Causal treatment effect decompositions with time-to-event outcomes under competing events

2026-05-19

AI Summary

In time-to-event analyses with competing risks, causal interpretation of treatment effects becomes complicated when the treatment may influence the competing event. This work proposes the first quadruple causal decomposition framework, which disentangles the total effect of treatment on the event of interest into four mutually exclusive causal pathways, explicitly characterizing the interaction between treatment and the competing event. By introducing cross-world counterfactual estimands and leveraging standard identifiability assumptions—such as exchangeability and consistency—the framework enables nonparametric identification of these effects under randomized controlled trial data. Empirical application to two real-world RCT datasets demonstrates that the proposed approach effectively clarifies the causal mechanisms underlying treatment effects in the presence of competing risks, substantially enhancing interpretability of the results.

Abstract

Inference about treatment effects for time-to-event outcomes is often obscured by the presence of competing events. A particularly complex situation arises when the treatment influences the occurrence of the competing event. A comprehensive assessment should then account for different mechanisms by which the treatment and the competing event together produce the apparent treatment effect. Here, we propose a decomposition of the treatment's effect on the event of interest (target), characterising how it arises due to four distinct mechanisms involving both the target and competing events. Based on a causal model, the decomposition relies on cross-world estimands reflecting counterfactual scenarios in which the treatment affects the two events as if set to conflicting levels. We specify exchangeability and consistency assumptions under which the decomposition can be estimated from observed data. We discuss how the new decomposition reveals the role of the competing event and serves as a basis for defining causally interpretable estimands in the presence of competing events. Finally, we demonstrate the use of the four-way decomposition with datasets from two randomised trials.

Problem

• causal inference
• competing events
• treatment effect
• time-to-event outcomes
• effect decomposition

Innovation

• causal decomposition
• competing events
• time-to-event outcomes
• cross-world counterfactuals
• treatment effect mechanisms