|
Activity Number:
|
378
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 1, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #309809 |
|
Title:
|
A Revisit to Sample Size Computation
|
|
Author(s):
|
Ruoyong Yang*+
|
|
Companies:
|
Pfizer Inc.
|
|
Address:
|
235 E 42nd St, New York, NY, 10017,
|
|
Keywords:
|
sample size computation ; non-informative prior ; average power ; simulation ; conditional power ; sample size re-estimation
|
|
Abstract:
|
In phase III clinical trials, we regularly compute sample size based on previously observed treatment effect. In this paper, sample size computation is explored with consideration of uncertainties around estimated treatment effects and corresponding standard deviations. Bayesian framework is introduced to compute average power under standard non-informative priors. The numerical integrations are easily implemented with a simple SAS macro. Standard sample size computation is found over optimistic in most cases. Sample size required under average power framework is explored. Standard conditional power computation is also found over optimistic in most cases. Sample size re-estimation method is incorporated into the average power framework. Substantially different additional sample size is required in some cases, compared with standard sample size re-estimation method.
|
- 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 2007 program |
