|
Activity Number:
|
233
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 3, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #303907 |
|
Title:
|
Sample Size Determination in Clinical Trials with Multiple Co-Primary Binary Endpoints
|
|
Author(s):
|
Takashi Sozu*+ and Tomoyuki Sugimoto and Toshimitsu Hamasaki
|
|
Companies:
|
Osaka University and Osaka University Graduate School of Medicine and Osaka University Graduate School of Medicine
|
|
Address:
|
Center for Advanced Medical Engineering and Informatics, 2-2 Yamadaoka, Suita, Osaka, 565-0871, Japan
|
|
Keywords:
|
arcsine transformation ; association measures ; continuity correction ; correlated endpoints ; Fisher's exact method ; multivariate Bernoulli
|
|
Abstract:
|
Clinical trials often observe two or more primary efficacy endpoints. One of major problems in such trials is how to determine a sample size suitable for multiple co-primary correlated endpoints. We develop fundamental formulas of power and sample size calculation when multiple primary endpoints are given binary variables. Under three association measures between primary endpoints, we discuss five methods of sample size calculation; asymptotic normal with/without continuity correction, arcsine transformation with/without continuity correction and Fisher's exact method. For all five methods, the required sample size decreases with increases in the association measure, although the dependency is not large when the effect sizes are far different between endpoints or when the association measure is negative. The arcsine with continuity correction is closer to Fisher's exact method.
|
