Reporting of interaction.

摘要

Interaction is the situation whereby the association of one risk factor with a certain outcome variable differs across strata of another risk factor. From a public health perspective, the assessment of interaction on an additive scale may be most relevant. Although additive models exist, logistic and Cox regression models are the most commonly used models in epidemiology. The resulting relative risks can be translated to an additive scale. The present paper presents surrogate measures to evaluate the presence of additive interaction when dealing with data on a multiplicative scale (relative risks). For a transparent presentation of interaction effects it is recommended to report the separate effect of each exposure as well as their joint effect compared to the unexposed group as joint reference category to permit evaluation of interaction on both an additive and multiplicative scale.