|
Activity Number:
|
75
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 6, 2006 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #306489 |
|
Title:
|
Stroke Clinical Trials and Response-Adaptive Randomization: an Ideal Match
|
|
Author(s):
|
Yuko Palesch*+ and Amy Bardeen and Renee Martin
|
|
Companies:
|
Medical University of South Carolina and Medical University of South Carolina and Medical University of South Carolina
|
|
Address:
|
Department of Biostatistics, Bioinformatics, and Epidemiology, Charleston, SC, 29425,
|
|
Keywords:
|
response-adaptive randomization ; acute stroke clinical trials
|
|
Abstract:
|
Generally, randomized clinical trials are conducted to benefit future patients, and response-adaptive randomization, such as the randomized-play-the-winner (RPW) method, would allow that as well as to help patients who participate in the trials. The RPW utilizes accumulating information and assigns the potentially better performing treatment to more subjects. Acute stroke trials prove ideal for this design due to their short follow-up time (24 hours) to obtain a reliable surrogate outcome measure. RPW was simulated with data from a placebo-controlled NINDS rt-PA Stroke Study in which rt-PA was shown to be effective. The RPW rule skewed the 50:50 treatment allocation to an average of 58:42 in favor of the rt-PA arm; hence 37% more subjects would receive the more successful treatment (rt-PA) under RPW. The statistical efficiency of RPW will be compared to the simple randomization method.
|
