|
Activity Number:
|
233
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 3, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #304119 |
|
Title:
|
NOCF! The future in Missing Data
|
|
Author(s):
|
Herbert Thijs*+
|
|
Companies:
|
Hasselt University
|
|
Address:
|
I-Biostat, Agoralaan 1, Diepenbeek, International, 3590, Belgium
|
|
Keywords:
|
BOCF ; Direct likelihood ; Incomplete data ; Pain clinical trials ; Efficacy ; Tolerability
|
|
Abstract:
|
In pain trials commonly used methods to analyze data subject to incompleteness are BOCF next to LOCF, BLOCF and others. The defense of those methods exists in the fact that with pain trials often a tolerability issue occurs which needs to be taken into account. From a statistical point of view however this cannot be tolerated. It can be shown easily that BOCF violates the information in the data tremendously and lead to biased results which can be both conservative and liberal. In this manuscript we state NO Carrying Forward (NOCF) as the strategy to deal with missing data in the future. We will shortly argue that different probabilities of dropout in different treatment arms actually is consistent with MCAR and we will provide a method to deal with tolerability issues by combining both efficacy and tolerability data in a direct likelihood approach.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2009 program |
