Organic Indirect Effects: A Product Method for Binary Mediators, with Application to Selection of Potential HIV Cures into Clinical Trials

• Activity Number: 226 - Novel Interventional Approaches to Causal Mediation Analysis: Theory and Practice
• Type: Invited
• Date/Time: Wednesday, August 11, 2021 : 10:00 AM to 11:50 AM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Judith J. Lok* and Ronald Bosch
• Companies: Boston University and Center for Biostatistics in AIDS Research, Harvard TH Chan School of Public Health
• Keywords: mediation analysis; indirect and direct effects; causal inference; Baron and Kenny; product method; HIV/AIDS

Abstract

Baron and Kenny (1986) proposed estimators of indirect and direct effects: the part of a treatment effect that is mediated by a covariate and the part that is not. Subsequent work on natural indirect and direct effects provides a formal causal interpretation, based on cross-worlds counterfactuals: outcomes under treatment with the mediator set to its value without treatment. Organic indirect and direct effects (Lok 2016, 2020) avoid cross-worlds counterfactuals, using “organic” interventions on the mediator. We show that the product method holds in linear models for pure and for organic indirect effects relative to “no treatment” even if the outcome model has treatment-mediator interaction. Moreover, we find a product method for binary mediators. Furthermore, we argue that the organic indirect effect relative to “no treatment” is very relevant for drug development. We illustrate the benefits of our approach by estimating the organic indirect effect of curative HIV-treatments mediated by two HIV-persistence measures, using ART-interruption data without curative HIV-treatments combined with an estimated/hypothesized effect of the curative HIV-treatments on these mediators.