Components of placebo effect: randomised controlled trial in patients with irritable bowel syndrome
BMJ 2008; 336 doi: https://doi.org/10.1136/bmj.39524.439618.25 (Published 01 May 2008) Cite this as: BMJ 2008;336:999
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
We congratulate Kaptchuk et al. (1) on their outstanding report, which was written with considerable ingenuity. Using single-blind placebo acupuncture needles, Kaptchuk et al. assessed three components of placebo effects: assessment and observation, a therapeutic ritual (placebo treatment), and a supportive patient-practitioner relationship. The conclusion that the patient-practitioner relationship is the strongest component is very insightful for evaluating the benefit of complementary and alternative medicine. They treated patients with irritable bowel syndrome, who were allotted to the following treatment groups: waiting list, placebo acupuncture alone (limited) and placebo acupuncture with a defined positive patient-practitioner relationship (augmented). In this report, the patients were well masked, i.e. 76%-84% and 56%-84% of the patients believed that they received genuine acupuncture at 3 and 6 weeks of treatment, respectively. This implies that a large proportion of the patients perceived specific sensations associated with skin pressure during blunt tip needle application, making them believe that the treatment received was genuine.
We believe that the sensations elicited by the placebo or real needle had significant psycho-physiological impact on the patients in terms of therapy. Since the practitioners were well trained to adhere to the protocol throughout the experiment, there might be no significant difference in the sensations during placebo application between limited and augmented treatments. However, if this is not the case, the role of psycho-physiological impact by placebo application cannot be excluded in bringing about a greater improvement in the augmented group compared to the limited group. For the single-blind needle used in this study, the amount of skin pressure by the blunt tip could not be controlled mechanically or automatically and was dependent on the unmasked practitioner's discretion, which might have led to a possible bias. We believe that the skin pressure should be kept equal throughout the experiment, and if the patient-practitioner masking placebo needle is used (2), the practitioner bias is no longer a cause for concern.
(1) Kaptchuk TJ, Kelley JM, Conboy LA, Davis RB, Kerr CE, Jacobson EE et al. Components of placebo effect: randomised controlled trial in patients with irritable bowel syndrome. BMJ 2008; doi:10.1136/bmj.39524.439618.25:1�E.
(2) Takakura N, Yajima H. A double-blind placebo needle for acupuncture research. BMC Complement Altern Med 2007;7:31.
Competing interests: None declared
Competing interests: No competing interests
Our team is appreciative of all the thoughtful discussion this study has generated. We can offer a few remarks.
Dr. Ernst’s raises the issue of “social desirability” as a possible contributor to a placebo response. Indeed, we believe that this could a factor in the placebo response. Dr. Bracken argues that our “augmented” medical encounter is closest to an ‘art” form and involves interpretations of meaning not reducible to empirical study. We disagree and believe that it is essential “to be specific about non- specifics.”
In terms of both of these points, we performed a nested series of in-depth interviews with an additional 27 patients which we expect will provide data to address Dr. Ernst’s question on how patients bonded with researcher and with Dr. Braken’s concern with non-reducible questions of meanings. Dr. Heyland points out that our “augmented” arm may represent a form of psychodynamic-interpersonal therapy. To the extent it does represent this (or as Dr. van Duppen suggests a form of cognitive behavioral therapy), we would say that such a therapeutic relationship might be a valuable component of any positive healing encounter. Actually, we are not sure if it is right or not to consider the patient-physician relationship a part of the placebo effect (endless debates have not resolved this), but it is certainly a “non-specific” component of medical care. Our study suggests that such a supportive relationship can significantly modify the placebo response and contribute towards enhanced clinical outcomes.
We agree with Dr. Pearce that our “limited” arm was unreal and may have exaggerated the benefits of the “augmented” arm. But randomized trials often need to create somewhat unrealistic conditions (e.g., patient populations without co-morbidity) to demonstrate an effect. Also, it should be noted that our study was not designed to compare “limited” with “augmented” but rather determine whether three different non-specific effects could produce outcomes analogous to “dose-dependent.”
