|
Activity Number:
|
280
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 4, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #303516 |
|
Title:
|
Statistical Methods for Interval-Censored Safety Events Based on Laboratory Data
|
|
Author(s):
|
Yan Zheng*+ and Meehyung Cho and William Stager and Gerard Derzko and Kaihong Jiang
|
|
Companies:
|
sanofi-aventis and sanofi-aventis and sanofi-aventis and sanofi-aventis and sanofi-aventis
|
|
Address:
|
200 Crossing Blvd, Bridgewater, NJ, 08807,
|
|
Keywords:
|
time-to-event distribution ; nonparametric ; maximum likelihood estimation ; uniform re-weighted Kaplan-Meier procedure ; empirical re-weighted Kaplan-Meier procedure
|
|
Abstract:
|
The lifetime data analysis of abnormal laboratory findings (for example, elevations in liver enzymes) requires addressing the issue of interval censoring. Since the laboratory assessments are performed only at preset times, the exact time for the first occurrence of a laboratory abnormality in question (defined as event) is only known to be in an interval between assessments. We review various analytical approaches found in the literature for interval censored data and develop two new methods which are extensions of Kaplan-Meier procedures. We then compare the new methods to commonly used approaches through simulations and applying to a motivating example for estimating time-to-event distributions, focusing on estimating the cumulative distribution function and hazard for data from a single treatment arm.
|
