|
Activity Number:
|
243
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #308162 |
|
Title:
|
Nonparametric Maximum Likelihood Estimation of the Incidence Rate Using Data from a Prevalent Cohort Study with Follow-Up
|
|
Author(s):
|
Vittorio Addona*+ and Masoud Asgharian and David Wolfson
|
|
Companies:
|
Macalester College and McGill University and McGill University
|
|
Address:
|
1600 Grand Avenue, Saint Paul, MN, 55105,
|
|
Keywords:
|
prevalent cohort ; right censoring ; left truncation ; incidence rate ; nonparametric maximum likelihood estimator
|
|
Abstract:
|
Prevalent cohort studies with follow-up often require fewer resources than incident cohort studies. We discuss nonparametric maximum likelihood estimation of a constant incidence rate that uses the well-known incidence-prevalence relationship. Efficient estimation of the incidence rate using this relationship, and based only on onset and failure/censoring times collected from a prevalent cohort study with follow-up, poses some major difficulties. In such studies, the onset times of prevalent cases are ascertained and subjects with shorter survival times are less likely to be recruited. We show how it is possible to adjust for the onset times of these "missing" cases when estimating the incidence rate. We apply our approach to data from the Canadian Study of Health and Aging to estimate age-specific incidence rates of dementia amongst elderly Canadians.
|
