Background
When performing clinical trials involving acupuncture, it can be difficult to choose a placebo control.
Objective
To validate the Park sham needle for use as a double-blind control intervention.
Design, setting, participants and interventions
Two different methods were employed. First, a researcher blinded to needle type administered the Park sham or a real needle, chosen at random, on the arms of 16 healthy volunteers. The researcher and the volunteers independently recorded which needle type they thought had been applied at each acupuncture point. Second, 19 patients with shoulder impingement syndrome were randomly assigned to receive acupuncture applied with the real needle or the Park sham needle, once a week for 6 weeks alongside a course of physiotherapy. At the end of the 6 sessions the patients recorded which needle type they thought was being used. The results were analyzed with a Fisher's exact test. This study was carried out in the Outpatient Department of Physiotherapy in Charing Cross Hospital, London, UK, on healthy volunteers and patients with shoulder impingement pain. The age range was 23 to 54 and 22 to 74 years respectively.
Results
Of the healthy volunteers, there was no difference (P = 0.23) between the number of needles that were correctly (n = 43) or incorrectly identified (n = 53). All patients thought that they had received the real needles. The researcher correctly identified all needles that were applied.
Conclusion
The researcher delivering the acupuncture recognized the needle type. However, both healthy and patient volunteers were blind to the needle type. This demonstrates that the Park sham needle is an effective single-blind control. It should be noted that the number of patients recruited was small and the study was underpowered to detect an effect of treatment.

Key words: Acupuncture, Placebos, Shoulder impingement syndrome, Validation studies as topic, Pilot projects