Selecting an appropriate placebo for a trial of spinal manipulative therapy

Selecting an appropriate control group or placebo for randomised controlled trials of spinal manipulative therapy is essential to the final interpretation and usefulness of these studies. Prior to starting a randomised controlled trial of spinal manipulative therapy for acute low back pain we wanted to ensure that the placebo selected would be considered appropriate by experts in the field thereby making the results more likely to be accepted and more likely to influence clinical practice. We developed ten placebo techniques that aimed to mimic spinal manipulative therapy as closely as possible which, while not including the active component of spinal manipulative therapy, were still credible. This list of placebo techniques with detailed descriptions was sent to 25 experts in the field from Australia and New Zealand including both clinicians and academics. We asked the experts to rate whether they believed each technique was appropriate for use as a placebo in a trial of spinal manipulative therapy. Sixteen (64%) of the experts responded. There were extremely low levels of agreement between the experts on which placebos were appropriate (kappa = 0.05, 95% CI 0.01 to 0.10). For nine of the ten placebos at least one expert considered the placebo to include the active component of spinal manipulative therapy while at least one other expert believed the same placebo was not only not active but also not credible. The results of this study demonstrate the different views of experts on what constitutes an appropriate placebo for trials of spinal manipulative therapy. Different beliefs about what is the active component of spinal manipulative therapy appear to be responsible for much of the disagreement.