Speaker
Theis Lange, Assistant Professor, Department of Biostatistics, University of Copenhagen
Abstract
An important problem within both epidemiology and many social sciences is to decompose the effect of given treatment or exposure into different causal pathways and quantify the importance of each of these pathways. Formal mediation analysis based on counterfactuals is a key tool when addressing this problem. During the last decade the theoretical framework for mediation analysis has been greatly extended to enable the use of arbitrary statistical models for outcome and mediator. However, the researcher attempting to employ these techniques in practice will often find implementation a daunting task as it tends to require special statistical programming.
In this paper we introduce a simple procedure based on marginal structural models which directly parameterize the natural direct and indirect effects of interest. It tends to produce more parsimonious results than current techniques, greatly simplifies testing for the presence of a direct or indirect effect, and has the advantage that it can be conducted in standard software. The talk will also discuss implementation examples in R and SAS.