Estimation of Controlled Direct Effects

Abstract

SUMMARY. When regression models adjust for mediators on the causal path from exposure to outcome, the regression coefficient of exposure is commonly viewed as a measure of the direct exposure effect. This interpretation can be misleading, even with a randomly assigned exposure. This is because adjustment for post-exposure measurements introduces bias whenever their association with outcome is confounded by more than just the exposure. By the same token, adjustment for such confounders stays problematic when these are themselves affected by the exposure. Robins (1999) accommodated this by introducing structural nested direct-effect models with direct effect parameters that can be estimated using inverse probability weighting by a conditional distribution of the mediator. The resulting estimators are consistent, but inefficient and can be extremely unstable when the intermediate variable is absolutely continuous. In this paper, we develop direct effect estimators which are not only more efficient, but also consistent under a less demanding model for a conditional expectation of the outcome. We find the one estimator which avoids inverse probability weighting altoghether to perform best. This estimator is intuitive, computationally straightforward and, as

Mediation Analysis: A Practitioner's Guide

“…Numerous other methods have been developed (1,2,9,13,24,26,35,58), which we could not address in this article. Some of these are discussed in the book length treatment of mediation (47).…”

How to use marginal structural models in randomized trials to estimate the natural direct and indirect effects of therapies mediated by causal intermediates

“…Moreover, the difference between a total effect and a natural direct effect can be interpreted as an indirect effect but not a controlled direct effect, in general [6][7][8][9]. In recent years, there have been many discussions about the conditions to identify a direct and indirect effect, especially in an epidemiologic research, and some analytical methods for this identification have been properly developed on the basis of counterfactual definitions [10][11][12][13].…”

Within-Subject Mediation Analysis in AB/BA Crossover Designs

“…If violated, we end up with time-dependent or intermediate confounding. It remains to be investigated how techniques such as inverse probability weighting [46] or Gestimation [47], that can deal with intermediate confounding concerning the estimation of the controlled direct effect in single-level settings, could be applied to multilevel settings.…”