Missing Data in Clinical Trials:Tryng to Select Reasonable Approaches and Sorting Out Their Effect on Analysis

• Activity Number: 319
• Type: Topic Contributed
• Date/Time: Tuesday, August 5, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Ralph D'Agostino Sr.*+
• Companies: Boston University
• Keywords: missing data ; sensitivity analysis ; worst case situations ; imputations

Abstract

Often in Clinical Trials there are missing data problems ranging from trivial to substantial missing data. Also, often the mechanism(s) for the missing data is (are)unknown. In this presentation we attempt to present a number of real examples and then suggest and discuss possible way of dealing with the issue. These ways range from worst case situations, to simple imputations, and to more elaborate mathematical models. Emphasis is given to selecting the primary analysis and sensitivity analyses under various reasonable assumptions for the missing data. Recommendations and suggestions applicable to different settings are presented and discussed.