Current course offerings

Please Note: Course profiles marked as not available may still be in development.

Course description

This unit covers the core concepts and methods to enable the assessment of causation from randomised and observational studies. The foundational concepts and notation of counterfactual (or potential) outcomes are first presented to define precisely the meaning of the causal effect of a treatment or exposure, and the concept of the target trial is discussed. Causal diagrams, in the form of directed acyclic graphs (DAGs), are presented as a visual representation of a postulated underlying causal structure, together with the key properties and rules used to determine the likely existence of confounding, selection and other biases that prevent unbiased estimation of causal effects. The unit then proceeds to methods of estimating average causal effects. Propensity score methods for assessment of single time point treatments/exposures are studied in detail, followed by longitudinal designs, in which the treatments/exposure and their effects can vary over time. The study of mediation effects, which are explanatory methods for determination of mechanisms and pathways by which an exposure affects and outcome, are covered using language and methods of the potential outcomes framework. The unit concludes with a discussion of what to do when participants do not fully adhere to treatments, with an introduction to principal stratification and instrumental variables.

Archived offerings