Causal Inference in Medical Decision Making

One of the most important tasks of decision analysts is to derive causal interpretations, on both the level of decision modeling and the level of statistical analyses of original data sets. Usually, an intervention, action, strategy, or risk factor profile is modeled to have a “causal effect” on one or more model parameters (e.g., probability, rate, or mean) of an outcome such as morbidity, mortality, quality of life, or any other outcome.

This entry introduces the key concepts of causal inference in medical decision making and explains the related concepts such as counterfactuals, causal graphs, and causal models and links them to well-known concepts of confounding. Finally, two examples are used to illustrate causal inference modeling for exposures and treatments.