Why does clustering matter in orthodontic trials?

Abstract

Clustering in RCTs occurs when participants or units are allocated to an intervention in a group rather than independently or when multiple measurements are taken from the same individual. Cluster RCTs occur frequently in clinical orthodontic research; however, only a quarter of published trials take account of the effects of clustering in the design and analysis of these trials. The effects of clustering needs to be considered when calculating the sample size required to detect a difference in treatment effect, obtaining consent for participation in the trial and finally the analysis of the data.