|
Activity Number:
|
193
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 10, 2004 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #301121 |
|
Title:
|
Statistical Models for Monitoring Enrollment and Projecting Patient Exposure in Ongoing Clinical Studies
|
|
Author(s):
|
Duane Snavely*+ and Guanghan F. Liu
|
|
Companies:
|
Merck & Co., Inc. and Merck & Co., Inc.
|
|
Address:
|
10 Sentry Parkway, BL 3-2, Blue Bell, PA, 19422,
|
|
Keywords:
|
enrollment ; patient-years ; Kaplan-Meier estimate ; piecewise linear model
|
|
Abstract:
|
During the enrollment period of a clinical study, the sponsor may monitor the patient enrollment process and estimate the time at which the study enrollment completes. Based on the estimated timeline, the sponsor can make preparations to close the patient screening and enrollment process at the optimal time point. The sponsor may also desire to "cut off" a study at a specific time for data analysis and reporting. The study may be cut off at the time when enough patient exposure data is accumulated to fulfill regulatory requirements. For this purpose, the study sponsor may estimate the total patient-years of exposure for a given future time point. We introduce a few statistical methods to project the cumulative patient enrollment curve and to project the patient-years of exposure at a given future time point. Survival models with piecewise linear regression are used to estimate the enrollment process and the probability of a patient staying in the study at given time point. Clinical trial data are used to illustrate that the methods have good flexibility to provide accurate estimation for the cumulative enrollment and for the total patient-years of exposure.
|
