You are here
Multiple questions for multiple mediators
Investigating the mechanisms that may explain the causal links between an exposure and a temporally distal outcome often involves multiple interdependent mediators. Until recently, dealing with multiple mediators was restricted to settings where mediators relate to exposure and outcome only linearly. Extensions proposed in the causal inference literature to allow for interactions and non-linearities in the presence of multiple mediators initially focussed on natural direct and indirect effects. These however are not all identifiable, with the rest requiring stringent, and often unrealistic, assumptions. More recent developments have focussed interventional (or randomised) direct and indirect effects to deal with these issues (Vansteelandt & Daniel, 2017). They can be identified under less restrictive assumptions, with generalizations dealing with time-varying exposures, mediators and confounders also possible (VanderWeele & Tchetgen Tchetgen, 2017).
The mediation questions that can be addressed when estimating interventional effects differ from those asked by natural effects in subtle ways. In this talk we will review them, discuss their differences in emphasis, assumptions, and interpretation, and propose ways of exploiting these differences to assess the robustness of conclusions. We will use an epidemiological investigation of the mechanisms linking maternal pre-pregnancy weight status and offspring eating disorders behaviour to illustrate these points.
Bianca De Stavola is Professor of Medical Statistics at UCL Great Ormond Street Institute of Child Health, London, UK. She recently joined UCL after 23 years at the London School of Hygiene and Tropical Medicine where she was co-Director of the Centre for Statistical Methodology. Bianca received her PhD from Imperial College London and MSc from the London School of Economics and Political Sciences, after graduating in Statistical and Economic Sciences at Padua University. Her main research activities involve the understanding, development and implementation of statistical methods for long-term longitudinal studies, with specific applications to life-course epidemiology. As these often involve causal enquiries, in particular related to understanding pathways towards disease development, mediation analysis is a main focus of interest of collaborations.
FBK-IRVAPP Vicolo Dallapiccola 12, 38122 Trento