Activity Number:
|
45
- Nonparametric and Semiparametric Modeling for Complex Lifetime Data
|
Type:
|
Contributed
|
Date/Time:
|
Monday, August 3, 2020 : 10:00 AM to 2:00 PM
|
Sponsor:
|
Lifetime Data Science Section
|
Abstract #313809
|
|
Title:
|
Non-Parametric Estimation in an Illness-Death Model with Component-Wise Censoring
|
Author(s):
|
Anne Eaton* and Yifei Sun and Jim Neaton and Xianghua Luo
|
Companies:
|
University of Minnesota and Mailman School of Public Health, Columbia University and University of Minnesota and University of Minnesota
|
Keywords:
|
composite endpoint;
survival analysis;
non-parametric;
kernel smoothing;
censoring
|
Abstract:
|
When a composite endpoint consists of death and a non-fatal event, if the non-fatal event can only be detected at clinic visits, the resulting composite endpoint exhibits "component-wise censoring". The method recommended by the FDA to estimate event-free survival for this type of data fails to account for component-wise censoring. We apply a kernel method previously proposed for a marker process in a novel way to produce a non-parametric estimator that accounts for component-wise censoring. The key insight that allows us to apply this method is thinking of non-fatal event status as an intermittently observed, binary marker variable rather than thinking of time to the non-fatal event as interval censored. We also obtain estimates of the probability of being alive with the non-fatal event, and the restricted mean time patients spend in disease states. The method can be used in the setting of reversible non-fatal events. We perform a simulation study to compare our method to existing multistate survival methods and apply the methods on data from a large randomized trial studying interventions for reducing the risk of coronary heart disease.
|
Authors who are presenting talks have a * after their name.