Is Hypnosis any Better than a Placebo?

Hypnotist Rob Macinnes

by Rob MacInnes

Perhaps as an informed potential client you’ve wondered about the relationship of hypnotism to a variety of suggestibility responsible for the placebo effect. Some of the readers who are practicing hypnotists, may have on occasion seen the confidence of a client get shaken by others who, in an offhanded way, commented that hypnosis is “just a placebo”, or “no more than a placebo”? Sometimes these statements are even made by otherwise knowledgeable helping professionals. Do these statements have any basis in fact? Most practitioners would probably assert from training or experience that this claim is untrue, that there is more going on with hypnosis than a placebo effect. And yet, if it were true, would there really be anything wrong with that? As Moerman and Jonas (2002) point out, to say that a treatment, such as acupuncture “isn’t better than a placebo” does not mean that it does nothing (Moerman & Jonas, 2002). Hans Eysenck (1991), one of the most prolific writers in the history of psychology, observes that placebo treatment in psychiatry is just about as effective as psychotherapeutic techniques (Eyesenck,1991).

Mosby’s Dictionary of Complementary and Alternative medicine offers two functional definitions of a placebo, “An inert substance used in control groups of clinical studies to maintain blinding” and “the beneficial effects of the meaning and context of treatment independent of the treatment itself” (Jonas, 2005). This article mainly discusses a study published in 1969 that elegantly illustrates how both of these aspects relate to hypnotism for the treatment of pain.

Placebo effects pervade all forms of health intervention, and these effects are often extraordinary; Placebo response with analgesic drugs can vary in proportion to the efficacy of the drug itself (Shea, 1991), Placebo analgesia can elicit the production of endogenous opiates (Moerman & Jonas, 2002), and experimental placebo effects can even mimic other pharmacological properties of the active agent with which it is being compared (Evans, 1974, 1985). People can even become addicted to placebos, showing many of the traits of drug dependance such as tolerance over time and withdrawal symptoms (as cited in Shea, 1991).

The use of hypnosis as a treatment of choice, just like the administration of a drug or any other health related intervention, can of course have strong placebo effects on the issue or condition being worked on; however, the practitioner and the would-be client can be assured that it is strongly suggested by empirical research that genuine hypnotic effects go significantly beyond, and are in fact unrelated to, the “non specific” effects produced by a placebo therapy.

A study published in 1969 by Thomas H. McGlashan MD, Fredrick J. Evans, PhD, and Martin T. Orne MD, PhD was designed to test the hypothesis that there are at least two mechanisms involved in hypnotic analgesia. One is the “non specific” placebo effects of using hypnosis as a treatment method, and the other is the distortion of the perception of the pain sensation specifically induced during deep hypnosis, that a placebo response for analgesia and that of hypnotically produced analgesia are two completely separate things.

When conducting experimental research on hypnotism, certain tests are employed which use standardized hypnotic inductions and have been developed to gauge an individual’s susceptibility to hypnosis as a specific score or measurement of how easily a person can be hypnotized. The most commonly used tests in research are the Stanford Hypnotic Susceptibility Scale (SHSS), which was developed and modified by Andre M. Weitzenhoffer and Ernest R. Hilgard: forms A, form B (as a follow up to re-test susceptibility on a second session), form C which attempts to order the items in increasing difficulty on a statistical basis (Udolf, 1987) and the Harvard Group Scale of Hypnotic Susceptibility (HGSHS) which is an adaptation of the SHSS by Ronald Shor and Emily Orne to test entire groups of people at a time.

M.T. Orne designed a double blind study with a hypnotic group of 12 “high” susceptible subjects and a control group of 12 “low” susceptible subjects as determined by performance on the HGSHS: form A and the SHSS: form C.

