Placebo Effect

Placebo Effect Definition

The word placebo derives from the Latin placere, meaning “to please.” Placebo literally means “I shall please.” In medicine, a placebo is a therapeutic intervention that is used intentionally or unintentionally for its nonspecific psychological or psychophysiological effects. Placebos may be used under a variety of circumstances, such as when the patient demands medication that the physician feels is unjustified, or when the physician wishes to provide hope and an expectation of improvement even though medication is unlikely to have any physical effect. The physician is, in effect, enacting the phrase “I shall please.” Patients receiving placebos do improve and this improvement is referred to as a placebo effect.

The Role of Placebos in Medical Research

Placebos play an important role in medical research. In evaluating a new medication, the standard research design is to randomly assign participants to receive either the medication or an identical-looking, placebo pill. In these studies, all who are directly involved should be “blind,” that is, unaware of whether a particular participant is receiving the pill or a placebo. The patient, clinician, and outcome evaluator all need to be blind so that they are not influenced by their various expectations for medication or placebo effects.

Unfortunately, in medication trials, the blinding procedures may not work perfectly. Many medications have side effects, and patients who experience side effects may be more convinced that they are receiving active medication, and patients who do not experience side effects may suspect that they are on a placebo. The clinician and independent evaluator may also be influenced by the report of side effects and may also guess the medication status of the patient. This problem has been the focus of a recent article criticizing the effectiveness of antidepressant medication by Kirsch and colleagues. They argued that most of the effect we see in research trials of antidepressant medication may be due to placebo effects heightened by positive expectations for patients with side effects.

There are research strategies to improve the maintenance of the blinding in research trials. Active placebos are medications that produce side effects similar to those of the medication under investigation, but without the active ingredient. For example, in studies assessing antidepressant medication, antihistamines are sometimes used as the placebo because they may cause dry mouth and blurred vision, which are side effects of certain antidepressants. Because participants receiving placebo may also experience side effects, all still remain blind.

Placebo conditions in medical research may go beyond pill placebos. For example, Moseley and colleagues studied the effectiveness of a particular form of knee surgery for arthritis. Two different forms of surgery were compared to placebo surgery in which an incision was made, but no further surgical work was done. In this study neither form of actual surgery did better than the placebo in reducing pain or in increasing knee functioning.

The Role of Placebos in Psychotherapy Research

Psychotherapy research has adopted the research design used for evaluating new medications. In analogous fashion, a new form of psychotherapy is generally compared to a placebo psychotherapy. Placebo psychotherapy usually consists in seeing a therapist for the same number and length of sessions, but the therapist in the placebo condition refrains from using what are thought to be the active ingredients in the new therapy. The placebo therapy controls for those components that are common to all therapies: a relationship with a caring therapist, a chance to examine closely the problems bothering the person, and so on. Often placebo therapies involve a specific intervention that is thought to be of minimal benefit. For example, in a study of a new therapy for phobias, the placebo condition might consist of relaxation training. Relaxations may be helpful, but alone are not as likely to be as effective as an active treatment that involves exposure to the feared object.

Blinding is quite different in psychotherapy research. Obviously clinicians administering the therapy cannot be blind as to what they are doing. Patients cannot be blind in that they are exposed to different rationales for the experimental and the placebo therapies, either explicitly or implicitly. Only evaluators can be blind, but they must be careful to ask the participant being interviewed to say nothing about the therapy itself that might let the interviewer know which condition they were in.

To be effective, a placebo should produce the same positive expectations of help as the experimental therapy. Borkovec and Nau had people rate the potential effectiveness of therapies from their rationales. They concluded that the results of actual outcome studies were entirely predictable from the rationale ratings. The weaker rationales of placebo conditions corresponded to weaker effects in published papers.

Morris Parloff discussed the various problems associated with placebo psychotherapy conditions and the various ways in which they fail to function in a way analogous to the placebo pill in medical research. He concluded that the psychotherapy placebo condition in psychotherapy research was an inadequate control and he told psychotherapy researchers to “go forth and sine qua non no more.”

A number of alternative research designs are available in the psychotherapy literature. One is to compare a new psychotherapy to the best-established therapy in common use. Another is to compare a standard therapy to a new therapy that improves on the standard with some additions. Disassembly designs compare the full treatment to therapy with one or another element removed (disassembled). This design allows for the evaluation of the components of a complex therapy program. Because medications for psychological disorders have been much more extensively studied than psychotherapies, some studies compare psychotherapies to a known effective medication and a placebo drug. The difference between the placebo and the active drug is used as the scale against which to evaluate the psychotherapy.

Factors Enhancing the Placebo Effect

The expectations of both the clinician and the patient are critical determinants of the placebo effect. Thus, therapist and patient belief that a powerful drug is being used enhances the placebo effect. When patients or clinicians “break the blind” by correctly guessing that a placebo treatment has been administered, expectancy for improvement may result in a diminished placebo effect. Guessing may be based on lack of improvement, lack of side effects, or higher dosage in comparison to medicated patients.