Both Dr. Thompson’s comments that a positive therapeutic relationship can engender a strong placebo effect and that physician need more face-to-face time with patients and Dr. Panasoff’s remarks that “physicians are placebo” and are helpful in the context of this entire discussion on whether our patient-practitioner relationship should or should not be considered a placebo effect. Whether our outcomes represent classical conditioning (Pearce) or expectancy (Heyland and Graz) or some combination of both is still unclear to many of our team and needs further study.
We appreciate Dr. Yun’s remarks on acupuncture. But from our perspective, our study has nothing to do with acupuncture. In order to study placebo effects, we used non-penetrating sham needles that scratched randomly selected non-acupuncture points. To our knowledge, mild scratching the skin for six or twelve sessions is unlikely to have any specific effects on digestion. However, the results of our sub-study comparing acupuncture to sham acupuncture will be reported elsewhere.
We agree with Dr. Julyan’s remarks that another arm of just patient-practitioner relationship would have been helpful. His remarks remind us that much work is required before we have a full and comprehensive understanding of placebo effects.
Sincerely,
Ted Kaptchuk
Harvard Medical School
Boston, Massachusetts 02215
Competing interests: None declared
Competing interests: No competing interests
While the study by Kaptchuk et al demonstrates very clearly the power of non-specific aspects of treatments for conditions such as Irritable Bowel Syndrome, it is questionable whether they are really justified in drawing the conclusions that they do.
Their main hypothesis is that the non-specific aspects of treatment can be theoretically and practically separated into three distinct components: the patient’s response to observation and assessment, their response to the application of a ‘therapeutic ritual’ (the placebo) and their response to the quality of the professional’s interaction with them. They wish to be specific about the non-specifics. They interpret their results as confirming their hypothesis. Their fundamental assumption is that by putting the randomised patients through the three different pathways they describe, they are effectively observing the differential impact of these different elements. However, their study does not control for the question of intensity of involvement with the healer. As well as involving ‘the successive addition of the three postulated elements of the non-specific clinical interaction’, the three arms of the study also involve different levels of time spent engaging with the practitioner. An alternative interpretation of their results would be that non-specifics factors are important and the more that ‘happens’ in treatment situations, the better. Perhaps, the greater the level of involvement with the practitioner, the greater is the degree of trust, confidence and expectation generated.
But there is a deeper assumption at work in this research: that the non-specific aspects of treatment response can be investigated with the same positivist tools that are applied in research on the specific (technical) aspects. The researchers are attempting to break the non- specific dimension of treatment into separate ‘variables’ that can be controlled in empirical studies such as this. This is questionable. Moerman (2002) argues that we should move from talking about the ‘placebo effect’ and instead speak of the ‘meaning response’. What we are dealing with when we study the non-specific aspects of healing are the ways in which medical encounters always involve negotiations around meanings: the meaning of pain, sickness, healing and sometimes death. What we are dealing with is really the ‘art of healing’ (Gadamer, 1996).
Understanding a piece of art is always primarily an act of interpretation and only secondarily something that involves empirical investigation. Appreciating Picasso’s Guernica involves looking at it as a whole and understanding the context (political, cultural and personal) in which it was produced. We can only understand the various elements of the painting by first grasping how it works as a whole. If the non-specific aspects of medical treatments are indeed something approaching an art, it is questionable how far empirical studies underscored by a logic of positivism will get us.
Gadamer HG. The Enigma of Health. Stanford: Stanford University Press, 1996.
Moreman D. Meaning, Medicine and the ‘Placebo Effect’. Cambridge: Cambridge University Press, 2002.
Pat.Bracken@hse.ie
Competing interests: None declared
Competing interests: No competing interests
Kaptchuk et al[1] make a stimulating contribution to our thinking about the effective elements of medical treatment in their study of patient response to three so-called placebo conditions – waiting list observation, sham acupuncture, and sham acupuncture plus practitioner interaction – for irritable bowel syndrome (IBS). We were not surprised to see that patients in the third arm of their study (those receiving ‘augmented interaction’) showed clinically significant improvements in several IBS-related domains. Why were we not surprised? First, because this ‘placebo’ condition permitted many of the non-specific factors of psychotherapy that are widely understood to be efficacious[2]. A closely related concept - the therapeutic alliance - has been consistently shown to be positively associated with clinical outcome, even for pharmacotherapy studies[3].
