Back to Agenda
Some Innovative Approaches to Handling Missing Data Problems in Clinical Trials
Session Chair(s)
Peiling Yang, PhD
Supervisory Mathematical Statistician
FDA, United States
In this session, innovative alternative designs will be proposed that may be applicable to trials in certain disease areas to mitigate missing data problems. With regard to sophisticated analyses, such as multiple imputation and pattern mixture model, which require simulating data sets to impute missing values, an illustration will be given as to how to pre-specify the computer algorithms and capture simulated data values in ADaM to enhance the traceability and reproducibility.
Learning Objective : Recognize the impact of missing data on the trial outcomes; Describe alternative approaches to mitigate dropouts; Explain how to pre-specify computer algorithms and capture simulated data values in ADaM.
Speaker(s)
Balaji Parameshvaran, MD
DIA, United States
Senior AMS Manager
Design Consideration to Drop Out Problem in Psychiatric Trials
Jinglin Zhong, PhD
FDA, United States
Mathematical Statistician, Office of Biostatistics, OTS, CDER
Academic Perspective
Sonia Davis, DrPH
RTI International, United States
Senior Research Statistician
Pre-specified, Traceable and Reproducible Multiple Imputation and Pattern Mixture Models Using ADaM and Define.XML
Mat D. Davis, PhD, MS
Teva, United States
Associate Director, Biostatistics
Have an account?