Assessing the effect of a teaching program on breast self-examination in a randomized trial with noncompliance and missing data

摘要

Many randomized studies suffer from noncompliance and missing data. We present an extended framework for the analysis of data from such experiments. We use an instrumental variables approach to link intention-to-treat effects to treatment effects and we adopt a Bayesian approach for inference and sensitivity analysis. This framework is illustrated in the context of a randomized trial of breast self-examination.