Second, although the authors attempted to control for psychotherapeutic factors by not allowing ‘specific cognitive and behavioural interventions that might be beneficial for irritable bowel syndrome’ in the ‘augmented interaction’ condition, they unfortunately made no reference to another form of evidence-based psychotherapy for IBS. Two substantial randomised controlled trials[4,5] have shown the effectiveness of psychodynamic-interpersonal (PI) therapy for IBS. This is a significant omission from the paper by Kaptchuk et al, as it means that PI interventions were not controlled for in the present study. In fact the permitted clinician behaviours in the augmented interaction group seem at face value to map closely onto interventions expected in PI therapy (eg exploring the patient’s causal attribution of their IBS, making links between IBS and relationships). It seems the authors were offering some of their subjects PI therapy without knowing it.
On a more general note, we would argue strongly that it is incorrect to label ‘patient-practitioner interaction’ as an element of the placebo effect at all. The term ‘placebo’ means that the treatment is inert, that it contains no active ingredient. Any resulting treatment effect is therefore due to expectancy on the part of the patient. This expectancy effect is what should be labelled placebo. Therefore to lump the patient- practitioner interaction in with expectancy effects and call them both placebo is wrong. In fact patient-practitioner interaction is the primary active ingredient of most if not all psychological treatment, whether for IBS or any other disorder, with or without a ‘therapeutic ritual’, as practised by general physician or specialist psychotherapist. The quality of this relationship has been shown to account for 30% of outcome variance, compared to just 15% for expectancy/placebo[6].
Simon Heyland, Specialist Registrar in Psychotherapy
simonheyland@doctors.org.uk
Jim Moorey, Consultant Clinical Psychologist
Gaskell House Psychotherapy Service, Manchester M13 0EU
Competing interests: JM trains and supervises clinicians practising psychodynamic-interpersonal therapy.
1. Kaptchuk TJ, Kelley JM, Conboy LA, Davis RB, Kerr CE, Jacobson EE, et al. Components of placebo effect: randomised controlled trial in patients with irritable bowel syndrome, BMJ 2008; 336: 999-1003.
2. Frank JD & Frank JB. Persuasion & Healing: a comparative study of psychotherapy. Baltimore: John Hopkins University Press, 1991.
3. Elkin I, Shea MT, Watkins JT, Imber SD, Sotsky SM, Collins JF et al NIMH Treatment of Depression Collaborative Research Program: general effectiveness of treatments. Arch Gen Psych 1989;46:971-82.
4. Creed FH, Fernandez L, Guthrie E, Palmer S, Ratcliffe J, Read N, et al. The cost-effectiveness of psychotherapy and paroxetine for severe irritable bowel syndrome. Gastroenterology 2003;124:303-17.
5. Guthrie E, Creed, Dawson D and Tomenson B. A controlled trial of psychological treatment for the irritable bowel syndrome. Gastroenterology 1991;100:450-57.
6. Lambert MJ. Psychotherapy outcome research: implications for integrative and eclectic therapists in Norcross JC and Goldfried MR (eds) Handbook of Psychotherapy Integration. New York: Basic Books, 1992.
Competing interests: JM trains and supervises clinicians practising psychodynamic-interpersonal therapy.
Competing interests: No competing interests
We congratulate Kaptchuck et al [1] on their important paper. We agree that a therapeutic relationship which engenders a strong placebo effect is the practitioner's most important tool in managing patients whose chronic symptoms have no apparent cause. In primary care, Thomas [2] found that the effectiveness of a doctor visit increased from 39 to 64%, when it included a diagnosis and a positive attitude, with or without a placebo pill. In another study [3], patients with non-cardiac chest pain who obtained reassurance from negative lab tests (though of no diagnostic value) were back to work faster, were more satisfied with care, and sought less help than controls.
