Fig. 1From: A systematic review of simulation studies which compare existing statistical methods to account for non-compliance in randomised controlled trialsTaxonomy of methods to deal with non-compliance to the protocol in a RCT, adapted from the taxonomies of Alshreef et al. and Mostazir et al. [7, 8]. Methods were categorised as simple, principal stratification, G-methods or “Other”. Methods highlighted are ones that have been identified within papers included in this review. *ITT does not attempt to deal with non-compliance directly but is included here as a “do nothing” approachBack to article page