A Unified Framework for the Transportability of Population-Level Causal Measures

TL;DR

This paper introduces a unified framework for transporting various population-level causal effect measures across different populations, addressing a gap in existing methods that focus mainly on risk difference.

Contribution

It develops a comprehensive approach to transport multiple causal measures under covariate shift, including identification, estimation, and evaluation methods.

Findings

The framework applies to both absolute and relative effect measures.

Classical and semiparametric estimators are derived and compared.

Simulation and real-world data validate the proposed methods.

Abstract

Generalization methods offer a powerful solution to one of the key drawbacks of randomized controlled trials (RCTs): their limited representativeness. By enabling the transport of treatment effect estimates to target populations subject to distributional shifts, these methods are increasingly recognized as the future of meta-analysis, the current gold standard in evidence-based medicine. Yet most existing approaches focus on the risk difference, overlooking the diverse range of causal measures routinely reported in clinical research. Reporting multiple effect measures-both absolute (e.g., risk difference, number needed to treat) and relative (e.g., risk ratio, odds ratio)-is essential to ensure clinical relevance, policy utility, and interpretability across contexts. To address this gap, we propose a unified framework for transporting a broad class of first-moment population causal effect measures under covariate shift. We provide identification results under two conditional exchangeability assumptions, derive both classical and semiparametric estimators, and evaluate their performance through theoretical analysis, simulations, and real-world applications. Our analysis shows the specificity of different causal measures and thus the interest of studying them all: for instance, two common approaches (one-step, estimating equation) lead to similar estimators for the risk difference but to two distinct estimators for the odds ratio.