|
Activity Number:
|
521
|
|
Type:
|
Invited
|
|
Date/Time:
|
Thursday, August 2, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Statisticians in the Pharmaceutical Industry
|
| Abstract - #307734 |
|
Title:
|
Statistical Considerations in Planning Drug Clinical Trials
|
|
Author(s):
|
H.M. James Hung*+
|
|
Companies:
|
Food and Drug Administration
|
|
Address:
|
10903 New Hampshire Ave Bldg 22 Rm 4238, Silver Spring, MD, 20993-0002,
|
|
Keywords:
|
sample size ; study-wise type I error ; effect size ; reproducibility
|
|
Abstract:
|
The usual regulatory evidential requirement for efficacy of a test drug is that at least two clinical trials provide sufficient statistical evidence of the therapeutic effect. Statistical significance for the estimated drug effect is dependent on the planned sample size for detecting the postulated effect size with a certain level of statistical power. Reproducibility of a statistically significant finding needs to be factored into the planning of the two registration trials. The postulated effect size is related to the effect size estimated from other trials. Moreover, per the current standard, a typical registration trial requires that the overall study-wise type I error rate be no larger than 5%. In this paper, we will discuss the relationships between these factors and share our insights into the implications of these considerations on the trial planning.
|
