Summary
The episode works through four questions that students commonly find difficult in the study of screening and diagnostic tests. It uses published accuracy figures for the PHQ-9 depression questionnaire to show why the positive predictive value of a sensitive test can fall below forty percent, and it explains the weighting of errors that is built into choosing a cutoff by the Youden index. It draws on a Cochrane review of rapid antigen tests for COVID-19 and a study in the Canadian Medical Association Journal to show how sensitivity and specificity vary with the patients studied and how verification bias can inflate sensitivity. It then converts a pre-test probability to post-test probabilities with likelihood ratios for a rapid strep test, and it extends the series testing formula to two tests that share their false positives, showing how far the positive predictive value falls when the conditional independence assumption fails. The episode closes with lead-time bias, length bias and overdiagnosis, illustrated by thyroid cancer screening in South Korea, and with a discussion of routine depression screening in which the hosts agree on a shared rule for judging screening programs.