The impact of unmeasured confounding on causal inference in observational studies: A plasmode simulation study of targeted maximum likelihood estimation

Unmeasured confounding can cause considerable problems for causal inference in observational studies and threaten the validity of the estimates of causal treatment effects. We investigate the robustness of a relatively new causal inference technique, targeted maximum likelihood estimation (TMLE), in terms of its robustness against the impact of unmeasured confounders. We benchmarked TMLE's performance with the inverse probability of treatment weighting (IPW) method. We utilized a plasmode-like simulation based on variables and parameters from the Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments (SUPPORT). We evaluated the accuracy and precision of the estimated treatment effects. Though TMLE performed better in most of the scenarios considered, our simulation study results suggest that both methods performed reasonably well in estimating the marginal odds ratio in the presence of unmeasured confounding. Nonetheless, the only remedy to unobserved confounding when making causal inference is by controlling for as many as possible confounders because not even TMLE can provide a safeguard against bias from unmeasured confounders.