Randomisation method for cluster randomised trials with unbalanced cluster sizes
Nokuthaba Sibanda1, Francesca Storey2, Fiona Cram3, Stacie Geller4, Bev Lawton2
1. School of Mathematics and Statistics, Victoria University of Wellington
2. Centre for Women’s Health Research, Victoria University of Wellington
3. Katoa Ltd, Auckland
4. Department of Medicine, University of Illinois
Cluster-randomized clinical trials (CRT) are trials in which the unit of randomization is not a participant but a group or cluster (e.g. a hospital or a GP practice). They are suitable when the intervention applies naturally to the cluster (e.g. practice change); when lack of independence among participants may occur or when it is most ethical to apply an intervention to all within a group (e.g. school-level immunization). Because participants in the same cluster receive the same intervention, CRT may approximate clinical practice, and may produce generalizable findings. However, when not properly designed or interpreted, CRT may induce biased results.
CRT designs have features that add complexity to statistical estimation and inference. Key among these is the cluster-level correlation in response measurements induced by the randomization. An important consideration is the experimental unit of inference; often it is desirable to consider intervention effects at the level of the individual rather than the cluster. Finally, given that the number of clusters available may be limited and these clusters may differ in size, simple forms of randomization may not achieve balance between intervention and control arms at either the cluster- or participant-level. Failure to account for imbalance may induce bias and reduce validity. This presentation focuses on the complexities of randomization in the design of CRTs, such as the imbalances in size and covariate factors across clusters.