Acupuncture, Touch, and the Placebo Response

Anthony Campbell

Reprinted from Acupuncture, touch, and the placebo response; 8(1):43-46, March 2000, Complementary Therapies in Medicine, by permission of the publisher, Churchill Livingstone. Users may not print out or otherwise reproduce copies of the material without the written permission of the copyright holder

Acupuncture practitioners who subscribe to the modern or non-traditional version of the therapy maintain that what they do is scientific, in the sense that it is founded on what mainstream science knows about the nervous system and other physiological systems[1]. Moreover, they can point to a modest but not insignificant body of research to support their claim, and there are plausible physiological mechanisms that may explain at least some of the observed phenomena[2]. And yet, if you talk to such practitioners, you quickly get the impression that they regard what they do as an art as well as a science. This is true of all of medicine, of course, but it is especially true of manual techniques such as acupuncture; most practitioners think that the skill with which the treatment is applied makes a lot of difference to the outcome.

Admittedly there are a few who don't share this attitude. A consultant neurologist who used acupuncture told me that he found the actual needling of patients boring and therefore deputed it to physiotherapists. However, this view is not widely held. Indeed, quite a number of acupuncturists, not necessarily adherents of the traditional system, can be induced to admit that they feel as if "something" were being transmitted via the needle to the patient, although they generally disavow, in the next breath, any suggestion of a mysterious or paranormal "life force". It is the nature of this "something" that I want to explore. My hope is that this can be done without plunging us into irrationality. The clue, I suggest, is to be found in the quality of touch.

Touch as a form of therapy

Touch is a form of communication, which differs from spoken language in that it is "primitive" and pre-verbal and can produce a sense of reassurance and calm in a patient. There are good reasons based in our evolutionary history why this should be the case: grooming is a standard method of social communication for our simian relatives and the rarity with which most humans indulge in it, outside a sexual or a mother-baby context, is probably an aberration from an evolutionary standpoint. Conventional doctors do touch their patients in the course of a physical examination but the ostensible function of the procedure is diagnostic, although even here there may be an element of reassurance and acceptance that has some therapeutic value. The psychological importance of touch is more obvious in the case of the manual therapies, in which diagnosis and treatment tend to overlap and begin to blend into each other.

In the case of acupuncture, for example, it is natural to focus on the needles as the central thing. But acupuncture generally also entails a more or less comprehensive preliminary physical examination, which in the traditional system includes palpation of the pulses and in the modern version is likely to include manual palpation of the spine and other areas for trigger points (TPs)[3]. I believe that this tactile component, though it is primarily regarded as a diagnostic procedure, may at times also have therapeutic effects independently of the needles. One reason for this is that the manual pressure used for detection may also inactivate the TPs when found, at least temporarily, by means of so-called "acupressure". And even if this does not happen, the process of examination is probably perceived unconsciously by the patient (and perhaps the acupuncturist) as therapeutic. Pressing the trapezius muscles to search for TPs is quite similar to giving a therapeutic massage; indeed, many patients may have asked a spouse or other relative to massage their shoulders in this way to relieve tension. The same applies to pressure designed to elicit tenderness in other parts of the body; this recalls the practice, widespread in Asian countries, of asking a relative to walk on one's body with bare feet in order to relieve pain.

There is probably a degree of touch-induced healing involved in every successful episode of manual therapy, including acupuncture. This is the theme of an important recent book by Bevis Nathan, an osteopath[4].

We may say that this or that technique is a local-tissue technique only and devoid of meaning, and we may have reasonable arguments upon which this statement rests. Nevertheless a patient may experience our touch (consciously or unconsciously) emotionally, whatever we may say... a positive experience of a practitioner's touch will induce feelings of improved self-image, increased self-worth and well-being. This will occur where touch is caring, sensitive, confident, competent and respectful, and is accompanied by requests for permission to perform techniques. These positive feelings will generate and reinforce the patient's belief that healing is taking place. This, in turn, will generate bodily physiological healing events. (p.125).

