[Skip to Content]
[Skip to Content Landing]
Letters
July 21, 1989

Sensitivity, Specificity, and Predictive Values in the 'Sensitivity and Specificity of Clinical Diagnostics'-Reply

Author Affiliations

University of New Mexico School of Medicine Albuquerque

University of New Mexico School of Medicine Albuquerque

JAMA. 1989;262(3):351. doi:10.1001/jama.1989.03430030038022

This article is only available in the PDF format. Download the PDF to view the article, as well as its associated figures and tables.

Abstract

In Reply.—  Several individuals have called our attention to an unfortunate error in typing that led to publication of an incorrect Table 1. The correct version, as they have all pointed out, is as shown (Table).Drs Woodruff and Baron make an important point in noting that predictive values (which we termed clinical accuracy in an attempt to make it more understandable to those not expert in decision theory) are sensitive to changes in prevalence. As a matter of fact, predictive values were added to the manuscript in response to the suggestion of one of the reviewers, who remarked that their use would be acceptable, so long as they were not going to be directly applied in the screening of other populations. In some of the diseases we studied, there was a marked change in prevalence (as in the case of tuberculosis) that accounted for much of the change in

×