|
Activity Number:
|
249
|
|
Type:
|
Invited
|
|
Date/Time:
|
Tuesday, August 4, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
WNAR
|
| Abstract - #303181 |
|
Title:
|
Causal Inference for Randomized Clinical Trials with a Survival Outcome and All-or-None Treatment Noncompliance
|
|
Author(s):
|
Gang Li*+
|
|
Companies:
|
University of California, Los Angeles
|
|
Address:
|
2868 Nicada Drive, Los Angeles, CA, 90077,
|
|
Keywords:
|
Right censoring ; Survival analysis ; Kaplan-Meier Estimate ; Causal Inference
|
|
Abstract:
|
We study the problem of assessing treatment efficacy in a randomized clinical trial with a time-to-event outcome and all-or-none treatment noncompliance. Specifically, we develop a new semiparametric method to estimate and compare treatment and control survival probabilities among potential treatment compliers who are typically not identifiable in the control arm. Our new estimator of the survival function is always monotonically nondecreasing which is a property not possessed by a commonly used estimator in the literature. We illustrate our method using both simulated and real data.
|
