clusterrandomised

Cluster randomised, or cluster randomized trials, are studies in which groups of participants are randomly allocated to intervention or control arms rather than individuals. Clusters can be hospitals, clinics, schools, communities, or workplaces, and all members within a cluster typically receive the same allocation. This design is used when the intervention is delivered at the group level, when there is a high risk of contamination between participants in the same setting, or for logistical reasons.

Key design features include the number of clusters, the average cluster size, and the intraclass correlation coefficient (ICC), which measures how similar outcomes are within clusters. Power calculations must adjust for clustering using the design effect: DE = 1 + (m − 1) × ICC, where m is the average cluster size. Analyses must account for clustering, commonly through mixed-effects models or generalized estimating equations, to obtain valid standard errors and p-values.

Randomization can be performed at the cluster level and may be stratified or restricted to balance cluster-level covariates such as size or location. Blinding is often challenging in cluster trials. Ethical considerations include consent, as randomization occurs at the cluster level; individuals may provide consent for participation, while governance may require cluster-level authorization and, in some contexts, opt-out or waived individual consent.

Strengths include reduced contamination and feasibility for group-delivered interventions; limitations include the need for more clusters to achieve power, potential baseline imbalance with few clusters, and reliance on accurate ICC estimates. Reporting follows CONSORT extensions for cluster randomised trials. Cluster randomised designs are common in public health, education, and community-based interventions.