|
Activity Number:
|
14
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Sunday, July 29, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #308964 |
|
Title:
|
Bayesian Forecasting of Multicenter Enrollment
|
|
Author(s):
|
Byron Jones*+
|
|
Companies:
|
Pfizer Inc.
|
|
Address:
|
Pfizer IPC 193, Sandwich, Kent, International, CT13 9NJ, United Kingdom
|
|
Keywords:
|
multicenter ; enrollment ; Bayesian ; prediction ; updating ; decision
|
|
Abstract:
|
A key design element of a multicenter trial is the number of centers and the rate at which each center enrolls patients. During the trial these rates are monitored to determine if the trial will complete enrollment on time. If enrollment is going too slowly then either more centers need be added or the rates of the existing centers increased. We describe (i) a Bayesian approach to monitoring enrollment and forecasting future rates and (ii) a decision scheme for deciding whether to add or drop centers or increase current enrollment rates.
|
