Introduction to Directed Acyclic Graphs (DAGs) in epidemiological research

When undertaking observational studies, we need to carefully consider potential confounding. Causal diagrams, such as directed acyclic graphs (DAGs), are powerful tools to do so in a transparent and logical way. DAGs can visually represent multiple variables and our assumptions about the relationships between them in a given setting.

By clarifying these assumptions, we can use rule-based tools to inform modelling decisions to decide which variables should be controlled for and which variables should not be controlled for.

In this workshop, we will:

· Introduce causal diagrams in general, and DAGs in particular

· Practice drawing DAGs and using them to help inform our modelling choices

· Learn how to use DAGitty, a common software for constructing and using DAGs

· Invite participants to bring research questions and discuss how they can be represented in a DAG

Know what’s Happening Next — before everyone else does.

Get App

Host or Publisher UCL Centre for Electronic Health Records research And Methodology (CEHRAM)