|
Abstract:
|
As defined in the Framework for FDA’s Real-world Evidence Program, real-world evidence (RWE) is clinical evidence derived from analysis of real-world data (RWD). A big question for both pharmaceutical industries and regulatory agencies is: how do we derive RWE from analysis RWD? In this topic-contributed session, we break this big question into four sub-questions as four key steps to derive RWE from the analysis of RWD: (1) Estimand: What are we going to estimate? (2) Data: What are we using to estimate the estimand? (3) Inference: How are we estimating the estimand based on the data? (4) Interpretation: How are we interpreting the findings from the estimation of the estimand? This presentation is about the fourth sub-question. ICH E9(R1) states that sensitivity analysis is intent to explore the robustness of causal inferences from the main estimator to deviations from its underlying modelling assumptions and limitations in the data. In other words, the robustness of RWE is determined by the sensitivity analysis. In this presentation, we will describe some recent developments in conducting sensitivity analysis, which offer us new ways to interpret RWE.
|
