The Design of Cluster Randomized Crossover Trials

Abstract

The inefficiency induced by between-cluster variation in cluster randomized trials can be reduced by implementing a crossover design. In a simple crossover trial each subject receives each treatment in random order. A powerful characteristic of this design is that each subject serves as its own control. In a cluster randomized crossover trial clusters of subjects are randomly allocated to a sequence of interventions. Under this design each subject is either included in only one of the treatment periods (crossover at cluster level) or in both periods (crossover at subject level). In this study the efficiency of both cluster randomized crossover trials relative to the cluster randomized trial without crossover is demonstrated. Furthermore, the optimal allocation of clusters and subjects given a fixed budget or desired power level is discussed.