14 Feb 2022 01:30pm to 15 Feb 2022 06:15pm

Causal Mediation Methods (Summer School 2022 days 5 & 6)

Event Location
Prof Stijn Vansteelandt
Ghent University, Belgium
Margarita Moreno Betancur
Margarita is co-lead of the Clinical Epidemiology and Biostatistics Unit (CEBU) at the MCRI and the University of Melbourne. Since completing her PhD in Biostatistics at Université Paris-Sud in 2014...
Prof John Carlin
John holds appointments with the Murdoch Children’s Research Institute and The University of Melbourne. Since completing a PhD in Statistics at Harvard University he has been engaged as a collaborator...
Marnie is a biostatistician at the Murdoch Children's Research Institute. She completed her PhD at The University of Melbourne in 2020, investigating the application of multilevel regression and poststratification for...
Ghazaleh Dashti
Ghazaleh completed her PhD in Epidemiology at the University of Melbourne in 2020. Her current research focuses on methods for handling missing data in the context of causal inference methods...

Many epidemiological questions concern the pathways that are presumed to mediate a relationship between a cause and its effect. Very often, the translational intent of such research questions is to inform potential intervention targets. However, until recently causal mediation analysis methods did not define mediation effects as effects of real-world interventions, and the assumptions underlying various methods were either too stringent or not assessable in practice, particularly in the context of multiple mediators. These issues have resulted in diverse views regarding the practical value of mediation analysis and related methods.

This course begins by providing an overview of the conceptual issues surrounding mediation analysis and related methods. We then present a recent approach that conceptualises mediation effects by mapping to a “target trial” evaluating interventions on one or several mediators. We describe how to define and emulate a target trial for mediation analysis and introduce an extended g-computat­ion approach for estimating the resulting “interventional” mediation effects.

Lectures and tutorials will help ground understanding of the methods, whilst a hands-on computer practical (in R and Stata) will cover their practical implementation. All lectures and tutorials include illustrations from real-world observational epidemiological studies. Electronic copies of presentation materials will be made available online.

Prerequisites:  It is strongly recommended that participants have previously taken either the ViCBiostat causal inference workshop (delivered in 2020 or in this same Summer School) or the course “Observational studies: Modern concepts & analytic methods” delivered by the Clinical Epidemiology and Biostatistics Unit (CEBU) at the Melbourne Children’s / MCRI. To do the computer practical, students must also have a sound working familiarity with Stata or R and have the corresponding software installed on their computer or laptop.


Please note day 5 (14 February) runs from 1:30pm to 5pm and day 6 (15 February) runs from 1:30-6:15pm.

Further detail is attached in the timetable.


Registration closes 5pm Friday 11th February

Register here