Challenges of Causal Inference in Health Studies

Remember when they were coming to take your gas stove away? Every so often a study about the effects of air pollution on health goes viral, and we’re reminded again that seemingly innocuous objects—like your kitchen cooker—could be bad for us in unexpected ways. How bad is air pollution? And is it getting any better?

In this episode of The Studies Show, Tom and Stuart look into the science of air pollution, trying to separate correlation from causality, and working out what scientists mean when they say that deaths are “attributable” to something (it’s more complicated than you think!).

The Studies Show is brought to you by Works in Progress magazine. We usually mention their long-form pieces at worksinprogress.co, but they also have a Substack newsletter at worksinprogress.news with shorter articles on the same topics. We commend it to you, and thank Works in Progress for sponsoring the podcast.

Show notes

* Recent news about “Ella’s Law” in the UK

* Tom’s 2019 Unherd article on air pollution

* “Death risk from London's toxic air sees ‘utterly horrifying’ rise for second year running”

* The Our World In Data “Deaths by Risk Factor” graph

* 2024 BMJ Open article about the health risks of coal power stations

* Dynomight’s long article on air quality

* The 1952 “Great Smog of London”

* More useful Our World In Data articles:

* An explainer on “attributable fractions” and summing up multiple risk factors

* On indoor air pollution

* Deaths from outdoor pollution

* Death rate from outdoor pollution

* Deaths from outdoor pollution vs. GDP per capita

* The WHO calls indoor air pollution “the world’s single largest environmental health risk”

* More on attributable fractions, with some examples

* Example of an experimental study on the effects of air pollution

* The article that sparked the Great Cooker Controversy of 2023

* Example of the media coverage at the time

* Biden forced to rule out a ban on gas cookers

* Recent story on how there’s “no safe level” of PM2.5

* Based on this 2024 paper in the BMJ

* How policy interventions can reduce (and have reduced) air pollution

* London report on the effect of ULEZ

Credits

The Studies Show is produced by Julian Mayers at Yada Yada Productions.

This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit sciencefictionspod.substack.com/subscribe