|
Activity Number:
|
233
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 3, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #303504 |
|
Title:
|
Beta-Mapping and Beta-Regression for Changes of Ordinal-Rating Measures on Likert Scales: A Comparison Among Multiple Treatment Groups
|
|
Author(s):
|
Kelly H. Zou*+ and Martin O. Carlsson
|
|
Companies:
|
Pfizer Inc. and Pfizer Inc.
|
|
Address:
|
235 East 42nd Street, Mail Stop 685/10/08, New York, NY, 10017,
|
|
Keywords:
|
Ordinal Rating Data ; Likert Scale ; Beta Distribution ; Beta Regression Analysis ; General Linear Mixed Model ; Likelihood Ratio Test
|
|
Abstract:
|
Observer evaluative or patient reported outcomes in clinical trials and post-hoc analyses often use instruments that measure responses on ordinal-rating or Likert scales. We proposed a flexible distributional approach by modeling all possible change scores from the baseline to the end point using independent beta distributions. The shape parameters of the fitted beta distributions were estimated by direct moment-matching or by binning. Covariates and the interaction terms were included in multivariate beta regression analyses under generalized linear mixed models. These methods were illustrated on the treatment satisfaction data in an overactive bladder drug study with four treatment arms. Monte-Carlo simulations were conducted to compare the Type 1 errors and statistical powers using a beta likelihood ratio test against its fully nonparametric or parametric competitors.
|
