Estimating high-dimensional heterogeneous treatment effects in the context of climate and health: a case study exploring the role of effect modification in the relationship between precipitation shocks and diarrheal illness

Abstract

A primary concern of public health researchers involves identifying and quantifying heterogeneous exposure effects across population subgroups. Understanding the magnitude and direction of these effects on a given scale provides researchers the ability to recommend policy prescriptions and assess the external validity of findings. Furthermore, increasing popularity in fields such as precision medicine that rely on accurate estimation of high-dimensional interaction effects has highlighted the importance of understanding effect modification. Traditional methods for effect measure modification analyses include parametric regression modeling with either stratified analyses and corresponding heterogeneity tests or including an interaction term in a multivariable model. However, these methods require manual model specification and are often impractical or not feasible to conduct by hand in high-dimensional settings. Recent developments in machine learning aim to solve this issue by automating heterogeneous subgroup identification and effect estimation. In this paper, we summarize and provide the intuition behind modern machine learning methods for effect measure modification analyses to serve as a reference for public health researchers. We discuss their implementation in R, provide annotated syntax and review available supplemental analysis tools by assessing the heterogeneous effects of drought on stunting among children in the Demographic and Health Survey data set as a case study.