|
Activity Number:
|
45
|
|
Type:
|
Invited
|
|
Date/Time:
|
Sunday, August 2, 2009 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #303099 |
|
Title:
|
Adjusting for Complex Sampling in Cohort Studies
|
|
Author(s):
|
Rebecca Betensky*+ and Matthew Austin
|
|
Companies:
|
Harvard School of Public Health and Harvard School of Public Health
|
|
Address:
|
Department of Biostatistics, Cambridge, MA, 02115,
|
|
Keywords:
|
truncation ; complex sampling
|
|
Abstract:
|
Observational studies often involve complex sampling, including multiple truncations requiring survival beyond a sequence of milestone events. For example, a study of risk factors for ALS sampled patients who had experienced certain risk factors prior to being diagnosed with ALS, and who were alive at study entry. A study of coronary heart disease (CHD) among diabetics sampled subjects from a cohort who were alive without CHD at an intermediary time for a blood draw and who had onset of diabetes at any timepoint. I will discuss the potential biases that can arise in standard analyses of these studies, and propose methods that avoid these through adjustment for the sequential truncations.
|