Conversely, of course, a practitioner whose touch is not caring, sensitive, and so forth will not generate these desirable effects. This may well be part of the reason why certain acupuncturists seem to obtain consistently better results than others.

The case of self-acupuncture is interesting in this context. Patients are sometimes taught to carry out their own acupuncture and this generally works quite well. However, it is not unusual for them to say that their own efforts are less successful than those of the therapist, and they may ask for an occasional "booster" to maintain the improvement. Such patients are usually doing the acupuncture perfectly well from a technical point of view, and therefore it seems that there is something extra that is being contributed by the therapist.

We could summarize all this by saying that success in acupuncture may not be wholly due to the needles but may also depend on messages transmitted and received, largely unconsciously, via the acupuncturist's hands during the preliminary diagnostic examination. The fact that touch appears to work at a more basic or primitive level than does verbal suggestion would explain why patients' prior beliefs in acupuncture seem to make no difference to the clinical outcome[5]. Verbal interaction no doubt plays a part, but is possibly less important than the touch component. Both patient and therapist will attribute all, or nearly all, the symptomatic relief to the needles, perhaps with the addition of a degree of "suggestion", but this may not be the whole story.

To put it differently, probably almost every successful episode of manual therapy is also, to some extent, an episode of psychotherapy. The reverse is also true: psychotherapy is likely to have effects on the body. It might therefore seem rational to use touch as part of psychotherapy, but this practice is almost entirely taboo in conventional psychotherapy, no doubt for understandable reasons. In recent years, however, there have been some attempts to overcome this barrier; the subject is well discussed, with references, by Nathan (pp. 133-193)[4].

Does this mean that acupuncture is "just placebo"?

At this point, if not before, scientifically minded acupuncturists are likely to feel uncomfortable, for we seem to be coming perilously near to conceding the critics' case that acupuncture is nothing more than a superior placebo. This is in fact still a possibility, as Ernst and White have pointed out (p. 156)[6]. And what makes the critics' charge even more telling is the fact that although there is a fair amount of scientific evidence to show that acupuncture in the broad sense has some effect, there is very little to show that any one method of doing it is better than another. It is difficult to argue convincingly that traditional acupuncture is better than modern acupuncture or vice versa, or that one version of modern acupuncture is better than another. Unanswered quite basic questions abound. Should the needles be left in for half a minute or twenty minutes? Should the needle be placed accurately in the trigger point or is superficial needling over the site enough[3]? Is manual stimulation necessary? Is electrical stimulation more or less effective than manual stimulation? Do classic acupuncture points have any real existence or special properties? Not much reliable guidance is to be found in the research literature on these or a host of other practical questions. (Incidentally, the same is true of many other kinds of unorthodox medicine, such as homeopathy, which also exist in a number of allotropic forms.) But if they all work to much the same extent, is it possible that none of them really works at all?

There seem to be two main strategies that might be adopted by therapists who wish to refute the charge that what they offer is "mere placebo". The first is to take refuge in mysticism, and the second is to admit that much of the effect may indeed be due to non-specific influences but to question the generally received understanding of placebo. I shall look at each of these in turn.

The mystical alternative

As Nathan points out, the mainstream manipulative professions have sought to gain acceptance from orthodox medicine by explaining their procedures in mechanical and physiological terms. The same is true of modern, non-traditional, acupuncture. However, the founders of osteopathy and chiropractic, Andrew Still and Daniel Palmer respectively, were vitalists, as is a minority of their followers even today. There are also several less conventional schools of manual therapy that are explicitly vitalistic. Among the manual therapies that are based on ideas of this kind are shiatsu, therapeutic touch, polarity therapy, and the various schools of "laying on of hands". Traditional Chinese acupuncture is expressly vitalistic, depending as it does on the concepts of yin-yang polarity and chi.

