|
Activity Number:
|
65
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, July 29, 2007 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #309226 |
|
Title:
|
Sample Size Determination for Binomial Endpoints Based on the Width of the Confidence Interval in Phase IIA Clinical Trials
|
|
Author(s):
|
Tai-Tsang Chen*+ and Vishwanath Iyer and Aparna B. Anderson
|
|
Companies:
|
Bristol-Myers Squibb Company and Novartis Healthcare Pvt. Ltd and Bristol-Myers Squibb Company
|
|
Address:
|
5 Research Parkway, Wallingford, CT, 06492,
|
|
Keywords:
|
Sample size ; Confidence interval ; phase II
|
|
Abstract:
|
Single-arm phase II clinical trials are usually undertaken to assess whether a new agent has sufficient activity and an acceptable safety profile to warrant further investigation. When the goal is to estimate the response rate, the design options are limited. We propose an approach to determine the sample size in a phase II study when the primary interest is to estimate a binary outcome. The sample size is chosen to ensure that among confidence intervals that contain the true parameter, the probability that the interval width is smaller than a prespecified threshold is sufficiently high. We will apply the sample size calculation to different methods of calculating confidence intervals and provide recommendations on how to choose the threshold value for the confidence interval width.
|
