Placebos in chronic pain: evidence, theory, ethics, and use in clinical practiceBMJ 2020; 370 doi: https://doi.org/10.1136/bmj.m1668 (Published 20 July 2020) Cite this as: BMJ 2020;370:m1668
All rapid responses
We appreciate and thank Dr. Braillon for his response to our article, in which he raises important topics that we address here. The first of these points is regarding what constitutes a “clinically meaningful” effect in the field of pain research. Although this term lacks uniform consensus, many clinical trials have settled on a two-point reduction (on a scale of 10) or a 30% decrement in baseline pain as a definition of “clinically significant”. A more conservative definition is a 50% reduction, which enjoys widespread agreement but is likely too conservative; some of this debate summarized here. In examination of large placebo meta-analyses in pain (Table 1), we chose to use the term “clinically meaningful” based on the effect sizes of group-level placebo responses frequently exceeding 30% reduction in pain scales (or absolute reductions in pain usually between 1 and 2). We did not attempt to perform a quantitative umbrella analysis on these results due to the high heterogeneity in outcome measures. Individual authors of these meta-analyses frequently use language implying clinical efficacy such as “placebo provides effective treatment” or “placebo is moderately effective.” Within these group-level effects of course, some patients achieved large reductions in pain, whereas others small or none at all. We interpreted this evidence on the whole as therefore clinically meaningful in the context of the trial duration. Information on the long-term durability of placebo responses are currently limited, and represent an important area of ongoing placebo investigation.
Recent open-label placebo (OLP) research further supports clinically meaningful placebo effects. The patients in over ten randomized, controlled OLP trials so far published generally demonstrate a statistically significant and clinically meaningful effect superior to no-treatment controls. For example, in the first RCT of OLP for irritable bowel syndrome (IBS) patients on OLP had on average a reduction of over 92 points on the standard IBS questionnaire used to assess clinical improvement compared to 46 in the no-treatment group (p=0.03); 59% of the placebo group achieved (clinically meaningful) “adequate relief” compared to 35% of the no-treatment, a significant 24% difference (p=0.03). . In the first OLP study of chronic low back pain, besides patients besides reporting significant pain reduction , the study showed that 64% of patients reported reductions in their medication use, 34% reported no change and 1 participant (2%) reported increasing their pain medication . Granted, patients are not cured and we agree with Braillon that the effects of placebo are generally limited to subjective domains. However, we would argue that subjective improvement is nonetheless an important basis of clinical medicine. The next step -- predicting individualized patient experiences with placebo -- is an area requiring further research.
Psychotherapy, as specifically mentioned by Dr. Braillon, is an important treatment modality with the potential to improve affective and cognitive responses to pain. We did not address this area as we felt it beyond the scope of our review, but meta-analyses demonstrate modest effects sizes in pain reduction (0.27) compared to no-treatment controls . This effect size is on par with that of OLP; it would be desirable to have a head-to-head comparison, although the hypothesized mechanisms of improvement are different and these could be complementary in nature as well. We only discuss the patient-physician relationship in relationship to whether it can enhance placebo effects in the paper, but we agree with Dr. Braillon that clinicians are increasingly headed toward a likely erosion of skills and opportunities necessary to facilitate optimal patient-physician interactions. In our experience and to our knowledge, the clinical relationship established in published OLP randomized controlled trials (RCT) was closer to routine than optimal as we did not want the no-treatment controls to produce excessively high outcomes. Certainly OLP requires less patient effort than psychotherapy. The possibility of their combination also needs exploration.
Billing for a placebo --a third point raised by Braillon-- is straightforward. Components of placebo are inherently enmeshed in a therapeutic encounter, which is billed as one normally would for an interaction with a healthcare provider. Prescribing an inert placebo as a substance is clearly not yet widespread in practice, although as we stress in the paper, if placebo is prescribed transparently, without deception, and with full informed consent, it is considered an ethical intervention. Most pharmacies do not yet carry inert “placebo” pills; that said, we highlight the widespread practice of prescribing “impure” placebos, which are readily covered by insurance. For example, in my (CH) own field of multiple sclerosis, multiple medications are regularly prescribed for fatigue, none of which have ever convincingly achieved superiority to placebo in randomized comparisons (for example, modafinil, armodafinil, and other stimulants). The list of these off-label treatments for various medical conditions is vast, and regularly billed to insurance as part of good clinical care.
