Effect modification is commonly assessed using stratified analysis or interaction terms in regression models. In stratified analysis, the data is divided into strata based on the levels of the potential effect modifier, and the association between the exposure and the outcome is examined within each stratum. If the associations differ significantly across strata, effect modification is likely present. Interaction terms in regression models can also be used to formally test for effect modification by including terms that represent the product of the exposure and the effect modifier.
