Causal Effect Estimation with TMLE: Handling Missing Data and Near-Violations of Positivity

TL;DR

This paper assesses TMLE's effectiveness in estimating treatment effects with missing data and positivity violations, comparing various methods through simulations and real data analysis.

Contribution

It introduces a comprehensive comparison of missing data handling methods with TMLE, highlighting the robustness of complete case analysis and MI CART in different scenarios.

Findings

Complete case TMLE with outcome-missingness model shows lower bias.

MI with CART achieves lower RMSE and maintains coverage.

Trade-offs exist between bias reduction and confidence interval accuracy.

Abstract

We evaluate the performance of targeted maximum likelihood estimation (TMLE) for estimating the average treatment effect in missing data scenarios under varying levels of positivity violations. We employ model- and design-based simulations, with the latter using undersmoothed highly adaptive lasso on the 'WASH Benefits Bangladesh' dataset to mimic real-world complexities. Five missingness-directed acyclic graphs are considered, capturing common missing data mechanisms in epidemiological research, particularly in one-point exposure studies. These mechanisms include also not-at-random missingness in the exposure, outcome, and confounders. We compare eight missing data methods in conjunction with TMLE as the analysis method, distinguishing between non-multiple imputation (non-MI) and multiple imputation (MI) approaches. The MI approaches use both parametric and machine-learning models. Results show that non-MI methods, particularly complete cases with TMLE incorporating an outcome-missingness model, exhibit lower bias compared to all other evaluated missing data methods and greater robustness against positivity violations across. In Comparison MI with classification and regression trees (CART) achieve lower root mean squared error, while often maintaining nominal coverage rates. Our findings highlight the trade-offs between bias and coverage, and we recommend using complete cases with TMLE incorporating an outcome-missingness model for bias reduction and MI CART when accurate confidence intervals are the priority.