|
Activity Number:
|
611
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 3, 2016 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
|
Abstract #319641
|
View Presentation
|
|
Title:
|
Handling Missing Data in Multiple-Attack Migraine Studies
|
|
Author(s):
|
Kaifeng Lu*
|
|
Companies:
|
Allergan
|
|
Keywords:
|
Consistency of effect ;
Missing data ;
Multiple imputation
|
|
Abstract:
|
We consider the missing data problem in the assessment of consistency of effect in multiple-attack studies in the acute treatment of migraine. A positive consistency response may be defined as relief or elimination of headache pain in at least three out of four treated migraine attacks per the EMA guidance on migraine treatment. Not all patients will have four migraine attacks observed during the treatment period due to the random nature of time to migraine attacks or patient dropouts. We use a simulation study to compare the performance of various approaches for the missing data problem in this setting. Multiple imputation approaches are shown to outperform ad-hoc approaches in terms of bias and efficiency.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2016 program
|
