Causal effect on a target population: a sensitivity analysis to handle missing covariates

TL;DR

This paper develops a sensitivity analysis framework to assess the bias in causal effect estimates caused by missing covariates when generalizing RCT results to a target population, using Gaussian assumptions and semi-parametric models.

Contribution

It introduces a method to quantify bias due to missing covariates in causal inference, including analysis of covariate imputation and proxy substitution, with theoretical and empirical validation.

Findings

Bias can be explicitly computed under Gaussian assumptions.

Linear imputation of missing covariates offers no bias reduction.

Proxies can partially mitigate bias from missing covariates.

Abstract

Randomized Controlled Trials (RCTs) are often considered the gold standard for estimating causal effect, but they may lack external validity when the population eligible to the RCT is substantially different from the target population. Having at hand a sample of the target population of interest allows us to generalize the causal effect. Identifying the treatment effect in the target population requires covariates to capture all treatment effect modifiers that are shifted between the two sets. Standard estimators then use either weighting (IPSW), outcome modeling (G-formula), or combine the two in doubly robust approaches (AIPSW). However such covariates are often not available in both sets. In this paper, after proving L1-consistency of these three estimators, we compute the expected bias induced by a missing covariate, assuming a Gaussian distribution, a continuous outcome, and a semi-parametric model. Under this setting, we perform a sensitivity analysis for each missing covariate pattern and compute the sign of the expected bias. We also show that there is no gain in linearly imputing a partially-unobserved covariate. Finally we study the substitution of a missing covariate by a proxy. We illustrate all these results on simulations, as well as semi-synthetic benchmarks using data from the Tennessee Student/Teacher Achievement Ratio (STAR), and a real-world example from critical care medicine.