NIOSHTIC-2 Publications Search

Specificity and sensitivity: a significant difference.

Liss GM; Singal M

Postgrad Med 1983 Jun; 73(6):34

20041139

Dr. Steven R. Gambert and colleagues in the September 1982 Issue (page 1471 discussed the important topic of the interpretation of laboratory tests in the elderly. In their article, however, we noted a misuse of a statistical term which should be clarified. In their discussion of liver function tests, the authors mention that "(results) maybe normal despite the presence of significant liver pathology; therefore, more specific tests of liver function should be done" (pages 150-151). The authors referred to "specific" when "sensitive" was probably intended. Sensitivity is the ability of a test to detect those who have disease (ie. It is the proportion of those with disease who have a positive test result). The false-negative rate is the proportion of those with the disease who have a negative test result. This rate is related to sensitivity in the following way. Sensitivity = 1 - FN. Thus, to detect those individuals with pathology or disease who have normal screening tests, more sensitive, not more specific, tests are needed. Specificity is a test's ability to identify correctly those who do not have the disease (ie. it is the proportion of those without disease who have a negative test result). In those whose test results are positive. However, one would also like to use specific tests (those with few false-positive results to help in confirming the diagnosis.

Statistical-analysis; Diagnostic-tests

POMDAS

19830601

Other

1983

6

0032-5481

DSHEFS

Postgraduate Medicine

OH