Activity Number:
|
396
|
Type:
|
Topic Contributed
|
Date/Time:
|
Thursday, August 15, 2002 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Biopharmaceutical Section*
|
Abstract - #300881 |
Title:
|
Mixed-effects Modeling and Simulation of Longitudinal Clinical Trials with Informative Dropouts
|
Author(s):
|
Ratna Ramaswamy*+ and S. Ouyang
|
Affiliation(s):
|
Johnson & Johnson Pharmaceutical R&D and Johnson & Johnson Pharmaceutical R&D
|
Address:
|
920 Route 202 South, Raritan, New Jersey, 08869, U.S.A
|
Keywords:
|
informative dropout ; modeling
|
Abstract:
|
In many therapeutical areas, analysis, design, and planning of longitudinal clinical trials are complicated by patient dropouts. These dropouts are usually informative in the sense that the probability of dropping out depends on the therapeutic outcome. Ignoring dropouts may cause large uncertainty of trial performance in planning future trials and unknown bias in the analysis results. Integrated modeling of symptom assessments and dropouts using mixed effects may provide a more realistic description of trial outcome. Through simulation the performance of future trials can be evaluated against competing trial designs. A case study is provided to illustrate the application.
|