In irritable bowel syndrome, we [4] have advocated a firm diagnosis accompanied by explanation and truly effective reassurance, such that lingering fears of serious disease are allayed. Psychosocial circumstances may impair a patient's ability to cope with symptoms, for exmple, an over- busy cell phone-dominated lifestyle can compete with basic biological activities such as eating, sleeping, defecation and even sex. Therapists’ greatest gift to such patients can be time spent exploring with them the implications of their lifestyle on their well-being – empathy in action.
If, in a clinical trial, therapeutic benefit = therapeutic gain from a treatment + disorder's natural history + placebo effect [5]; then, another possibility is: therapeutic loss = therapeutic gain from a treatment + disorder's natural history - nocebo effect.
In the former case, a harmful treatment (such as bloodletting) may achieve a net benefit if it is accompanied by a large placebo effect from the personality and reputation of the practitioner and the mystique of the procedure. In the latter case, a useful treatment such as dietary advice may be undermined by poor doctor/patient interaction. Healthcare systems that assign more value to technological procedures than to consultations and which provide inducements for rapid patient turnover are nocebos in this group of patients. Doctors need face-to-face time to bring to bear their personality, reputation, authority, and reassuring compassion. There is no need for dummy pills; the placebo is the doctor [5].
1. Kaptchuk TJ, Kelley JM, Conboy LA, Davis RB, Kerr CE, Jacobson EE, et al. Components of placebo effect: randomised controlled trial in patients with irritable bowel syndrome. BMJ 2008; doi: 10.1136/bmj.39524.439618.25
2. Thomas KB, General practice consultations: is there any point in being positive? BMJ 1987; 294:1200-2
3. Sox HCJr, Margulies I, Sox CH. Psychologically mediated effects of diagnostic tests Ann Int Med 1981; 95:680-5
4. Thompson WG, Heaton KW. Irritable Bowel Syndrome, 2nd edn. Health Press, Abingdon, 2003
5. Thompson WG The Placebo Effect in Health and Disease: Combining Science and Compassionate Care. Prometheus Press, Amherst NY, 2006
Competing interests: None declared
Competing interests: No competing interests
Editor
On the basis of their study on the placebo response in irritable bowel syndrome, Kaptchuk et al. conclude that "the patient-practitioner relationship is the most robust component" of the placebo effect [1].
Despite some significant limitations, including extremely brief follow-up and potential bias in patient recruitment, their findings fit with previous observations that the therapeutic relationship is correlated to beneficial outcomes [2].
However, the inclusion of another comparison group would have shed light on an important issue they do not disuss - how would patients respond to the augmented patient-practitioner relationship in the absence of sham acupuncture (or any other intervention)?
It is possible that the "doctor as drug" effect alone may be stronger than the study indicates [3]. Doctors often feel under pressure to "do something", when much of the time our patients may benefit most when we are free to just "be someone" - the one who helps them feel better.
1. Kaptchuk T.J., Kelley, J.M., Conboy, L.A., Davis, R.B., Kerr, C.E., Jacobson, E.E., et al. (2008) Components of placebo effect: randomised controlled trial in patients with irritable bowel syndrome. BMJ, 336, 999-1003.
2. Martin, D.J., Garske, J.P., & Davis, M.K. (2000) Relation of the therapeutic alliance with outcome and other variables: A meta-analytic review. Journal of Consulting and Clinical Psychology, 68(3), 438-450.
3. Balint, M. (2000) The Doctor, His Patient and The Illness. Churchill Livingstone, 2nd Edition.
Competing interests: None declared
Competing interests: No competing interests
This article just confirms one of the earliest teachings I got as a young first year medical students in the Buenos Aires faculty of Medicine : the physician is the first (and most important) part of the treatment that the patient recieves.Empathy,understanding and showing our patients that we really care for them as human beings and not as a "disease" help improve the chances of success.