In addition to expectancy effects, various characteristics of the treatment have been found to enhance the placebo effect. Literature reviews report the following general findings: (1) injections are more powerful than pills, (2) capsules are perceived as stronger than tablets, (3) size is positively correlated with perceived strength, and (4) pill color is related to perceived effect. These results underscore the importance of matching the placebo treatment to the active treatment as closely as possible.

Finally, according to Shapiro and Shapiro, greater placebo effects are observed “in pleasant, non-threatening, efficient clinical settings with doctors who are perceived by patients as warm, likable, and interested in them.” From a therapeutic perspective, the placebo effect is enhanced by a clinician who allays anxiety and provides a meaning to the illness.

Possible Mechanisms of Placebo Effects

Some common psychological and physiological mechanisms that have been proposed to explain the placebo effect include classical conditioning, response expectancy, anxiety reduction, and an opioid-mediated pathway.

According to the classical conditioning model, the placebo response can be conceptualized as a conditioned response. The drug or treatment is the unconditioned stimulus, and its physiological or pharmacologic effects are the unconditioned response. Over the course of one’s lifetime, neutral behavioral and environmental stimuli (i.e., clinical setting, doctor, pills, injections, etc.) act as conditioned stimuli, which through repeated pairing with the unconditioned stimulus (i.e., receipt of drug) eventually produce a conditioned response—the placebo response. Experimental animal data, including results that show a conditioned immunosuppressive response can be elicited in rats, provide some support for the conditioning model of the placebo response. Limitations of the classical conditioning theory include the finding that some conditioned responses to drugs occur in the opposite direction of the placebo and drug responses. Thus, the classical conditioning model cannot account for the placebo phenomenon in its entirety.

More contemporary theories of learning, such as Kirsch, suggest that the placebo effect is explained by expectations such as “the anticipation of the occurrence of non-volitional responses, such as pain, sadness, joy, intoxication, vomiting, and alertness.” In this framework, although various modes of learning, including conditioning, reading, and observation, can shape expectancies, the expectancies themselves are considered the key mediating mechanism of the placebo response. Placebo effects do not always mimic drug effects, but may elicit responses that are consistent with cultural expectancies. Positive expectancies may be engendered not only in the patient, but in the patient’s family and friends. These expectancies may lead to activation of efforts to solve problems and improve the patient’s life circumstances, making the patient feel generally better.

The hypothesis that anxiety reduction is a mechanism for the placebo effect is supported by reports that increased pain tolerance is associated with lower situational anxiety following placebo administration. However, findings that placebo effects can be highly specific (e.g., local to a specific part of the body) imply that global changes, such as anxiety reduction, are not satisfactory explanations.

Opioid pathways likely are responsible for at least some instances of the placebo response, namely placebo pain relief. Evidence from animal and human studies supports the idea that an opioid-mediated neural circuit is related placebo pain relief. Specifically, medications that block opioids reduce the pain relief of placebo responders.


  1. Borkovec, T. D., & Nau, S. D. (1972). Credibility of analogue therapy rationales. Journal of Behavior Therapy and Experimental Psychiatry, 3, 257-260.
  2. Evans, F. J. (1985). Expectancy, therapeutic instructions, and the placebo response. In L. White, B. Tursky, & G. E. Schwartz (Eds.), Placebo: Theory, research and mechanisms (pp. 215—228). New York: Guilford.
  3. Fields, H. L., and Price, D. D. (1997). Toward a neurobiology of placebo analgesia. In A. Harrington (Ed.), The placebo effect: An interdisciplinary exploration (pp. 93-116). Cambridge, MA: Harvard University Press.
  4. Kirsch, I. (1997). Specifying nonspecifics: Psychological mechanisms of placebo effects. In A. Harrington (Ed.), The placebo effect: An interdisciplinary exploration (pp. 166—186). Cambridge, MA: Harvard University Press.
  5. Kirsch, I., Moore, T. J., Scoboria, A., & Nicholls, S. S. (2002). The emperors new drugs: An analysis of antidepressant medication data submitted to the U.S. Food and Drug Administration. Prevention and Treatment, 5, Article 23.
  6. Moseley, J. B., O’Malley, K., Petersen, N. J., Menke, T. J., Brody, B. A., Kuykendall, D. H., et al. (2002). A controlled trial of arthroscopic surgery for osteoarthritis of the knee. Journal of Family Practice, 51, 813.
  7. Parloff, M. B. (1986). Placebo controls in psychotherapy research: A sine qua non or a placebo for research problems? Journal of Consulting and Clinical Psychology, 54, 79-87.
  8. Ross, S., & Buckalew, L. W. (1985). Placebo agentry: Assessment of drug and placebo effects. In L. White, B. Tursky, & G. E. Schwartz (Eds.), Placebo: Theory, research and mechanisms (pp. 67-82). New York: Guilford.
  9. Shapiro, A. K., and Shapiro, E. (1997). Is it much ado about nothing? In A. Harrington (Ed.), The placebo effect: An interdisciplinary exploration (pp. 12-36). Cambridge, MA: Harvard University Press.
  10. Thompson, W. G. (2000). Placebos: A review of the placebo response. American Journal of Gastroenterology, 95, 1637-1643.

Back to Health Psychology.