1. Rowbotham MC (2001) What is a “clinically meaningful” reduction in pain? Pain 94:131–132. https://doi.org/10.1016/S0304-3959(01)00371-2
2. Kaptchuk TJ, Friedlander E, Kelley JM, et al (2010) Placebos without deception: A randomized controlledtrial in irritable bowel syndrome. PLoS One 5:. https://doi.org/10.1371/journal.pone.0015591
3. Carvalho C, Caetano JM, Cunha L, et al (2016) Open-label placebo treatment in chronic low back pain: A randomized controlled trial. Pain. https://doi.org/10.1097/j.pain.0000000000000700
4. Carvalho C, Kirsch I, Kaptchuk TJ (2017) Reply. Pain 158:536–537. https://doi.org/10.1097/j.pain.0000000000000795
5. Williams AC de C, Eccleston C, Morley S (2012) Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database Syst Rev. https://doi.org/10.1002/14651858.CD007407.pub3
6. Miller P, Soundy A (2017) The pharmacological and non-pharmacological interventions for the management of fatigue related multiple sclerosis. J Neurol Sci 381:41–54. https://doi.org/10.1016/j.jns.2017.08.012
Competing interests: No competing interests
“Medicine is occasionally cure, often relieves, and always consoles.” Paré (1510-1590), a French barber surgeon.
Kaptchuk and colleagues carefully examined the effect of placebo treatment in chronic pain during three different conditions (double blind RCTs, deceptive placebo experiments, and open label placebo) to review theories of placebo effects.(1) However, their conclusion went further and deserves comment.
Firstly, could Kaptchuk and colleagues explain and document the “clinically meaningful” effect of placebo as expressed in their conclusion pledging for placebo use. Indeed, placebos do not have long-lasting or powerful objective clinical effects (quality of life, pain medication tapering …). The subjective patient reported alleviation is small, observed in only one third of the subjects and only under certain conditions (if you slip a placebo into a person’s drink, it does not work).(2)
Secondly, they rightly stressed “developing an optimal patient centered relationship is important” but wrongly ended with “regardless of whether it improves clinical outcomes” overlooking that psychotherapy (a term they overlooked) has been robustly evidence based for long.(3) Psychotherapy must focuses on the cognitive, affective, and behavioral components of the pain experience which is associated with psychosocial comorbidities. It helps patients to increase their feelings of control or feelings of self efficacy.(4) Sadly, very few patients with chronic pain have access to psychotherapy and a placebo cannot replace it. Further, I’m afraid that the term “optimal patient centered relationship” is flying in the face of real life practice. Are clinicians adequately trained for being open, listening, removing barriers, sharing authority, promoting active participation of the patient by the expression of his/her values and last but not least, analyzing if there is no misunderstanding by rephrasing (“reflective listening”)?(5) Are visits long enough for implementing these technical skills when accounting for the growing burden of administrative tasks?
Thirdly, could Kaptchuk and colleagues explain how to bill for a placebo?
1 Kaptchuk TJ, Hemond CC, Miller FG. Placebos in chronic pain: evidence, theory, ethics, and use in clinical practice. BMJ. 2020;370:m1668.
2 Braillon A. Placebo is far from benign: it is disease-mongering. Am J Bioeth. 2009;9:36-38.
3 McArdle JM, George WD, McArdle CS et al. Psychological support for patients undergoing breast cancer surgery: a randomised study. BMJ 1996;312:813-4.
4 Dalton JA, Coyne P. Cognitive-behavioral therapy: tailored to the individual. Nurs Clin North Am. 2003;38:465-vi
5 Braillon A, Taiebi F. Practicing "Reflective listening" is a mandatory prerequisite for empathy. Patient Educ Couns 2020. Online Apr 4. doi:10.1016/j.pec.2020.03.024
Competing interests: No competing interests