Effect Measure Modification by Covariates in Mediation: Extending Regression-Based Causal Mediation Analysis

Yi Li, Maya B Mathur, Daniel Solomon, Robert J. Glynn and Kazuki Yoshida
Additional contact information
Yi Li: Brigham and Women's Hospital

No 3gf64, OSF Preprints from Center for Open Science

Abstract: In this paper, we generalize the closed-form regression-based mediation analysis approach proposed by Valeri and VanderWeele (2013, 2015) to accommodate effect measure modification by the covariates. We show that covariate levels can affect the presence and magnitude of EMM of the conditional NDE and NIE, and that the dependence of the NDE and NIE depend on covariates is affected by the link functions of mediator and outcome models as well as the strength of EMM and of exposure-mediator interaction. Our proposed approach is implemented in R package regmedint (version 1.0.0), available at https://cran.r-project.org/web/packages/regmedint/index.html.

Date: 2022-04-05
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://osf.io/download/624c5ad867553806aa624c0e/

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:osf:osfxxx:3gf64

DOI: 10.31219/osf.io/3gf64

Access Statistics for this paper

More papers in OSF Preprints from Center for Open Science
Bibliographic data for series maintained by OSF ().