The Use of the Directed Acyclic Graph to Disentangle Confounding Variables from Causal Factors in Observational Studies.

The principal objectives of biomedical research are to identify causal relationships, to understand mechanisms that underlie normal functions and the pathophysiological mechanisms involved in diseases, in order to preserve health. Randomized controlled trials (RCTs) are the gold standard to demonstrate causality due to their design. However, they are not always feasible or are unethical in some clinical circumstances due to their nature or involved economic costs. Conventional observational studies are an alternative in many clinical and epidemiological investigations; however, they may imply bias risks due to the difficulty of separating confounding and causal factors and adjusting for baseline covariates.