Planning your Target Trial and navigating real-world metadata: foundations for causal inference

Off

If you would like to register your interest, please complete this form and we will be in touch if new dates are released. 


ºù«Ӱҵ the course

This course provides an introduction to a key structural foundation for causal inference: the Target Trial framework. The course will focus on applications of the Target Trial framework to real-world data sources, including electronic health records and registries. The course covers how to:

  • Understand the basic principles and purpose of causal inference
  • Design the ‘ideal’ trial based on seven key components: eligibility criteria, treatment strategies, assignment procedure, follow-up period, outcomes, estimands, and analysis plan
  • Develop an appropriate estimand statement (i.e. what is intended to be estimated)
  • Identify and critique single-source or linked real-world data sources
  • Navigate metadata to understand the potential emulated observational study
  • Identify potential bias in the emulated study such as selection bias
  • Navigate core information governance processes including data sharing agreement

Learning outcomes

After this course, you should be able to:

  • Distinguish causal inference from prediction and descriptive studies, and their relative uses
  • Understand interpretation issues caused by bias such as confounding and selection bias
  • Understand how to develop your own target trial and engage with relevant stakeholders
  • Write a clear estimand statement
  • Understand strengths and limitations of real-world data including navigating metadata
  • Have an awareness of the range and variety of national and localised real-world data sources
  • Understand core principles associated with information governance when requesting and storing real-world data

Who will benefit from the course?

This course is not an analytical causal inference course, such that no statistical software (e.g. Stata or R) will be used. The Target Trial framework is a structural methods for causal inference, whereby the study design itself aids avoid bias in any intended causal estimates.

As such, the course is aimed at individuals who wish to inform the design of real-world causal inference studies based on the Target Trial framework, but who are not necessarily those who will be conducting the causal inference analyses. As developing an appropriate target trial design requires input from a variety of stakeholders, this course is aimed at any individual interested in the use and design of target trials, real-world evidence, and causal inference. No previous knowledge of trials, causal inference or statistics is required for the individual to benefit from this course.

Course faculty

Dr Matt Franklin, Senior Research Fellow, School of Medicine and Population Health

A global reputation

ºù«Ӱҵ is a world top-100 research university with a global reputation for excellence. We're a member of the Russell Group: one of the 24 leading UK universities for research and teaching.