Beyond Conditional Averages: Estimating The Individual Causal Effect Distribution

TL;DR

This paper introduces a method to estimate the distribution of individual causal effects from observational data under specific assumptions, using latent variable models, with a practical case study on liver disease and heart failure risk.

Contribution

It demonstrates that the ICE distribution is identifiable under certain independence assumptions and proposes flexible latent variable models for its estimation from cross-sectional data.

Findings

Estimated 20.6% of the population has a harmful effect greater than twice the average.

Provided a case study on Hepatic Steatosis and heart failure risk.

Validated the latent variable modeling approach in practical settings.

Abstract

In recent years, the field of causal inference from observational data has emerged rapidly. The literature has focused on (conditional) average causal effect estimation. When (remaining) variability of individual causal effects (ICEs) is considerable, average effects may be uninformative for an individual. The fundamental problem of causal inference precludes estimating the joint distribution of potential outcomes without making assumptions. In this work, we show that the ICE distribution is identifiable under (conditional) independence of the individual effect and the potential outcome under no exposure, in addition to the common assumptions of consistency, positivity, and conditional exchangeability. Moreover, we present a family of flexible latent variable models that can be used to study individual effect modification and estimate the ICE distribution from cross-sectional data. How such latent variable models can be applied and validated in practice is illustrated in a case study on the effect of Hepatic Steatosis on a clinical precursor to heart failure. Under the assumptions presented, we estimate that 20.6% (95% Bayesian credible interval: 8.9%, 33.6%) of the population has a harmful effect greater than twice the average causal effect.