Abstract

Frequently claims are made in the clinical literature that a particular test or data reduction procedure provides diagnostic information that is superior in quality to that provided by alternative tests or procedures. Often these claims are based solely on an evaluation of the tests’ ’’sensitivity”, that is, an estimate of the percentage of diseased patients diagnosed correctly by the test. Relatively few such studies have also evaluated a test’s ’’specificity” an estimate of the percentage of patients not at risk, or normals who are diagnosed correctly by the test.

© 1986 Optical Society of America