An Open Path to DAG: Navigating Causal Inference in Epidemiological Research.

Abstract

Directed acyclic graphs (DAGs) are a valuable tool in epidemiology for illustrating causal relationships between variables in epidemiological research. DAGs enhance the transparency and robustness of the causal inference by delineating causal paths and identifying confounders, mediators, and colliders. It helps the researcher to identify the minimal number of variables that need to be adjusted to eliminate the effect of confounders on the causal relationships. However, this is not being widely used in epidemiological research, and when used, there is considerable variation in the way it is utilized and reported. One reason is the researchers' lack of practical knowledge on using and reporting DAGs in their studies. This article will introduce the basic concepts of DAGs to help public health researchers familiarize themselves with the various terms used in DAGs. In addition, the illustrative example provided in the article will help researchers to draw, interpret, and report DAGs in their epidemiological studies.