Cluster randomised crossover designs: repeated inclusion of clusters to improve power

Often referred to as re-randomisation designs, randomised trials that allow participants to be included in a randomised clinical trial multiple times (being randomised independently each time), have been shown to increase trial recruitment rates. That is, in these studies, participants are included repeatedly. Provided certain assumptions are valid, treatment effect estimators from these repeated inclusion designs will be unbiased, and lead to more precise estimates of the treatment effect over designs without repeated inclusions of participants. The potential for such repeated inclusion of clusters in cluster randomised trials has not yet been explored. Re-inclusion of clusters may be useful when the number of available clusters is limited or when cluster recruitment is difficult: allowing some clusters to participate multiple times in the same trial. In this PhD project, the statistical theory of repeated inclusion cluster randomised trials, including longitudinal variants such as cluster randomised crossover designs, where clusters switch back and forth between control and intervention conditions, will be explored.