Individualised Treatment Effects Estimation with Composite Treatments and Composite Outcomes

TL;DR

This paper introduces H-Learner, a hypernetwork-based method for estimating individual treatment effects involving complex composite treatments and outcomes, addressing data scarcity and enabling applications in diverse fields.

Contribution

The paper presents a novel hypernetwork approach, H-Learner, for ITE estimation with composite treatments and outcomes, improving data sharing and applicability in complex scenarios.

Findings

H-Learner outperforms existing methods in empirical tests.

Effective in handling binary and arbitrary composite treatments.

Addresses data scarcity in causal inference.

Abstract

Estimating individualised treatment effect (ITE) -- that is the causal effect of a set of variables (also called exposures, treatments, actions, policies, or interventions), referred to as \textit{composite treatments}, on a set of outcome variables of interest, referred to as \textit{composite outcomes}, for a unit from observational data -- remains a fundamental problem in causal inference with applications across disciplines, such as healthcare, economics, education, social science, marketing, and computer science. Previous work in causal machine learning for ITE estimation is limited to simple settings, like single treatments and single outcomes. This hinders their use in complex real-world scenarios; for example, consider studying the effect of different ICU interventions, such as beta-blockers and statins for a patient admitted for heart surgery, on different outcomes of interest such as atrial fibrillation and in-hospital mortality. The limited research into composite treatments and outcomes is primarily due to data scarcity for all treatments and outcomes. To address the above challenges, we propose a novel and innovative hypernetwork-based approach, called \emph{H-Learner}, to solve ITE estimation under composite treatments and composite outcomes, which tackles the data scarcity issue by dynamically sharing information across treatments and outcomes. Our empirical analysis with binary and arbitrary composite treatments and outcomes demonstrates the effectiveness of the proposed approach compared to existing methods.