Propensity score methods for estimating relative risks in cluster randomized trials with low‐incidence binary outcomes and selection bias

• DOI: 10.1002/sim.6185
• Publication Date: 2014-04-28T04:09:57Z
• Authors: Clémence Leyrat; Agnès Caille; Allan Donner; Bruno Giraudeau
• Keywords: Inverse probability weighting

Abstract Despite randomization, selection bias may occur in cluster randomized trials. Classical multivariable regression usually allows for adjusting treatment effect estimates with unbalanced covariates. However, binary outcomes low incidence, such a method fail because of separation problems. This simulation study focused on the performance propensity score (PS)‐based methods to estimate relative risks from trials incidence. The results suggested that among different approaches used (multivariable regression, direct adjustment PS, inverse weighting and stratification PS), only PS fully corrected moreover had best statistical properties. Copyright © 2014 John Wiley & Sons, Ltd.