|
Activity Number:
|
35
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 6, 2006 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
| Abstract - #307518 |
|
Title:
|
On Informative Detection Bias in Screening Studies
|
|
Author(s):
|
Arvid Sjölander*+ and Juni Palmgren
|
|
Companies:
|
Karolinska Institutet and Karolinska Institutet
|
|
Address:
|
Department of Medical Epidemiology and Biostatistics, Stockholm, 17177, Sweden
|
|
Keywords:
|
informative detection bias ; directed acyclic graph ; counterfactuals ; sensitivity analysis ; bounds
|
|
Abstract:
|
A problem which has drawn some attention in the causal literature is to estimate the causal effect of hormone replacement therapy (HRT) on breast cancer. Since HRT is considered a risk factor for breast cancer HRT users are more likely to have a mammogram taken than non users. The observed discrepancy in cancer prevalence between users and non users is thus likely to suffer from detection bias. Standard statistical control for this bias, such as stratification on screening pattern, will severe the bias whenever screening pattern and disease status is confounded. We present the problem within the standard counterfactual framework. A sensitivity analysis is performed and bounds on the causal HRT effect are derived. We compare this method with the standard statistical approach.
|
