Sample Size Reestimation in Clinical Trials
- PDF / 918,351 Bytes
- 15 Pages / 504 x 720 pts Page_size
- 32 Downloads / 235 Views
Sample Size Reestimation in Clinical Trials Zhenming Shun Drug Information Journal 2001 35: 1409 DOI: 10.1177/009286150103500437 The online version of this article can be found at: http://dij.sagepub.com/content/35/4/1409
Published by: http://www.sagepublications.com
On behalf of:
Drug Information Association
Additional services and information for Drug Information Journal can be found at: Email Alerts: http://dij.sagepub.com/cgi/alerts Subscriptions: http://dij.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations: http://dij.sagepub.com/content/35/4/1409.refs.html
>> Version of Record - Oct 1, 2001 What is This?
Downloaded from dij.sagepub.com at UNIV OF UTAH SALT LAKE CITY on November 25, 2014
Drug Information Journal, Vol. 35, 1409-1422, 2001 Printed in the USA. All rights resewed.
0092-86 15/2001 Copyright 0 2001 Drug Information Association Inc.
SAMPLE SIZE REESTIMATION IN CLINICAL TRIALS ZHENMING SHUN Aventis Pharma, Bridgewater, New Jersey
This paper describes an approach to pegorming interim sample size reestimation based on the observed treatment difference in clinical trials. The approach combines the advantages of the group sequential and sample size reestimation methods and provides an efficient design for clinical trials. It provides flexibility but still maintains the integrity of the trial. To control the overall type I error level, a method is proposed to adjust the group sequential stopping boundaries adjusted for sample size reestimation and negative stops. The adjusted stopping boundaries are flexible to different rules of sample size reestimation and reuse the alpha values saved by negative stops. The adjustment is based on the exact type 1 error change and, therefore, the penalty for the type I error inflation due to such an interim reestimation is kept to a minimum. The eflciency of sample size reestimation without positive stops is compared with the group sequential method using unconditional power and expected sample size. All results are based on sufficient mathematical justifications. Key Words: Conditional power; Qpe I error; Efficiency; Group sequential method; Stochastic curtailment
INTRODUCTION THE SAMPLE SIZE of a clinical trial is a “reciprocal measurement” of the variability in the outcome of the trial: the more subjects that are planned, the more accurate the estimate of the treatment difference will be. We always hope the sample size is large enough, maybe just large enough, to see the treatment difference as we expected. How to size a clinical trial is usually a challenge to biostatisticians because the information (delta and standard deviation) used for sample size calculation may not be sufficient. In some cases, the sample size is determined by a so called “clinically meaningful minimal delta.” In such cases, the sample size can be poorly planned due to inaccuracy in the selected standard deviation. Some procedures have been developed to reestimate the sample size using
Data Loading...
