Abstract - Acupuncture for osteoarthritic pain: a randomized controlled trial evaluating the relative effects of real acupuncture and two placebos with respect to patient empathy, empowerment, and the practitioner on treatment outcome

Method: This two centre RCT involved three acupuncture practitioners delivering either mock electrical stimulation (MES), real acupuncture (RA) or a Streitberger needle (SN) for OA of the hip or knee. Our primary outcome was pain over the week following the course of treatment using a VAS with a significant result being a 20% improvement in pain. We assessed empowerment, empathy, needling sensation and patient belief in the intervention as well as quality of life. Patients allocated to these three interventions were randomised to received empathic (talking and caring) or nonempathic (just treatment) consultations to evaluate the impact of the therapeutic relationship on pain. The treatment was provided by the three acupuncturists who delivered consistent interventions for all three treatments. Eight 30min treatments occurred over 4 weeks with 1 week of initial baseline recording. The credibility of the three interventions was assessed.Results: A total of 1472 patients were approached, 221 were randomised and 209 completed the study. This is consistent with our initial sample size calculations so the study is adequately powered to detect clinically significant differences. Analysis (ANCOVA) of the differences between empathic and nonempathic consultations indicates that there are no significant differences between these two conditions. There is a nonsignificant trend suggesting that RA is most effective and MES the least effective, The only clinically and statistically significant predictors of outcome was the therapist (P=0.007 95% CI ?19.31 to ?2.51) providing the intervention and if patients thought the treatment was ‘real’ whatever actual treatment they had received (P=0.05), while gender age and affected joint were not related to outcome. Most patients thought their treatments were real (96% RA, 93% SN, 75% MES).Discussion: The main factor that appears to affect the outcomes of acupuncture treatment for OA pain is the practitioner delivering the treatment and to some extent the belief in the patient that they are receiving real acupuncture. There is no clinically significant difference between the three interventions. Kaptchuk et al.'s suggestion of a dose-related placebo response to an acupuncture-related ritual is not confirmed by our adequately powered study. This study also suggests we are able to deliver convincing acupuncture placebos/controls and that a ‘placebo’ needle (SN) appears as clinically effective as MES. This implies that an enhanced placebo response is not associated with acupuncture needling. This suggests we should consider that the most important factor influencing treatment outcome for pain may be related to the therapist and not the procedure they employ. This hypothesis will require further investigation across a number of different interventions but may be important for pain management.