Handling Multivariable Missing Data in Causal Mediation Analysis Estimating Interventional Effects

2024-03-26

Epidemiology

AI Summary

This study addresses bias in causal mediation analysis induced by multivariate missing data, particularly in estimating indirect effects. We systematically evaluate the performance of multiple imputation (MI) under seven distinct missingness mechanisms, employing fully conditional specification (FCS) imputation and g-computation for effect estimation, while comparing MIBoot and BootMI for variance estimation. Our key contribution is the proposal and validation of a “substantive-model-compatible” FCS–BootMI joint strategy, which demonstrates superior bias–variance trade-offs. Bias is largest when missingness is driven by mediators, confounders, or the outcome itself; under MCAR, all methods yield nearly unbiased estimates except under severe model misspecification; and BootMI substantially reduces variance estimation bias relative to MIBoot. Simulation studies calibrated to the Victorian Adolescent Health Cohort provide methodological guidance for handling missing data in causal mediation analysis.

Technology Category

• Reasoning under Uncertainty: Causality
• Machine Learning: Causal Learning
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories

Application Category

• Social Networks and Social Media: Fairness and bias in social network and social media analysis
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs

Abstract

The interventional effects approach to causal mediation analysis is increasingly common in epidemiologic research given its potential to address policy-relevant questions about hypothetical mediator interventions. Multiple imputation is widely used for handling multivariable missing data in epidemiologic studies. However, guidance is lacking on best practices for using multiple imputation when estimating interventional mediation effects, specifically regarding the role of missingness mechanism in the performance of the method, how to appropriately specify the multiple imputation model when g-computation is used for effect estimation, and appropriate variance estimation. To address this gap, we conducted simulations based on the Victorian Adolescent Health Cohort Study. We considered seven missingness mechanisms, involving varying assumptions regarding the influence of an intermediate confounder, a mediator, and/or the outcome on missingness in key variables. We compared the performance of complete case analysis, six multiple imputation approaches by fully conditional specification, differing in how the imputation model was tailored, and a “substantive model compatible” multiple imputation-fully conditional specification approach. We evaluated MIBoot (multiple imputation, then bootstrap) and BootMI (bootstrap, then multiple imputation) approaches for variance estimation. All multiple imputation approaches, apart from those clearly diverging from best practice, yielded approximately unbiased estimates when none of the intermediate confounder, mediator, and outcome variables influenced missingness in any of these variables and nonnegligible bias otherwise. We observed the largest bias for interventional effects when each of the intermediate confounders, mediators, and outcomes influenced their own missingness. BootMI returned variance estimates with a smaller bias than MIBoot.

Problem

• Addressing missing data in causal mediation analysis for interventional effects
• Evaluating multiple imputation methods for mediator and confounder missingness
• Comparing variance estimation approaches in mediation analysis with missing data

Innovation

• Multiple imputation for missing data handling
• Substantive model compatible imputation approach
• BootMI for improved variance estimation