All Times EDT
Keywords: Estimands, Causal Inference, Clinical Trial Simulation, ICH E9(R1), Intercurrent Events, Repeated Measures
Clinical trials are the gold standard for evaluating new pharmaceutical interventions. Although clinical trials are often designed with randomization and well-controlled protocols, complications will inevitably arise in the presence of intercurrent events (ICEs) such as treatment discontinuation. These can lead to missing outcome data based on the strategy adopted to handle the ICEs, and confound causal inferences when the missingness is a function of a latent stratification of patients defined by intermediate outcomes. Although established methods for clinical trials such as intention-to-treat (ITT) analyses may address causal questions, they might not be relevant in light of the ICEs. The pharmaceutical industry has been focused on developing new methods that can yield pertinent causal inferences in trials with ICEs. However, it is difficult to compare and contrast the properties of different methods developed in this endeavor. This is because real-life clinical trial data cannot be easily accessed or shared to provide benchmark datasets, and because different methods consider distinct assumptions for the underlying data generating mechanism. We develop a novel simulation model and corresponding Shiny application in R for clinical trials with ICEs that can directly evaluate different methodologies. Our simulator is named the Clinical Trials With Intercurrent Events Simulator (CITIES). It is formulated under the Rubin Causal Model, and the considered treatment effects account for ICEs in clinical trials with repeated measures. We illustrate the utility of CITIES via two case studies involving real-life clinical trials.