Courses will be on Tuesday, April 8th at Campus Ledeganck (Karel Lodewijk Ledeganckstraat 35, 9000 Ghent).
Registration will be open from 8:30 on April 8th and again during lunch from 13:00 to 13:30.
See below for details on the course you have registered for.
Full day course
Introduction to Causal Inference, by Rhian Daniel (Cardiff University) and Erin Gabriel (University of Copenhagen)
Location: Auditorium 1, Ledeganck
Start time: 9:00 with a coffee break in the morning, lunch, and another coffee break in the afternoon.
This course (which had its first incarnation at the UK-CIM 2016) is a whistlestop tour of the concepts and methods of causal inference, aimed at an audience of newcomers to the area, but who have a working knowledge of topics such as regression models. The emphasis is on giving enough background on the basic ideas so that the Euro-CIM meeting can be enjoyed without feeling lost. The material covered might therefore change slightly once the final programme for the meeting is known, but is likely to include:
- the different languages of causality, e.g. do-notation, potential outcomes
- how these languages express causal effects
- the sorts of assumptions often relied upon to identify causal effects, and the meaning of identification
- graphical models used in causal inference, including DAGs and SWIGs
- regression models as causal models
- methods based on the propensity score
- instrumental variable methods, including Mendelian randomisation
- sustained exposures and time-varying confounders
- target trial emulation
- mediation analysis
The fact that the list above is far too long for a one-day course gives an impression of the nature of the workshop - rather than dwell on all the intricacies of the various methods and approaches, and how one might apply them in practice using computer packages, the focus will instead be on imparting the main ideas before moving swiftly on to the next topic.
Half-day courses
Both half-day courses include a coffee break. When registering for both, a lunch will also be included. There is a discount when registering for both half-day courses.
Morning – Target Trial Emulation in Practice, by Ruth Keogh (London School of Hygiene and Tropical Medicine) and Karla Diaz Ordaz (UCL)
Location: Auditorium 2, Ledeganck
Start time: 9:00 with a coffee break in the morning
This course will give an introduction to using the target trial emulation framework to address causal questions using observational data. The idea of ‘target trial emulation’ was set out and popularised by Hernan and Robins (2016) in their paper on “Using Big Data to Emulate a Target Trial When a Randomized Trial Is Not Available” (AJE), though similar approaches had been used earlier. Trial emulation is increasingly used across a range of clinical areas to answer questions that would not be feasible in a randomized trial or to enhance evidence from randomized trials.
The trial emulation approach starts with specifying a protocol for a ‘target trial’ – the trial we would like to conduct if it were feasible - and then emulating each element of the target trial using observational data. Trial emulation incorporates principles of study design combined with an analysis that accounts for the lack of randomization to treatment within the observational data. This course will discuss how we can use trial emulation to help us to: formulate sensible and precise causal questions, avoid common biases that can occur through study design and analysis steps, facilitate communication between different experts involved during a study. A range of examples will be discussed covering different types of treatment strategies, data sources, and clinical areas. The focus will not be on details of statistical analysis, but we will touch on different methods that have been used in practice, emphasising that trial emulation is not restricted to any particular statistical analysis method. We hope that at the end of the course participants will have built confidence in designing and implementing target trial emulation studies.
Afternoon – A Gentle Introduction to Targeted Learning, by Antoine Chambaz (Université Paris Cité)
Location: Auditorium 2, Ledeganck
Start time: 13:30 with a coffee break in the afternoon
Coined by van der Laan and Rubin in 2006, further developed and championed by a global community of researchers, targeted learning is a data analysis methodology that bridges causal analysis, statistics and machine learning. It emphasizes the estimation of well-defined target parameters while leveraging flexible machine learning techniques to ensure robust and interpretable results.
The methodology is detailed in two seminal books co-edited by Rose and van der Laan in 2011 and 2018. They serve as foundational texts, offerring both theoretical insights and practical guidance for implementing targeted learning in real-world settings.
In this lesson, I will present the main ideas behind targeted learning and demonstrate how to operationalize them.