The “low” susceptible subjects were selected to show that any analgesic effect they exhibited was unlikely to have been the result of inadvertent hypnosis, these effects would be understood as caused by secondary placebo effects which could accompany a hypnotic procedure. These control subjects were given suggestions that produced a deep state of relaxation and were subsequently given suggestions that one of their hands would become analgesic. Electric shocks were administered in each hand, and with some manipulation (covertly lowering the voltage), they were led to believe they had produced a marked hypnotic analgesia in the selected hand. In this manner, the control subjects were made unaware for the duration of the study, that they were part of the group selected for their unusually low susceptibility to hypnosis.

On the second session, all subjects were then assigned a hand pumping task to establish a baseline threshold for their tolerance for ischemic muscle pain. Blood flow to the arms was obstructed with a blood pressure cuff inflated to 200 millimeters of mercury. Later in this second session, a relaxation method of hypnotic induction was performed on all subjects in each group and analgesia was suggested in one arm. Tolerance for pain was assessed via this hand pumping task once more.

In a third session, a placebo “drug” was administered as a powerful new analgesic named, N-methyl-O-isopropyl oxazolidine and again the ischemic pain task was administered as before. The explanation given to all subjects for this third session included information about the pharmacology of analgesia and the particular drug being used. This subterfuge was meant to give the impression to both high and low susceptible groups that the “efficacy of hypnosis” was being measured against a powerful physiologically active drug rather than betray the fact that the purpose of the experiment was to distinguish analgesia in deep hypnosis from the “non specific” placebo effects of using hypnosis as a treatment method.

The results showed no relationship between the level of placebo analgesia and that of susceptibility to hypnosis. Both the truly hypnotized and non-hypnotized subjects  (the group selected for low hypnotic ability that was surreptitiously led to believe they were capable of producing hypnotic analgesia) produced elevated ischemic pain thresholds, however the effect was much greater for the true (highly susceptible) hypnotic subjects in the second session.

The responses of both of these groups to the placebo “drug”, without any hypnotic procedure, were identical. The placebo effect was equally distributed regardless of experimentally measured hypnotic susceptibility. The high susceptible subjects performance in their “drug” placebo session was not as great as their own hypnotic analgesia performance in session two. This confirmed their hypothesis that susceptibility to hypnosis (at least in an experimental setting) is not intrinsically correlated with placebo responsivity.

Evans (1969) speculates:
It is not possible to specify the mechanisms involved in the alteration of pain perception achieved by some of the deeply hypnotized Ss. A variety of negative hallucinations involving most sensory modalities can be induced with deeply hypnotized Ss, and it is plausible that the somesthetic sensation of pain can be subjected to the same kind of cognitive changes as those involved in other negative hallucinations. (p. 243)

One conclusion drawn from this is that when the subject believes in the efficacy of the procedure, hypnotically induced analgesia can be effective regardless of the client’s aptitude for hypnosis, however McGlashan et al. (1969), emphasizes that the cognitive distortion produced by hypnotic analgesia in highly susceptible subjects is unique to hypnosis and is unrelated to the effects of nonspecific placebo factors accompanying the hypnosis procedure (McGlashan et al., 1969).


Eyesenck, H (1991). Is suggestibility?  In J Shumacker (ed.) Human suggestibility: advances in theory research and application (pp. 76-90) New York: Routledge

Jonas. (2005) . Placebo. In Mosby’s Dictionary of Complementary and Alternative Medicine. retrieved from

McGlashan, T.H., Evans, F.H., Orne, M.T. (1969) . The nature of hypnotic analgesia and placebo response to experimental pain. Psychosomatic medicine, Vol XXXI (No 3), 227-246

Moerman, D. E. & Jonas, W.B. (2002)   Deconstructing the Placebo Effect and Finding the Meaning Response [electronic version] . Annals of internal medicine, 136 no. 6 471-476

Udolf R. (1987) . Handbook of hypnosis for professionals. Second edition. New York: Van Nostrand Reinhold Co.

This entry was posted in E-Zine Articles. Bookmark the permalink.

Leave a Comment

You must be logged in to post a comment.