Comparison of Frameworks for Tipping Point Analyzes

• Activity Number: 607 - Recent Advances in Missing Data Methods: From Estimands to Assumptions for Primary and Sensitivity Analyses
• Type: Topic Contributed
• Date/Time: Thursday, August 3, 2017 : 8:30 AM to 10:20 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Achim Guettner* and Summer Xia
• Companies: Novartis and Novartis
• Keywords: missing data ; tipping point analyses ; ICH E9 addendum ; sensitivity analyses ; multiple imputation

Abstract

The final concept paper "Addendum to Statistical Principles for Clinical Trials on Choosing Appropriate Estimands and Defining Sensitivity Analyses in Clinical Trials" emphasizes the importance for having a framework on sensitivity analyses. Sensitivity analyses can cover the presentation of results of analyses in which different choices and departures of assumptions are considered. These analyses include assumptions made on missing data across the different endpoints. This can be addressed with tipping point analyses. Different frameworks for tipping point analyses will be presented and discussed. Some approaches allow the inclusion of covariates in the analyses. However, this inclusion adds on additional complexity to the analysis and interpretation of the results. In addition to the statistical properties, the ease of clinical interpretation also needs to be considered in the choice of the methodology. Simulations to compare the impact of covariates on the tipping point analysis were conducted and results will be presented. Further considerations of tipping point analyses relevant at study planning will become obvious.