The modern theories underlying such treatments often make mention of "energy fields", a metaphor derived from physics, but the origins of these ideas are much older. Nathan connects them, I am sure correctly, with mystical teachings such as those that were elaborated in great detail by the Neoplatonists in Alexandria in the early centuries of the Christian era, although it would be a monumental undertaking to trace the routes by which these ideas have descended to their present form in unconventional medicine. A more immediate source can be found in the work of Wilhelm Reich. He was an associate of Freud and worked with him in Vienna after the First World War. Like Freud, he was much preoccupied with sex and placed the attainment of satisfactory orgasm at the root of his therapeutic theory. Later in life he went to the USA, where he developed what sounds like a complex delusional system concerned with "orgone", a kind of cosmic energy. This brought him into conflict with the US Food and Drug Administration and eventually he was committed to prison, where he died. Some of Reich's ideas were developed by others, notably Alexander Lowen, and gave rise to a system of therapy called Bioenergetics. Other offshoots include Biodynamic therapy and Biosynthesis.

Nathan is prepared to regard "energetics, field theory, subtle-body theory, spirituality and even theosophy" as ideas that deserve to be researched and taken seriously, because they offer "the only possibility for a comprehensive understanding of human life - one that explicitly allows for the wealth of history of human experience". I can't follow him in this. It seems to me that to allow oneself to be involved in these murky areas is to risk submerging beyond hope of rescue in a sea of confusion. I prefer to think that we will ultimately be able to offer an explanation of the psychological and emotional effects of the manual therapies in neurophysiological terms, though this will entail a considerably better understanding of the way placebos work than we have at present.

Don't be frightened of the placebo

The placebo concept still requires a lot of elucidation[7], and indeed may not be easily applicable to the manual therapies, but meanwhile we should at least remember that the "placebo effect" is a neurophysiological phenomenon. This may seem obvious when one thinks about it, but it is often forgotten. As I have argued elsewhere, there seems to be an unwarranted assumption on the part of many doctors that placebo effects are somehow unreal and not very reputable[8]. They think in this way because they are unconsciously using a dualistic model of the organism. They say, in effect, that there is the body, which is a physiological system on which drugs and other medical treatments are supposed to operate, and there is the mind, which can be affected by suggestion to produce a spurious effect. Between the mind and the body there is an ill-defined no-man's land rather inadequately occupied by something called psychosomatic medicine. Doubtless few would admit to holding a crude view of this kind, yet this is what seems to be implied in the pronouncements of some medical critics of complementary medicine. What is odd about it is that it is incompatible with the materialism which otherwise characterizes most medical thinking, even (or perhaps especially) in psychiatry. As a recent Lancet leading article remarked: "With a few dissenters, psychiatry accepts that in principle its disorders are disorders of the brain that can and should be investigated as such."[9]. Exactly the same applies to the placebo phenomenon: it is produced by the brain and should be approached from that direction.

There is good scientific evidence available to show that the endogenous opioids are mediators of placebo analgesia although this may not be the whole explanation, especially in view of the wide variations in what constitutes a placebo[10]. We are not yet able to specify in any detail which brain areas and mechanisms are important in relation to the subjective phenomena that occur in patients receiving manual treatments. I have elsewhere proposed that the limbic system may be involved in the production of emotional phenomena such as euphoria, laughter, and crying in acupuncture patients[11]. Should this idea turn out to be even partially correct, there is no reason to think that acupuncture would be the only treatment that is capable of modifying the activity of the system. Other manipulative methods, such as those used in physiotherapy, osteopathy, and chiropractic might act in the same way, and so too might less conventional therapies, all of which can on occasion give rise to emotional abreactions. There are numerous descriptions by patients of emotionally charged experiences while receiving such treatments, especially the more vigorous, such as Rolfing. The limbic system might thus represent a final common path (though probably not the only one) by which such treatments work. The touching that is entailed in examining a patient for TPs would then be simply one more means of accessing the limbic system. To ask whether this is a placebo effect seems to be almost meaningless in this context.

Nathan also favours the limbic system as the part of the brain that is likely to be involved in psychophysical phenomena of this kind, for he postulates that a state of chronic hyperarousal in the anterior cingulate cortex (part of the limbic system) results in a "chronic but incoherent efferent discharge into the soma... which not only explains chronic muscular hypertonia, but also provides a baseline explanation for all psychosomatic pathology"(p. 188)[4].


