Advanced Topics in Causal Research (online)

This course offers an introduction to causal epidemiological research from a modern perspective. We begin by covering a framework and language that enables us to make a large variety of causal research goals explicit. These include goals relating to mediation, effect modification, interaction, and the effects of time-fixed or time-varying treatments.

With the foundations in place, we turn to the key obstacles on the path to reaching such research goals. Emphasis is placed on confounding and selection bias. Other sources of bias, notably missing data and measurement error are not addressed in detail. You are also introduced to directed acyclic graphs as tools to identify structural sources of bias and guide analysis.

Having described key obstacles, we address ways (other than classical multivariable regression) to overcome them, incl. propensity score analysis, inverse probability weighting and instrumental variable analysis.

At the end of the course, you should be able to:

• be able to approach causal inference from a counterfactual outcomes perspective.
• be able to distinguish between conditional and marginal causal effects and understand the concept of collapsibility.
• be able to represent assumptions in directed acyclic graphs (DAGs) and be able to read off the implications of these assumptions.
• be able to recognize key obstacles to causal inference, including time-fixed and time-varying confounding and selection bias.
• understand more advanced approaches to account for confounding, including propensity score methods and instrumental variable methods, and be able to apply these.
• understand the concepts of effect modification and interaction and be able to compute and interpret different measures of effect modification and interaction.
• understand principles and pitfalls of mediation analysis.

Our courses are aimed at clinical researchers, nurses, general practitioners, and other health professionals who want to improve their skills in epidemiology, statistics and (clinical) research.

3 weeks, 14 hours per week (online)

Even though you can manage your own time our courses are not intended as individual education. We offer personalized online learning with lots of interaction with peer students, the E-moderator and lecturers. Flexibility from students, a positive attitude towards teachers and peers and the willingness to learn together and help each other is invaluable to our courses. To experience maximum interaction, we advise you to log on several times per week.

Note that the starting dates of courses, interim deadlines, and dates of exams are fixed, but you can choose when and where you want to watch web lectures and work on assignments. The e-moderator of the course will inform you about the beginning of the course and about deadlines during the course.

The average required study workload for the courses of MSc Epidemiology Postgraduate Online is 14 hours per week. You will need this time to study, to keep up with the assignments and course material.