Efficacy in Longitudinal Psychosis Studies When Missing Mechanism Unknown

• Activity Number: 360
• Type: Contributed
• Date/Time: Tuesday, August 3, 2010 : 10:30 AM to 12:20 PM
• Sponsor: Biometrics Section
• Author(s): Jun Zhao*+ and Everton Rowe and Ram Suresh
• Companies: Merck & Co., Inc. and Merck & Co., Inc. and Merck & Co., Inc.
• Keywords: Drop-out ; Missing ; LOCF ; MMRM ; Time to Failure ; Time to Response

Abstract

In late phase of anti-psychotic drug development, we face high drop-out rate due to various reasons. This causes severe problem in analyzing longitudinal data we collected from studies. Traditionally we use parametric ANCOVA to analyze the data at study Endpoint using missing data imputation methods such as last observation carried forward (LOCF). Recently researchers and regulatory agencies are aware of the issues of missing data, and started using less biased mixed effect models such as MMRM to the longitudinal data collected from the trials. The approaches are still under questions due to unknown mechanism of missingness. In order to reduce the bias introduced by the missing data, we use non-parametric method to analyze the time to event data to compare the efficacy between treatment groups. Two events are defined in our research, the time to failure and time to (sustained) response.