Causal Inference in Epidemiology: Recent Methodological Developments
Course
In London
Description
-
Type
Course
-
Level
Intermediate
-
Location
London
-
Duration
Flexible
-
Start date
Different dates available
Causal inference is a central aim of many empirical investigations, and arguably most studies in the fields of medicine, epidemiology and public health. However, traditionally, the role of statistics is often relegated to quantifying the extent to which chance could explain the results, whilst concerns over systematic biases due to the non-ideal nature of the data are relegated to their qualitative discussion. The field known as causal inference has changed this state of affairs, setting causal questions within a coherent framework which facilitates explicit statement of all the assumptions underlying a given analysis, in many settings developing novel, flexible analysis methods, and allowing extensive exploration of potential biases.
This course will discuss the current state of the art with respect to these issues, while retaining a practical focus. The potential outcomes framework, causal diagrams, standardization, propensity scores, inverse probability weighting, instrumental variables, marginal structural models, causal mediation analysis and examples of sensitivity analysis will be discussed. Participants will acquire awareness of the common threads across these new methods and competence in applying them in simple settings.
Facilities
Location
Start date
Start date
About this course
Participants will be expected to be numerate epidemiologists, or applied statisticians with an interest in epidemiology and clinical trials. An MSc in Epidemiology or Medical Statistics, or previous attendance to the Advanced Course in Epidemiological Analysis, would be an advantage.
There will be no formal assessment. Participants will receive a Certificate of Attendance.
Reviews
Subjects
- Mediation
- Language
- Causal language
- Estimands
- Diagrams
- Unmeasured
- Confounders
- Regression
- Propensity
- Instrumental
Course programme
The topics covered will be:
- Causal language, estimands, and diagrams
- Methods to deal with the bias introduced by measured and unmeasured confounders. These will include: standard regression methods, propensity score-based and instrumental variable methods
- Marginal structural models for dealing with time-dependent confounding in longitudinal studies
- Causal mediation analysis
- Sensitivity analysis
- Practical experience of the above methods in Stata
Causal Inference in Epidemiology: Recent Methodological Developments