|
Activity Number:
|
328
- New Guidance on Specifying the Target Difference (Effect Size) for a Randomised Controlled Trial
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Tuesday, August 1, 2017 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
|
Abstract #323728
|
View Presentation
|
|
Title:
|
The Impact of the Treatment Effect on Sample Size Estimation
|
|
Author(s):
|
Jerry Hintze*
|
|
Companies:
|
NCSS
|
|
Keywords:
|
effect size ;
sample size ;
power analysis ;
difference to detect ;
type I error ;
type II error
|
|
Abstract:
|
When designing a clinical trial an important step is the estimation of the sample size. In turn an important step in the sample size estimation is the quantification of the effect size. This is defined as the clinically important difference which would enable the conclusion of a treatment effect. There are other factors in the sample size calculation such as the treatment allocation, anticipated population variance (for a continuous outcome) and the Type I and Type II error. These are in the main determined by: precedent or development need (in the case of effect size); a quantified estimate from a past trial (the variance estimate) or the regulatory or other requirements (Type I and Type II error) The effect size has a subjective element in its quantification and yet the sample size estimation is critically dependent on it. If you half the effect size you will quadruple the sample size. This talk will demonstrate the sensitivity of sample size calculations to the effect size and highlight the need for guidelines on the setting of the effect size when designing a trial.
|
Authors who are presenting talks have a * after their name.
