S5E4: Mind the Modifier: When Causal Effects Refuse to Be Average

S5E4: Mind the Modifier: When Causal Effects Refuse to Be Average

From SERious EPI by Sue Bevan - Society for Epidemiologic Research

April 15, 2026 · 56 min

About this episode

The episode discusses effect modification in causal inference, emphasizing the variability of treatment effects across different population characteristics.

Hailey and Matt are joined by guest co-host Dr. Mabel Carabali to discuss Chapter 4 (Effect Modification) from Causal Inference: What If . We start off our discussion about heterogeneity of treatment effects, emphasizing that there is often no single causal effect but effects that vary across groups depending on population characteristics. Mabel helps to explain effect (measure) modification as variation in the exposure outcome effect across levels of a third variable. She also explains the concept of qualitative effect modification. We talk about how these concepts connect to transportability and generalizability. The end of the episode focuses on effect modification on the additive vs. multiplicative scales, continuing our (neverending?) debate about when and why we should care about effect modification on the relative scale versus the absolute scale.

People in this episode

Hosts: Hailey, Matt

Guest: Dr. Mabel Carabali

Topics covered

  • effect modification
  • heterogeneity of treatment effects
  • causal inference
  • transportability
  • generalizability
  • additive vs multiplicative scales

Keywords

  • causal effects
  • effect modification
  • treatment effects
  • population characteristics
  • qualitative effect modification
  • transportability
  • generalizability

Mentioned in this episode

Books & works: Causal Inference: What If

More episodes of SERious EPI

Explore listener stats, chart rankings, contacts and more on the SERious EPI podcast page.