Using SAS Macros for Multiple Mediation Analysis in R

Abstract

Mediation analysis refers to the process of making inferences on effects of third variables that intervene in the relationship between an exposure and response variable. The relationships among variables can be modelled by generalized linear models (GLM). However, GLM are not sufficient to describe relationships among variables when there are nonlinear relationships and potential interaction effects. A general mediation analysis method was developed using not only GLMs, but also multiple additive regression trees and smoothing splines by Yu and Li (2017). The method is implemented in the R package, mma. In this paper, we developed SAS macros so that functions in the mma package can be called and the mediation analysis performed in the SAS environment. Funding statement: Research reported in this publication was supported by the National Institute On Minority Health And Health Disparities of the National Institutes of Health under Award Number R15MD012387. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Keywords

• mediation analysis
• confounding analysis
• sas macros
• r package mma