Estimating causal effects in trials involving multitreatment arms subject to non-compliance: a Bayesian framework

Data analysis for randomized trials including multitreatment arms is often complicated by subjects who do not comply with their treatment assignment. We discuss here methods of estimating treatment efficacy for randomized trials involving multitreatment arms subject to non-compliance. One treatment effect of interest in the presence of non-compliance is the complier average causal effect, which is defined as the treatment effect for subjects who would comply regardless of the treatment assigned. Following the idea of principal stratification, we define principal compliance in trials with three treatment arms, extend the complier average causal effect and define causal estimands of interest in this setting. In addition, we discuss structural assumptions that are needed for estimation of causal effects and the identifiability problem that is inherent in this setting from both a Bayesian and a classical statistical perspective. We propose a likelihood-based framework that models potential outcomes in this setting and a Bayes procedure for statistical inference. We compare our method with a method-of-moments approach that was proposed by Cheng and Small in 2006 by using a hypothetical data set, and we further illustrate our approach with an application to a behavioural intervention study.