|
Activity Number:
|
285
- Weighting Methods and Mediation Analysis for Causal Inference
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 9, 2022 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
|
Abstract #322234
|
|
|
Title:
|
Mediation Analysis with Exposure Measurement Error Under Main Study/Validation Study Designs
|
|
Author(s):
|
Chao Cheng* and Donna Spiegelman and Fan Li
|
|
Companies:
|
Yale School of Public Health and Yale School of Public Health and Yale School of Public Health
|
|
Keywords:
|
Expected estimating equation;
Mediation proportion;
Natural indirect effect;
Regression calibration;
Validation study
|
|
Abstract:
|
The difference method is widely used in mediation analysis to quantify the extent to which a mediator may explain the mechanisms underlying the pathway between an exposure and a health outcome. In many health science studies, the exposures or treatments are often measured with error, which could lead to bias in the estimators provided by the difference method. The current work develops appropriate statistical approaches to account for bias caused by exposure measurement errors, when the outcome is either continuous or binary. We assume a main study/validation study design such that a set of validation samples are available to characterize the relationship between the true exposure and its error-prone counterparts. Extensive simulation studies were carried out to demonstrate the validity and efficiency of the proposed approaches in finite samples. We apply the proposed approaches to the Health Professional Follow-up study to investigate the impact of body mass index in mediating the effects of physical activity on the risk of cardiovascular diseases.
|
Authors who are presenting talks have a * after their name.
