Missing Not at Random Models for Masked Clinical Trials with Dropouts

• Activity Number: 537
• Type: Invited
• Date/Time: Wednesday, August 6, 2014 : 2:00 PM to 3:50 PM
• Sponsor: ENAR
• Author(s): Shan Kang*+ and Roderick Little and Nikon Kaciroti
• Companies: University of Michigan and University of Michigan and University of Michigan
• Keywords: Clinical Trials ; Blinding ; Longitudinal Study ; Missing at Random ; Missing Not at Random ; TROPHY

Abstract

Missing data is an unavoidable problem in clinical trials. Most existing missing data approaches assume the missing data are missing at random (MAR, Little and Rubin 2002). However, the MAR assumption is very questionable when the real causes of missing data are not well known, and cannot be tested from observed data. We propose a specific missing not at random (MNAR) assumption, which we call masked MNAR (MMNAR), which may be more plausible than MAR for masked clinical trials. We formulate models for categorical and continuous data under this assumption and we further extend these models to longitudinal studies. Simulations are conducted to examine the finite sample performance of our methods and compare them with other methods. We also applied our methods to the TRial Of Preventing HYpertension (TROPHY) study.