Statistical Methods for Mediation Analysis

Coronavirus (COVID-19) information

The Short Course Programme in Population Health Sciences has been temporarily suspended.

Booking

Bookings for 2020-21 courses will open later in the autumn.

Information on this page relates to the last run of the course and is for reference only. 

Mailing List

Sign up to our mailing list to be notified when bookings reopen. 

We may need to make responsive changes to our future programme to follow the latest Public Health, Government and University guidance on coronavirus (COVID-19).

Please be aware that all information about short courses planned for 2021 is provisional and subject to change.

Course dates

2 - 3 July 2020

Course duration

2 days (approximately 14.5 hours of teaching).

Registration will start at 9am on the first day, the course will finish by 4pm on the final day.

Course tutors

Dr Gemma HammertonDr Laura Howe, Dr Jon Heron (course organisers) and others.

Course aims

To provide an understanding of the statistical principles behind, and the practical application of, mediation analyses in epidemiology.

Course objectives

  1. To be aware of traditional and counterfactual approaches to mediation analysis and the assumptions underlying these methods;
  2. To have a working knowledge of the Stata commands to run models for traditional and counterfactual approaches to mediation analysis;
  3. To understand mediation analysis in the presence of exposure-mediator interaction;
  4. To be aware of methods to incorporate multiple mediators and intermediate confounders; and
  5. To have a working knowledge of the Stata commands to run models with multiple mediators or intermediate confounders.

Who the course is intended for

Medical statisticians, and epidemiologists with a quantitative background and knowledge of linear and logistic regression. 

Prerequisites: Participants should have knowledge of regression analyses and their implementation in Stata of at least the level achieved in the 'Introduction to Linear and Logistic Regression Models' short course. 

Course outline

  1. Review and critique traditional methods for mediation analyses;
  2. Describe counterfactual-based approaches to mediation analysis;
  3. Describe mediation analysis in the presence of exposure-mediator interactions;
  4. Review traditional and counterfactual methods to incorporate multiple mediators; and
  5. Introduce counterfactual methods to incorporate intermediate confounders.

Please note: Practical sessions of this course will be held in a computer lab, so you will not need to bring a laptop. Stata version 15 will be used during the course.

Recommended reading

Aitken, Z., Simpson, J.A., Gurrin, L., Bentley, R. & Kavanagh, A.M., 2018. Do material, psychosocial and behavioural factors mediate the relationship between disability acquisition and mental health? A sequential causal mediation analysis. International journal of epidemiology, 47(3), pp.829-840.

It was an excellent and interesting course, and the teaching style was great.

Course feedback, July 2019

Course fee

£440

More information on course fees, fee waivers and reduced prices.

Course venue

Bristol Medical School
Canynge Hall
39 Whatley Road
Bristol
BS8 2PS
United Kingdom

Map and directions

Course refreshments

We provide morning and afternoon refreshment breaks, including tea and coffee, biscuits and fresh fruit.

If you have specific dietary needs we ask that you let us know in advance.

Lunch is not included. There are a range of local cafes and supermarkets nearby for students to purchase lunch. 

Accommodation

Information about accommodation in the area.

Contacts

For further information please email short-course@bristol.ac.uk.

Edit this page