To the Editor.—
The recent report by Calin et al (237:2613, 1977) suggests that a medical history is a sensitive and specific screening test for ankylosing spondylitis. However, the authors confuse the concept of specificity (the proportion of patients without ankylosing spondylitis who have a negative test) with the concept of predictive value (the proportion of patients with a positive test who prove to have ankylosing spondylitis). Specificity is not dependent on prevalence since it is determined by patients free of the disease in question, whereas predictive value depends strongly on prevalence and may be calculated from the specificity and sensitivity of the test and the prevalence of the disease.1The Table compares the sensitivity, specificity, and predictive values of a positive HLA-B27, four or five affirmative answers, or five affirmative answers to the five most discriminating questions, using the author's values for prevalence, sensitivity, and specificity. Note that
Crawford G. Ankylosing Spondylitis. JAMA. 1977;238(20):2142. doi:10.1001/jama.1977.03280210034006
Customize your JAMA Network experience by selecting one or more topics from the list below.
Create a personal account or sign in to: