🤖 AI Summary
In clinical practice, relative risk (RR) is more clinically interpretable than absolute risk difference; however, mainstream heterogeneous treatment effect (HTE) methods—such as causal forests—rely on absolute risk differences for recursive partitioning, often overlooking RR heterogeneity. To address this, we propose the first RR-oriented variant of causal forest: a nonparametric node-splitting criterion grounded in generalized linear model comparisons, explicitly designed to maximize statistical power for detecting RR heterogeneity. Our method imposes no strong distributional assumptions and automatically identifies covariates—and their interaction structures—that drive RR heterogeneity. Simulation studies and empirical analyses demonstrate that the proposed framework substantially improves identification of clinically relevant subgroups, successfully capturing RR heterogeneity patterns missed by conventional causal forests. It achieves this while preserving computational feasibility, thereby enhancing both clinical interpretability and statistical power.
📝 Abstract
The estimation of heterogeneous treatment effects (HTE) across different subgroups in a population is of significant interest in clinical trial analysis. State-of-the-art HTE estimation methods, including causal forests (Wager and Athey, 2018), generally rely on recursive partitioning for non-parametric identification of relevant covariates and interactions. However, like many other methods in this area, causal forests partition subgroups based on differences in absolute risk. This can dilute statistical power by masking variability in the relative risk, which is often a more appropriate quantity of clinical interest. In this work, we propose and implement a methodology for modifying causal forests to target relative risk, using a novel node-splitting procedure based on exhaustive generalized linear model comparison. We present results that suggest relative risk causal forests can capture otherwise undetected sources of heterogeneity.
