This Article  • References  • Full text PDF  • Send to a friend  • Save in My Folder  • Save to citation manager  • Permissions  Citing Articles  • Contact me when this article is cited  Related Content  • Similar articles in JAMA  Social Bookmarking   What's this?

Linda M. Frazier, MD
Duke University Durham, NC

JAMA. 1993;269(3):354-355.

	Since this article does not have an abstract, we have provided the first 150 words of the full text PDF and any section headings.

To the Editor.

—In the excellent article by Deyo et al,¹ the sensitivity and specificity values provided in Tables 1 and 3 are useful. But predictive value can be even more helpful for making patient care decisions. For example, sciatica is listed by the authors as having a sensitivity (Se) of 0.95 and specificity (Sp) of 0.88 for a surgically important herniated disk. These high numbers seem to indicate that a history of sciatica is a pretty good test. But when prevalence is low, even tests with fairly high sensitivity and specificity have low predictive values.

In primary care settings, for instance, a surgically important disk is present in less than 2% of patients with acute low back pain¹ (prevalence [P] = 0.02). The positive predictive value (+PV) of sciatica as an indicator of an important disk problem can be calculated² as +PV = (Se) (P) / (Se) (P) + (1–Sp) . . . [Full Text PDF of this Article]

CiteULike    Connotea    Del.icio.us    Digg    Reddit    Technorati    Twitter     What's this?