|
Activity Number:
|
104
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 4, 2008 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Health Policy Statistics
|
| Abstract - #302453 |
|
Title:
|
Longitudinal Data with Follow-Up Truncated by Death: Communicating a Match Between Analysis Method and Research Aims
|
|
Author(s):
|
Laura L. Johnson*+ and Brenda Kurland and Paula Diehr
|
|
Companies:
|
National Center for Complementary and Alternative Medicine and Fred Hutchinson Cancer Research Center and University of Washington
|
|
Address:
|
6707 Democracy Blvd, Bethesda, MD, 20892-5475,
|
|
Keywords:
|
longitudinal data ; truncation by death ; missing data ; generalized estimating equations ; random effects models
|
|
Abstract:
|
How to summarize trajectories of longitudinal data truncated by death not non-response? Unconditional models such as random effects models may implicitly impute data beyond the time of death. Fully conditional models stratify longitudinal response trajectories by time of death. Partly conditional models reflect the average response in survivors at a given time point, rather than individual trajectories. Joint models of survival and longitudinal response describe the evolving health status of the entire cohort. Researchers using longitudinal data should consider which method of accommodating deaths is consistent with research aims, and use analysis methods accordingly. While subtly different wording is used to describe the analyses, the results can vary dramatically, impacting clinical and public interpretation of an analysis.
|
