mechanism that forced a rethinking of the mind-body relationship and provided placebo analgesia with a physiological substrate it had previously lacked.(Haller, 2014) The placebo was no longer simply a psychological phenomenon; it had measurable neurochemical correlates.

The pharmacological specificity of placebo analgesia was confirmed by naloxone challenge: when an opioid antagonist was administered to patients experiencing placebo-mediated pain relief, the relief was abolished, demonstrating that their brains were genuinely releasing endogenous opioids rather than producing a purely psychological report of reduced pain.(Sapolsky, Robert M., 2004) The same opioid pathway that explains surgical-wound pain reduction appears to underlie the classic finding that hospital patients with a window view of trees requested less post-operative pain medication than patients whose windows faced a blank wall.(Sapolsky, Robert M., 2004)

This finding invited a reconceptualization. University of Michigan anthropologist Daniel Moerman and physician Wayne Jonas argued that “placebo” was a misnomer: an inert substance cannot be the direct cause of anything, so what produces the effect is not the pill but the meaning the brain associates with it. They proposed “meaning response” as a more accurate term — “the psychological and physiological effects of meaning in the treatment of illness.” (Haller, 2014) The reframing had practical stakes: if the mechanism is meaning rather than deception, then the therapeutic relationship itself — not only the inert ingredient — is the active variable.

Open-Label Placebos and the Deception Problem

The standard assumption about placebos was that their effectiveness depended on the patient not knowing they were receiving one: deceive the patient, produce the effect; inform the patient, cancel it. Recent research has challenged this. Evidence now suggests that placebos can be effective even when patients know they are receiving one — so-called open-label placebos — which would allow their therapeutic use without the deception that had made Sissela Bok and subsequent bioethicists uneasy.(Stegenga, 2018) Ted Kaptchuk, the Harvard-based researcher whose work has done most to map this territory, argues that the path forward requires reconciling “solid objectivity arrived at independent of belief and culture with the absolute uncontestable importance and meaningfulness of subjectivity.”(Haller, 2014) Whether the clinical and regulatory apparatus of evidence-based medicine can accommodate that reconciliation remains unsettled.

The Nocebo Effect

The placebo has a mirror. Walter Kennedy introduced the term “nocebo” in 1961 to name the inverse phenomenon: negative expectations producing real harmful effects, including dizziness, depression, nausea, and palpitations.(Haller, 2014) If positive expectation can trigger endorphin release and reduce pain, negative expectation can produce measurable physiological harm through comparable mechanisms. Jacob Stegenga observes that both placebo and nocebo effects arise from expectation mechanisms whose underlying physiological basis remains poorly understood, and that most expectation effects are beneficial rather than harmful — most are not nocebo.(Stegenga, 2018) The nocebo raises particular ethical concerns in clinical practice: informed consent requires disclosing side effects, but disclosing expected side effects can itself produce them.

See Also

• Randomized Controlled Trial
• Evidence-Based Medicine
• Vis Medicatrix Naturae
• Homeopathy
• Vitalism
• Mechanism
• Clinical Judgment
• Therapeutic Nihilism

Sources

Auto-generated from evidence card IDs listed in frontmatter.