23 May 2012 10:00am

Associating with DAGs can be beneficial: A tour through counterfactuals, causal graphs, challenges and opportunities


In the past 15 years there has been an explosion of analytical methodological development based on counterfactuals (potential outcomes) and utilisation of causal diagrams to understand and inform non-randomised comparisons in epidemiology and related areas.

These largely stem from the seminal works of Rubin (1974) regarding potential outcomes, of Pearl (2000) for causal diagrams, and Robins (1986) for repeated exposures over time. In this talk I will provide an outline of the key aspects of these approaches and the correspondence between them, and will illustrate select features using applications from personal experience and the literature.

Challenges associated with implementation will be discussed together with opportunities for further work in this rapidly expanding area.