|
Activity Number:
|
135
- Multiplicity, Missing Data and Other Topics
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 9, 2021 : 1:30 PM to 3:20 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
|
Abstract #317671
|
|
|
Title:
|
Exploring the Impact of Different Endpoint Definitions for MMRM and Related Missing Data Problems
|
|
Author(s):
|
Yaoyuan Vincent Tan* and Fengjuan Xuan
|
|
Companies:
|
Vertex Pharmaceuticals and Vertex Pharmaceuticals
|
|
Keywords:
|
COVID-19;
Longitudinal data;
Missing not at random
|
|
Abstract:
|
The mixed effect model for repeated measures (MMRM) is a commonly used method for the analysis of longitudinal continuous endpoints in clinical studies. It is often of interest to define the clinical endpoint /estimand under different scenarios and characterize their performance. In particular, the at time t versus average time t endpoint definition is of interest. In order to equip the study biostatistician with a tool to respond to such discussions, we derive the estimated treatment effect of using these different endpoint definitions and standard error where there is no or minimum missing data. During the Pandemic, when more missing data are expected, we propose to use simulation to determine the properties of these endpoints. Various different scenarios are explored including 1. Data contains multiple sources such as home or clinic assessed data and 2. missingness mechanism is missing not at random.
|
Authors who are presenting talks have a * after their name.
