The control of confounding by intermediate variables

TLDR

In epidemiologic studies, crude exposure‑disease associations can be biased by confounding, and while prior work has focused on single‑timepoint exposures, longitudinal studies introduce time‑dependent covariates that may act as both confounders and intermediates. The study defines confounding for longitudinal data and introduces the extended standardized risk difference estimator to adjust for covariates that are simultaneously confounders and intermediates. The authors develop a formal definition of longitudinal confounding and derive the extended standardized risk difference estimator based on repeated measures of exposure, covariate, and outcome.

Abstract

Abstract In epidemiologic studies of the effect of an exposure on disease, the crude association of exposure with disease may fail to reflect a causal association due to confounding by one or more covariates. Most previous discussions of confounding in the epidemiologic literature have considered only point exposure studies, that is, studies that measure exposure and covariate status only once, at start of follow‐up. In this paper we offer definitions of confounding suitable for longitudinal studies that obtain data on exposure, covariate, and vital status at several points in time. An important difference between longitudinal studies and point exposure studies is that, in longitudinal studies, a time‐dependent covariate can be simultaneously a confounder and an intermediate variable on the causal pathway from exposure to disease. In this paper I propose an estimator, the extended standardized risk difference, that provides control for confounding by a covariate that is simultaneously a confounder and an intermediate variable.