PPT Slide
- The methodology of cluster randomised trials is well understood and robust.
- A cRCT is the best choice if the intervention under study can only be delivered to a group (cluster) of individuals.
- A cRCT is an excellent choice if the intervention understudy provides maximum benefit when delivered to a group (cluster) of individuals.
- If the potential for ‘contamination’ exists is significant, a cRCT should be considered.
- Methodological reviews demonstrate that up to 75% of all cRCTs have inappropriate sample size calculations and 50% have inappropriate analyses.
Donner A, Klar N. Design and Analysis of Cluster Randomisation Trials in Health Research. Arnold:London, 2000.
Statistics in Medicine. 2001;20: 329 - entire issue.
www.EvidenceBased.net/talks