Summary
The episode works through five questions that students commonly find difficult in the introduction to observational studies. The hosts use the reanalysis of hormone therapy and coronary heart disease in the Nurses' Health Study to show why prevalent users mislead and how target trial emulation aligns time zero, and they discuss what the hierarchy of evidence can and cannot establish. Worked examples compare the prevalence ratio with the prevalence odds ratio, classify exposures by their vulnerability to reverse causation and recall bias, and show how a time-invariant factor that only lengthens illness can appear to be a cause. The episode closes by estimating yearly incidence from two prevalence figures and by examining the assumptions behind that estimate.