Abstract Details
Activity Number:
|
42
|
Type:
|
Contributed
|
Date/Time:
|
Sunday, August 3, 2014 : 2:00 PM to 3:50 PM
|
Sponsor:
|
Section on Statistics in Epidemiology
|
Abstract #313649
|
View Presentation
|
Title:
|
Estimating the Relationship Between a Transient Effect and the Onset of an Acute Event: A Comparison of a Case-Crossover Design and a Retrospective Cohort Design
|
Author(s):
|
Carlin Brickner*+ and Dirk Moore and Shou-En Lu and Kitaw Demissie and Robert J. Rosati
|
Companies:
|
Visiting Nurse Service of New York and Rutgers School of Public Health and Rutgers School of Public Health and Rutgers School of Public Health and CenterLight Healthcare
|
Keywords:
|
case crossover ;
simulation
|
Abstract:
|
The case-crossover design is an epidemiological method developed to estimate the transient effect of an exposure and an acute event, requiring only data on subjects with the event. Traditionally leveraged in primary data collection settings, the prominence of data warehouses enables researchers to observe exposure information on entire populations at little cost. The case-crossover design estimates the hazard ratio from a conditional likelihood comparable to the likelihood in a Cox Hazards analysis with time-dependent covariates: a method commonly applied in retrospective cohort designs. The case-crossover design assumes the baseline hazard is constant or small and the exposure distribution is globally exchangeable. In series of simulation studies, we generate a cohort from a hazard function with a time-dependent covariate and apply both designs. We show that the Cox Hazards analysis is superior in efficiency for all scenarios. However, in the presence of an unmeasured confounder, the case cross-over design is preferred because its estimates remain unbiased. An application studying medications changes and fall risk in a home health care population is presented.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2014 program
|
2014 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Professional Development program, please contact the Education Department.
The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Copyright © American Statistical Association.