To the Editor: The recent letter by Wagner et al. (Nov. 15 issue)* requires comment. When their data for combined ear-lobe crease and ear-canal hair is used, the sensitivity — which is defined as the percentage of afflicted people detected by a test — is 56.25 per cent (18 of 32), not the 90 per cent they cite, which is the predictive value of a positive test in their test population. Their false positive rate is 18 per cent (2 of 11 men without coronary artery disease). The value of this type of test in screening is a function also.
ASJC Scopus subject areas