MethodsEffect Decomposition in the Presence of Treatment-induced Confounding A Regression-with-residuals ApproachWodtke, Geoffrey T.a; Zhou, XiangbAuthor Information From the aDepartment of Sociology, University of Chicago, Chicago, IL bDepartment of Sociology, Harvard University, Cambridge, MA. Submitted May 14, 2019; accepted January 23, 2020. Supported by a grant from the Social Sciences and Humanities Research Council of Canada (Grant No. 435-2018-0736). Disclosure: The authors report no conflicts of interest. Supplemental digital content is available through direct URL citations in the HTML and PDF versions of this article (www.epidem.com). The study is exempt from IRB review because it involves only minimal risk and anonymous secondary data. Replication files are available from the corresponding author upon request. Correspondence: Geoffrey T. Wodtke, Department of Sociology, University of Chicago, 1126 E. 59th Street, Chicago, IL 60637. E-mail: firstname.lastname@example.org. Epidemiology: May 2020 - Volume 31 - Issue 3 - p 369-375 doi: 10.1097/EDE.0000000000001168 Buy SDC Metrics Abstract Analyses of causal mediation are often complicated by treatment-induced confounders of the mediator–outcome relationship. In the presence of such confounders, the natural direct and indirect effects of treatment on the outcome, into which the total effect can be additively decomposed, are not identified. An alternative but similar set of effects, known as randomized intervention analogues to the natural direct effect (rNDE) and the natural indirect effect (rNIE), can still be identified in this situation, but existing estimators for these effects require a complicated weighting procedure that is difficult to use in practice. We introduce a new method for estimating the rNDE and rNIE that involves only a minor adaptation of the comparatively simple regression methods used to perform effect decomposition in the absence of treatment-induced confounding. It involves fitting (a) a generalized linear model for the conditional mean of the mediator given treatment and a set of baseline confounders and (b) a linear model for the conditional mean of the outcome given the treatment, mediator, baseline confounders, and a set of treatment-induced confounders that have been residualized with respect to the observed past. The rNDE and rNIE are simple functions of the parameters in these models when they are correctly specified and when there are no unobserved variables that confound the treatment–outcome, treatment–mediator, or mediator–outcome relationships. We illustrate the method by decomposing the effect of education on depression at midlife into components operating through income versus alternative factors. R and Stata packages are available for implementing the proposed method. Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved.