|
Activity Number:
|
381
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 9, 2006 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #307634 |
|
Title:
|
Analyzing Change in Hazard for Time-to-Event Endpoints in Clinical Trials
|
|
Author(s):
|
Rafia Bhore*+ and Sandra Gardner
|
|
Companies:
|
U.S. Food and Drug Administration and Sunnybrook Health Sciences Centre
|
|
Address:
|
10903 New Hampshire Ave., Silver Spring, MD, 20993-0002,
|
|
Keywords:
|
change point ; hazard function ; survival ; time to event ; exponential distribution ; inference
|
|
Abstract:
|
Risk of a serious adverse event with long-term exposure to a drug may increase or decrease over time, or similarly, the efficacy of a drug may show changes in responses through time to onset of response, time to relapse of a disease, etc. Change-point methodology is relatively new in its application to clinical trials with censored survival data where the underlying hazard rate abruptly changes over time. This talk will review and compare likelihood-based methods in the literature for estimating change points in the hazard function. Through simulated data based on parametric survival models with change point, we show how to determine point estimates and interval estimates of the change point for different methods. Methods will be compared with respect to length and coverage of confidence intervals.
|
