Tutorial: a nontechnical explanation of the counterfactual definition of effect modification and interaction

Causal questions addressed by (clinical) epidemiological studies usually focus on the causal effect of a single exposure on a certain outcome. For instance, whether and how exposure to a specific risk factor or medical treatment causes the occurrence or cure of some disease. It is well-recognized, however, that most outcomes are not caused by single exposures in isolation but by multiple exposures in combination. That is why experimental and observational studies frequently report results of subgroup or stratified analyses [4-12], to explore if the effect of the primary exposure under study depends on the level of another factor or on the effect of a secondary exposure.

Source: Journal of Clinical Epidemiology - February 3, 2021 Category: Epidemiology Authors: Martijn J.L. Bours Source Type: research