Competing interests: None declared
Competing interests: No competing interests
Kaptchuk et al. are surely correct in asserting the important contribution of warmth, attention,empathy and confidence to the placebo effect they demonstrate. But the converse must also be true: witholding these items is likely to have negative or deleterious effects on symptoms. Unfortunately this negative element was introduced into group 2 patients who were told: "practitioners introduced themselves and stated they had reviewed the patient’s questionnaire and "knew what to do." They then explained that this was "a scientific study" for which they had been "instructed not to converse with patients." This is likely to have falsely exaggerated the benefits of group 3 patients . Their results are therefore difficult to interpret.
It is worth emphasizing that the commonly held view that a placebo response is determined by psychological genesis is false. The placebo effect is highly complex, but evidence suggests it is effected via organic, possibly neuro- humoral mechanisms. The great error is to regard responders as sham, or fake, or to interpret symptoms that do respond as being psychogenic or non-organic.
Competing interests: None declared
Competing interests: No competing interests
Kaptchuk et al. found that the most robust component of what they consider as a placebo effect is the patient-physician relationship. (1) Although this ‘augmented patient-physician relationship’ as applied in this study presupposes some basic skills of cognitive behavioural therapy (CBT). Questions like ‘how irritable bowel syndrome is related to relationships and lifestyle’ , ‘how the patient understood the "cause" and "meaning" of his or her condition’, as well as ‘ active listening’, ‘empathy’ or ‘communication of confidence and positive expectation’ as response to patients anxiety, negative perceptions of symptoms or catastrophically way of thinking are basic techniques and primary steps of CBT.
This study proves that these basic interventions, feasible in physicians daily consultations, can be effective for IBS and most probably for other functional somatic symptoms and syndromes. Training of physicians in applying these ICE (questioning ideas, concerns and expectations of the patient) techniques in their communication with patients can be very useful.
Because of the continuity in his relationship with the patient and the holistic approach the GP has strong opportunities for a positive patient- practitioner relationship. Therefore it would be useful that this kind of research would be done in family practice.
1. Kaptchuk TJ, Kelley JM, Conboy LA, Davis RB, Kerr CE, Jacobson EE, et al. Components of placebo effect: randomised controlled trial in patients with irritable bowel syndrome. BMJ 2008 Apr 3.
Competing interests: None declared
Competing interests: No competing interests
Which are the placebo effects - comment on Kaptchuk et al’s IBS placebo study
“A placebo can only be assumed to be inert according to current knowledge” [1]. The sham procedure used in Kaptchuk et al’s IBS placebo study [2] is already known not to be inert. One cannot touch the body without biological effects. Some of these effects may in theory be attributable to placebo, others are normal reactions to touch and can have many dimensions to them [3, 4]. Therefore the sham acupuncture which necessarily involves touch and pressure is not an inert placebo (something admitted by its proponents [5]) and cannot have effects solely attributable to the ritual of therapy [6] as the authors claim [7]. Thus placebo effects in both the sham treatment arms are necessarily overstated. Additionally some aspects of the effects of touch are probably specific to the acupuncture therapy [8], a possibility acknowledged by the lead author in recent discussions about the role of touch in taiji chuan [9, 10].
In these articles about taiji chuan the lead author also demonstrates knowledge of complex interventions and the difficulties of doing research on them. Among other things, the evidence he cites comes from acupuncture related studies showing how many aspects of patient-practitioner psycho- social-verbal interactions are specific aspects of acupuncture treatment [11]. This and other supporting studies have demonstrated the complex nature of acupuncture as an intervention [8, 12]. Unfortunately in the third arm of the IBS placebo study since sham acupuncture was used to investigate placebo effects, not only is there a problem with the sham not being inert, but the study will have attributed to placebo some effects due to these non-placebo related specific components of acupuncture intervention. There is no discussion of this and no attempt to tease apart placebo related treatment components from these acupuncture specific non- placebo related patient-practitioner interactional components [12]. Thus the study will necessarily have further overestimated placebo effects in this third arm, due to this mislabeling of treatment components.
