Sham in CAM

Several unique methodological challenges exist concerning the choice of the most appropriate comparator or control group for clinical research in complementary and alternative medicine (CAM). This article reviews the four major types of control groups (active control, placebo or sham, no intervention control, and different protocols of the experimental intervention), noting the different questions they answer and the different contexts in which they may be used, and discusses how this framework may be applied to CAM research in an effort to avoid methodologically flawed study designs. The choice of comparator depends on complex factors such as the research question being asked, the most plausible competing rival hypotheses, and logistical considerations related to ethics, methodology and feasibility. For example, for sham to be an effective comparator it must capture the ‘nonspecific’ elements of the treatment without containing those that are ‘specific’ to the research question at hand. However, as demonstrated by the examples of acupuncture and energy medicine, designing such sham intervention is not always easy. Controlling for social interaction between practitioners and subjects, addressing issues around ethics, having good-quality measurement of treatment integrity and intensity, and selecting practitioners of sufficient competence when doing a sham process are other important parts of correct design choices. In the light of these complexities, clinicians and researchers alike need a better match between the research questions and the design employed in order for their research to be both scientifically valid and clinically meaningful.