Evid Based Med 18:31 doi:10.1136/eb-2012-100665
  • Therapeutics
  • Randomised controlled trial

Mock treatments and real acupuncture affect osteoarthritic pain similarly: patients who believe in a treatment they are receiving have less pain

  1. M McDonough Suzanne
  1. Health and Rehabilitation Sciences Research Centre, Institute of Nursing and Health Research, School of Health Sciences, University of Ulster, Newtownabbey, Co Antrim, UK
  1. Correspondence to: Suzanne McDonough
    Room 1F118, Health and Rehabilitation Sciences Research Centre, Institute of Nursing and Health Research, School of Health Sciences, University of Ulster, Newtownabbey Co Antrim, BT37 0QB, UK; s.mcdonough{at}

Commentary on: [CrossRef][Medline][Web of Science]Google Scholar


There has been increased interest in explaining the placebo effect given findings of no difference between placebo and true needle acupuncture.1 ,2 A lack of difference may be explained by an ‘enhanced’ placebo effect given the ritual nature of acupuncture and it is known that the placebo effect can be altered by the therapeutic context (eg, by the device itself as well as patient and therapist–patient interactions).3


This elegant trial by White and colleagues is very timely. The design is innovative, using a mixed methods approach to explore therapeutic contextual factors that might explain the clinical benefits observed in recent placebo-controlled trials. It investigated the effect of (1) the type of placebo treatment, (2) the type of consultation and (3) the practitioner effect on pain in patients with knee and hip osteoarthritis. Patients were adults suffering moderate to severe chronic osteoarthritic pain from either the knee or hip, awaiting joint replacement surgery (N=221). The authors conducted a multifactorial randomised controlled trial involving three interventions: real acupuncture and two forms of simulated treatment, Streitberger non-penetrating needle and mock transcutaneous electrical stimulation (MES); two consultation types (empathetic and non-empathetic); and three practitioners. A nested qualitative component was key to contextualising the quantitative data.


This trial has sufficient power to detect differences in consultation type and treatment type on pain scores, but not practitioner effects. The results showed that all treatment groups improved over time with no statistically significant differences between the groups. However, those that believed that treatment was real recorded significantly lower pain scores. The type of consultation did not have a significant effect on pain following treatment, while practitioner 3 had a significant effect compared with the other two practitioners.


The clinical implications of this trial are that both placebo devices produced an equivalent analgesic effect, and although there was no pharmacological placebo in the current study, a recent review has shown that physical placebos (where MES was most commonly used) have larger effects than placebo pills.4 This clinical benefit seems to be key in confirming to patients that treatment was real, leading to greater confidence in scoring symptom improvement. The placebo effect could therefore be enhanced, as suggested by the authors, by helping patients to understand that placebo treatments can legitimately relieve their pain; equally lack of belief could potentially reduce the real effects of an unconvincing treatment.

Some caution is required around the finding of a practitioner effect, given that this comparison may have been underpowered in the main analysis (a possible type 1 error is noted by the authors). The use of only three practitioners limits the ability to generalise this finding more widely, despite the qualitative findings that add support for a practitioner effect. Greater clinical benefit was associated with a paternalistic male authority figure/expert in the field, and the potential influence of practitioners needs further investigation.

It is very difficult to interpret the findings with respect to empathy, because of the difficulties controlling for this in a clinical trial setting. It would appear from this study that the nature of the consultation did not influence treatment response; however, the potential nocebo effect of non-empathic consultations may have been blunted by the fact that patients were informed about this as part of the consent process. The issue of consent is problematic in placebo research, larger placebo effects have been observed in trials where patients were not informed about the possibility of a placebo,4 and other studies that have shown the benefit of enhanced consultations when consent was not required to different consultation types.5 It would be interesting to use a trial design with a more patient-centred approach whereby patients are only consented to the treatment they will receive, to establish how this influences the placebo response.6


Free Sample

This recent issue is free to all users to allow everyone the opportunity to see the full scope and typical content of EBM.
View free sample issue >>

Don't forget to sign up for content alerts so you keep up to date with all the articles as they are published.