Because of Bayes' theorem, a test for a rare disease (1-in-1000 prevalence) with a 5% false-positive rate yields a positive result that means only about a 1-in-51 (under 2%) chance of actually having the disease — not 95% — so screening for very rare conditions can be a waste even with a 'good' test.

factualpending

Speaker

Arnold Kling

Evidence Quote

“you have a 1 out of 51 chance of having the disease or less than 2% ... which is not 95”

Source

Arnold Kling on the Economics of Health Care and the Crisis of Abundance 11/5/2007— EconTalk
Created: 6/13/2026, 7:03:55 PM

Your notes

Loading notes...