A generalized theory of separable effects in competing event settings

• Published in: Lifetime data analysis Vol. 27; no. 4; pp. 588 - 631
• Main Authors: Stensrud, Mats J., Hernán, Miguel A., Tchetgen Tchetgen, Eric J, Robins, James M., Didelez, Vanessa, Young, Jessica G.
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.10.2021; Springer Nature B.V
• Subjects: Blood pressure; Causality; Decomposition; Economics; Epidemiology; Finance; Health Sciences; Humans; Identification; Incidence; Insurance; Kidneys; Management; Mathematics and Statistics; Medicine; Operations Research/Decision Theory; Public health; Quality Control; Reliability; Safety and Risk; Statistics; Statistics for Business; Statistics for Life Sciences; Causal inference; Effect decomposition; Hazard functions; Competing events; Separable effects; G-formula
• ISSN: 1380-7870; 1572-9249; 1572-9249
• DOI: 10.1007/s10985-021-09530-8

In competing event settings, a counterfactual contrast of cause-specific cumulative incidences quantifies the total causal effect of a treatment on the event of interest. However, effects of treatment on the competing event may indirectly contribute to this total effect, complicating its interpretation. We previously proposed the separable effects to define direct and indirect effects of the treatment on the event of interest. This definition was given in a simple setting, where the treatment was decomposed into two components acting along two separate causal pathways. Here we generalize the notion of separable effects, allowing for interpretation, identification and estimation in a wide variety of settings. We propose and discuss a definition of separable effects that is applicable to general time-varying structures, where the separable effects can still be meaningfully interpreted as effects of modified treatments, even when they cannot be regarded as direct and indirect effects. For these settings we derive weaker conditions for identification of separable effects in studies where decomposed, or otherwise modified, treatments are not yet available; in particular, these conditions allow for time-varying common causes of the event of interest, the competing events and loss to follow-up. We also propose semi-parametric weighted estimators that are straightforward to implement. We stress that unlike previous definitions of direct and indirect effects, the separable effects can be subject to empirical scrutiny in future studies.