|
Activity Number:
|
368
- SPEED: Statistics for Biopharmaceutical Studies
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2018 : 11:35 AM to 12:20 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
|
Abstract #332596
|
|
|
Title:
|
Extended Rank Tests for Analyzing Recurrent Event Data
|
|
Author(s):
|
Qiang Zhao* and Mark Chang and Michael LaValley and Joseph M. Massaro and Bin Zhang and Kathryn Lunetta
|
|
Companies:
|
and Veristat and Boston University and Boston University and Seqirus and Boston University
|
|
Keywords:
|
bias;
informative number of events;
intra-class correlation;
rank tests;
recurrent events
|
|
Abstract:
|
In practice, rank tests are commonly used to compare survival curves. Recurrent events can be treated as a type of correlated survival data, which is frequently observed when an event of interest is non-fatal and subjects may experience multiple events during follow-up. In addition to the inter-subject heterogeneity in analyzing recurrent event data, Wang and Chang studied the bias in estimation of the marginal survival curve of recurrent event data and came up with an unbiased Kaplan-Meier-like estimator for recurrent event data. However, there were no corresponding rank tests to compare Wang and Chang's survival estimates between different groups. In this paper we extended three commonly used rank tests so that they can be used to compare the Kaplan-Meier-like survival estimates developed using Wang and Chang's method. We also studied the empirical power differences between our new method and Jung and Jeong's method for clustered survival data and found that with moderate/large inter-subject heterogeneity our new method has greater power.
|
Authors who are presenting talks have a * after their name.
