|
Activity Number:
|
187
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, August 7, 2006 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #306047 |
|
Title:
|
Using Auxiliary Information in Clinical Trials
|
|
Author(s):
|
Shu Han*+ and Donald Berry
|
|
Companies:
|
Guidant Corporation and The University of Texas
|
|
Address:
|
1150 Cushing Circle, Saint Paul, MN, 55108,
|
|
Keywords:
|
auxiliary information ; Bayesian ; clinical trials ; nonparametric ; kernel density
|
|
Abstract:
|
We develop parametric and nonparametric models that utilize early endpoints in making treatment comparisons concerning the primary endpoint. Using all available information in this way enables earlier decisions about treatment benefits and futility. It also improves the efficiency of adaptive designs. Our parametric model is Bayesian and assumes the data are normally distributed. Our nonparametric model uses kernel density estimation. Our simulation results demonstrate that both parametric and nonparametric models have advantages over conventional methods. The parametric model performs slightly better than the nonparametric model when the distributions of the early and primary endpoints are bivariate normal. The nonparametric model is robust in that it performs better than the parametric model when these distributions deviate from normality sufficiently.
|
