Bias in identifying and recruiting participants in cluster randomised trials: what can be done?

Abstract

Blinded recruitment of participants presents particular challenges for cluster trials, but careful design can minimise the risk of selection bias