What I have suggested in this paper is that touch is important in manual therapies, including acupuncture, but its effects depend on the brain and are conditioned by a very wide range of influences, not all of which are easy to identify or delineate. In the case of acupuncture, we need to consider not only the patient's reactions to the insertion of the needle but also the effects of the manual examination. There seems no good reason to be alarmed or defensive about admitting the importance of this for the final therapeutic outcome.

This analysis seems to offer plenty of scope for the quite widespread impression among acupuncturists that there is "something else" involved besides the effects of the needle. That "something else" may indeed prove to be what is often called the placebo effect, but this should be understood to be a neurophysiological phenomenon and hence as "real" as anything else.

On a practical level, this has important implications for how practitioners of modern medical acupuncture think about what they are doing. Ernst and White suggest that if it turns out that sham acupuncture is as effective as "real" acupuncture, research should concentrate on how to maximize the placebo effect and therapists should be trained in placebo acupuncture (p. 156)[6]. However, this seems to pose a problem for modern therapists, for whom the placebo effect still retains a certain element of charlatanry. Would it be ethical to deceive patients (in effect), even if it were for their own good? And, if not, would telling the patient that this was a placebo procedure negate its effectiveness? There seem to be serious problems in the deliberate use of placebo today, though the practice was widespread in medicine before the Second World War[10].

Perhaps, therefore, we should consider abandoning the use of the word "placebo" in this context. If we could establish some of the mechanisms that underlie the subjective phenomena produced by manual treatments, we might be able to offer a description that avoids the dreaded word "placebo" in favour of one that was both more accurate and more acceptable to both patients and therapists. We might, for example, find ourselves able to say something like this:

Our therapy is designed to help your brain to reprogram itself in a more healthy way. It takes advantage of the inbuilt healing propensities of the body, so rather than manipulating the body from outside we are working with the mechanisms that exist and encouraging them to take over and restore health.

This would be more than a verbal sleight of hand. It would not be simply a matter of relinquishing the "placebo" terminology; rather, it would direct the attention of scientists towards the subjective phenomena, in the hope that research on these could be integrated with research on the local tissue phenomena which at present are the main focus of scientific interest. Admittedly, these subjective phenomena are more difficult to research, which is one reason why they have been relatively neglected, but there seems to be no good reason why they should not be integrated into our general theoretical framework.


  1. Campbell A. A doctor's view of acupuncture: traditional Chinese theories are unnecessary. Complementary Therapies in Medicine 1998;6:152-155.
  2. White A, Neurophysiology of acupuncture analgesia. In: Ernst E, White A (eds) Acupuncture: a scientific appraisal. Lnodn, Butterworth-Heinemann, 1999.
  3. Baldry PE. Acupuncture, Trigger Points and Musculoskeletal Pain. Edinburgh: Churchill Livingstone, 1998.
  4. Nathan B. Touch and emotion in manual therapy. Edinburgh: Churchill Livingstone, 1999.
  5. Collier S, Phillips D, Camp V, Kirk A. The influence of attitudes to acupuncture on the outcome of treatment. Acupuncture in Medicine 1995;13:74-77.
  6. Ernst E, White A. Acupuncture: A Scientific Appraisal. London: Butterworth-Heinemann, 1999.
  7. ter Riet G, de Craen AJM, Kessels AGH. Is placebo analgesia mediated by endogenous opioids? A systematic review. Pain 1998;76:272-275.
  8. Campbell A. Cartesian dualism and the concept of medical placebos. Journal of Consciousness Studies 1994;1:230-233.
  9. Anon Editorial. Lancet 1994;343:681-682.
  10. Kaptchuk TJ. Powerful placebo: the dark side of the randomized controlled trial. Lancet 1994l351:722-725.
  11. Campbell A. The limbic system and emotion in relation to acupuncture. Acupuncture in Medicine 1999;17:124-130.