Re: Does Acupuncture Have Additional Value to Standard Poststroke Motor Rehabilitation?
To the Editor:
The results of Sze et al1 are surprising because they show virtually no placebo effect of acupuncture. This intervention is regularly associated with powerful placebo effects. In fact, the most conclusive interpretation of the earlier studies is that the positive effect of acupuncture was to a large extent due to nonspecific effects of acupuncture.2,3⇓ The finding of a total absence of a placebo response in the study by Sze et al thus contradicts all previous trials. The authors do not address this contradiction, but I think it cries out for an explanation.
- ↵Sze FK, Wong E, Yi X, Woo J. Does acupuncture have additional value to standard poststroke motor rehabilitation? Stroke. 2002; 33: 186–194.
- ↵Johansson BB, Hacker E, von Arbin M, Britton M, Långström F, Terént A, Ursing D, Asplund K, for the Swedish Collaboration on Sensory Stimulation After Stroke. Acupuncture and transcutaneous nerve stimulation in stroke rehabilitation: a randomized, controlled trial. Stroke. 2001; 32: 707–713.
Prof Ernst’s letter first raised the issue of placebo effect of acupuncture and then questioned why our randomized controlled trial (RCT) did not even show such an effect.
I share his view on the placebo effect of acupuncture. Actually our recent meta-analysis (in submission to Stroke) has also revealed that the positive effect of acupuncture in the literature for poststroke motor recovery was likely due to placebo effect and/or selection bias. The placebo effect, or nonspecific effect, of an intervention like acupuncture could be related to the expectations of the patient, or due to interaction between the interventionist and the patient. Johansson et al’s study1 using an expectation scale demonstrated that the expectations of a stroke patient did not differ between the real and sham acupuncture groups. As for the interaction between the acupuncturist and the patient, it involves the quality of their relationship, the degree of trust, their culture, etc.2 Obviously, the level of interaction between the therapists and the patient in the control group is very important. If their interaction is as good as that between the acupuncturist and the patient, this placebo effect will be lost. I agree with the comment by Johansson et al1: “A control group receiving the same attention as the active treatment group seems essential in trials of specific rehabilitation interventions such as sensory stimulation.” In other words, if we can take into consideration in designing acupuncture RCTs the 2 factors, namely the expectations of the patient and the interaction between the therapists and the patient in the control group, the placebo effect may be avoided.
In our stroke rehabilitation unit where the RCT was carried out, all stroke patients receive good psychological and social support from a multidisciplinary team including therapists, nurses, and social workers. Therefore, the interaction between the therapists and the patient in the control group is as good as in the acupuncture group. This is the main reason why our RCT did not show a placebo effect of acupuncture, assuming the expectations of the patient in the 2 groups were the same.