This placebo study chose to use sham acupuncture as its ‘placebo’ treatment. This was an unfortunate choice. No sham acupuncture treatment has ever been demonstrated to be inert, raising questions about bias in acupuncture studies [1] and thus the suitability of sham acupuncture in trials of acupuncture [13]. Recently experts have raised the issue of whether there should be a moratorium on sham acupuncture studies due in part to these difficulties [14]. The authors of this study have chosen to ignore the same evidence and arguments about complex interventions and the inherent difficulty of separating their placebo effects that they have used and cited elsewhere [8, 9, 10], raising other questions about this placebo study.
It would have been much more interesting and relevant to answer the questions about placebo that this study attempted to investigate if they had chosen a sham (placebo) standard pharmaceutical intervention administered in normal GP practice where the doctor usually does not have time to talk much with the patient, and use as a third arm an extended discussion treatment arm added to the placebo medication.
References
1. De Craen AJM, Tijssen JGM, Kleijnen J. Is there a need to control the placebo in placebo controlled trials? Heart. 1997;77:95–96.
2. Kaptchuk TJ, Kelley JM, Conboy LA, Davis RB, Kerr CE, Jacobson EE, Kirsch I, Schnyer RN, Nam BH, Nguyen LT, Park M, Rivers AL, McManus C, Kokkotou E, Drossman DA, Goldman P, Lembo AJ. Components of placebo effect: randomized controlled trial in patients with irritable bowel syndrome. BMJ, 2008:336(7651):999-1003.
3. Fields T. Touch Therapy. London, Churchill Livingstone. 2000.
4. Leder D, Krucoff MW. The touch that heals: the uses and meanings of touch in the clinical encounter. J Alt Complem Med. 2008:14(3):321-327.
5. White AR. Acupuncture research methodology. In Lewith G, Jonas WB, Walach H, eds. Clinical Research in Complementary Therapies. Edinburgh: Churchill Livingstone, 2002:307–323.
6. Birch S. Comment on 'sham device v inert pill: randomised controlled trial of two placebo treatments. February 9, 2006 & Yes let’s get real: what placebo isn’t. March 10, 2006. http://www.bmj.com/cgi/eletters/332/7538/391#129658
7. Kaptchuk TJ, Stason WB, Davis RB, Legedza ART, Schnyer RN, Kerr CE, Stone DA, Nam BH, Kirsch I, Goldman RH. Sham device v inert pill: randomized controlled trial of two placebo treatments. BMJ 2006; 332:391- 397.
8. Schnyer R, Birch S, MacPherson H. Acupuncture practice as the foundation for clinical evaluation. In MacPherson H, Hammerschlag.R, Lewith G, Schnyer R (eds). Acupuncture Research: Strategies for Building an Evidence Base. London, Elsevier, 2007:153-179.
9. Wayne PM, Kaptchuk TJ. Challenges inherent in T’ai Chi research: Part I - t’ai chi as a complex multicomponent intervention. J Alt Complem Med. 2008:14(1):95-102.
10. Wayne PM, Kaptchuk TJ. Challenges inherent in T’ai Chi research: Part II - defining the intervention and optimal study design. J Alt Complem Med. 2008:14(2):191-197.
11. Paterson C, Dieppe P. Characteristic and incidental (placebo) effects in complex interventions such as acupuncture. BMJ, 2005:330:1202- 1205.
12. MacPherson H., Thorpe L, Thomas K. Beyond needling - therapeutic processes in acupuncture care: a qualitative study nested within a low back pain trial. J Alt Complem Med, 2006:12(9):883-880.
13. Birch S. A review and analysis of placebo treatments, placebo effects and placebo controls in trials of medical procedures when sham is not inert. J Alt Complem Med, 2006: 12(3):303-310.
14. Paterson C. The colonization of the lifeworld of acupuncture: The SAR conference. J Alt Complem Med. 2008:14(2):105-106.
Stephen Birch Foundation (Stichting) for the Study of Traditional East Asian Medicine (STEAM), Amsterdam, the Netherlands
Mark Bovey Coordinator, Acupuncture Research Resource Centre, Thames Valley University, London, UK
Competing interests: None declared
Competing interests: No competing interests