A Global Sensitivity Analysis Paradigm for Analyzing Randomized Clinical Trials with Informative Loss of Follow-up
View Presentation View Presentation
*Daniel Oscar Scharfstein, Johns Hopkins University 

Keywords: Missing Data, Sensitivity Analysis

Loss of follow-up in randomized trials is a pervasive problem, one which serves to jeopardize the scientific and regulatory inferences drawn from these studies. Any one analysis will rely on strong untestable assumptions and thus is inadequate for inference. FDA guidance calls for the use of sensitivity analysis to evaluate the robustness of the inferences to model assumptions. In this talk, we present a coherent, global sensitivity analysis paradigm for reporting inferences.