Section outline

  • Mediation and Effect Decomposition
     

    Faculty:  

    Location:      Rock Hall 102

    • Watch URL
    • Required Reading: VanderWeele File
    • Required Reading: Judd File
    • Required Reading: Nandi File
    • Required Reading: Kim and VanderWeele Applied Example File
    • Required Reading: Nguyen IORW File
    • Required Reading: Rudolph File

    • Required Reading:

      There has been a lot of work on mediation in the past 20 years, much of it published by Tyler VanderWeele.  His Annual Review paper is the key reading for this week and next.  Don't get lost in the formulas: they all come down to the same idea.  Once you understand the linear setting, you can wade through the algebra and realize the formulas for non-linear models are doing the same thing.  Focus on understanding the counterfactual contrasts that we refer to as direct or indirect effects and the assumptions you need to estimate them.

      The applied examples are just to show you conventional approaches and more modern examples (I've tried to order roughly in order of the development of the field, from the most conventional approach as with Judd to still-developing approaches such as Rudolph).  Consider these articles spread across this week and next week (May 17 and May 24).      

      1. VanderWeele TJ. Mediation Analysis: A Practitioner's Guide. Annual Review of Public Health 2016.

      Applied paper examples

      1. Judd et al., Dietary Patterns Are Associated With Incident Stroke and Contribute to Excess Risk of Stroke in Black Americans

      2. Nandi, Glymour, Kawachi, and VanderWeele. Using Marginal Structural Models to Estimate the Direct Effect of Adverse Childhood Social Conditions on Onset of Heart Disease, Diabetes, and Stroke. Epidemiology 2012. 

      3. Kim ES, VanderWeele TJ. Mediators of the association between religious service attendance and mortality. American journal of epidemiology. 2018 Sep 27;188(1):96-101.
      4. Nguyen TT, Tchetgen EJ, Kawachi I, Gilman SE, Walter S, Glymour MM. The role of literacy in the association between educational attainment and depressive symptoms. SSM-population health. 2017 Dec 1;3:586-93.

      5. Rudolph KE, Sofrygin O, Schmidt NM, Crowder R, Glymour MM, Ahern J, Osypuk TL. Mediation of neighborhood effects on adolescent substance use by the school and peer environments. Epidemiology. 2018 Jul 1;29(4):590-8.

    • Optional Reading:

    • Assignment:

      Please identify a quantitative research article evaluating mediation in your field and provide the citation.

      What is the primary discipline of the authors?

      Draw a DAG representing the implicit or explicit causal model explored in this paper (you do not need to post your DAG, but we will try to discuss in class).

      What is the exposure of interest?

      What is the outcome of interest?

      What is the hypothesized mediator of interest and how is it measured?

      Describe the modeling approach and briefly report the estimated total, direct, and indirect effects (if these are reported). 

      If the direct effect is reported, would you describe this as a natural direct effect, a controlled direct effect, or something else?

      Do you think there is potential measurement error in the mediator and how would that affect the results?

      Do you think there are unmeasured confounders of the mediator-outcome association and how would that affect the results of the mediation analysis?

      Do you have any critiques of the paper? 

    • Mediation slides File