|
Activity Number:
|
65
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, July 29, 2007 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #308460 |
|
Title:
|
Small-Sample Tests for Efficacy in Fixed-Dose Drug Combination Studies
|
|
Author(s):
|
Melinda M. Holt*+ and Jesse Linder and Ferry Butar Butar
|
|
Companies:
|
Sam Houston State University and Southeastern Louisiana University and Sam Houston State University
|
|
Address:
|
Mathematics and Statistics, Huntsville, TX, 77341,
|
|
Keywords:
|
Binomial ; Clinical Trials ; Resampling ; Bootstrap ; Min Test
|
|
Abstract:
|
To be approved by the U.S. Food and Drug Administration, a drug combination must demonstrate efficacy; that is, it must be superior to each of its components. Here we consider fixed-dose clinical trials designed to show the efficacy of a combination drug when the response of interest is binary. Wang and Hung (1997) developed large-sample tests to demonstrate efficacy. Based on the results of a performance analysis, the authors recommended their tests for samples of size 20 or more and called for further study into small-sample methods. We present small-sample procedures based on the bootstrap and permutation methods. We also provide a performance analysis comparing the power and Type I error of the small-sample approaches to those of the large-sample approach.
|
