|
Activity Number:
|
56
- Causal Inference
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 8, 2021 : 3:30 PM to 5:20 PM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
|
Abstract #318668
|
|
|
Title:
|
Estimation of Attributable Events in Stratified Case-Case Analyses
|
|
Author(s):
|
Sarah E. Hegarty* and Dylan S. Small and Pamela Shaw
|
|
Companies:
|
University of Pennsylvania and University of Pennsylvania and Kaiser Permanente Washington Health Research Institute
|
|
Keywords:
|
case-case analyses;
attributable events;
inverted Mantel-Haenszel;
Monte-Carlo simulations;
common odds ratio;
homogeneity
|
|
Abstract:
|
Surveillance databases contain readily available information about events ranging from disease occurrence to traffic accidents or firearms fatalities. Such databases typically record information on cases only, with no information on controls or the potential denominator of individuals at risk. Case-case analyses estimate treatment effects when only case data are available by classifying cases into two types, one of which the treatment is known to have no effect on. Previous work has demonstrated that the odds ratio of exposure between the two types of cases can have a causal interpretation when treatment assignment is random and established an approach for estimating the minimum attributable events due to a treatment. We extend existing methods to a setting where the treatment assignment is only independent of outcome after stratification. We compare the performance of multiple approaches for estimating the attributable events. We conduct numerical studies to determine the type I error, coverage and bias of the novel stratified methods and compare them to the unstratified approach. We further illustrate methods using surveillance data on traffic accident deaths.
|
Authors who are presenting talks have a * after their name.
