Activity Number:
|
11
|
Type:
|
Topic Contributed
|
Date/Time:
|
Sunday, August 11, 2002 : 2:00 PM to 3:50 PM
|
Sponsor:
|
IMS
|
Abstract - #300686 |
Title:
|
Estimating Survival From Length-biased Data with Unknown Initiation Times
|
Author(s):
|
David Wolfson*+ and Xun Zhang
|
Affiliation(s):
|
McGill University and Montreal General Hospital
|
Address:
|
805 Sherbrooke Street West, Montreal, Quebec, H3A 2K6, Canada
|
Keywords:
|
length bias ; survival analysis ; density estimation
|
Abstract:
|
Survival data are sometimes collected by identifying subjects through a cross-sectional survey and following them until censoring or failure. It is well-known that the censored survival times are left truncated, termed-length biased if the onset times are "stationary." It is often the case in such studies that initiation dates are either unknown or known with uncertainty. When stationarity may be assumed we propose a method to estimate the unbiased survival function when faced with unknown initiation times. The forward recurrence times alone are used. The method may be adapted to estimation of the survival function, without conducting follow-up of the subjects following a cross-sectional survey. We illustrate our approach with survival data from patients with dementia, collected as part of the Canadian Study of Health and Aging.
|