What Do RCTs Tell Us, and Could They Tell Us More? Looking Within and Beyond the Study Sample

Randomised controlled trials (RCTs) are generally regarded as the gold-standard method to answer questions of treatment effectiveness. Carefully designed and rigorously implemented, they can provide unbiased estimates of average treatment effectiveness in a wide range of clinical presentations. However, they have two potential drawbacks. First, owing to the variability of treatment effects at the level of individual patients, the aggregate findings of an RCT may be difficult to apply to a particular patient or type of patient. Subgroup analyses can produce more specific findings in this regard but have some shortcomings of their own and must be conducted, and reported, carefully. Second, while RCTs can normally claim a high degree of internal validity, their external validity can be more questionable. This results from the fact that participants in an RCT may not be representative of the target population to which their findings are intended to apply, and from differences between the experimental context in which RCTs are conducted and the world of everyday clinical practice. This chapter will discuss these difficulties and suggest ways in which they may be mitigated.