Last update:


Arch Hellen Med, 21(2), March-April 2004, 179-194


ROC curve in diagnostic research

Laboratory of Clinical Epidemiology, School of Nursing, University of Athens, Athens, Greece

The clinical usefulness of a diagnostic test (laboratory test or clinical finding) is determined by the discriminating power of the test, meaning the ability to correctly classify patients with or without the disease. The most commonly used measures of the diagnostic value of a test are the true positive rate (sensitivity) and the true negative rate (specificity). These rates are the operating characteristics of the diagnostic test. The relationship between the true positive rate and the false positive rate (one minus the true negative rate), resulting from continuously varying the cut-off point over the entire range of results observed, is demonstrated graphically by a curve called Receiver Operating Characteristic (ROC) curve. The selection of the cut-off point (particular value of the measured variable above which the test result is considered positive and under which it is considered negative) of a diagnostic test is of great importance as it determines the discriminating power of the test. The ROC curve provides means for (a) visual (shape and position of the curve) and quantitative (area under the ROC curve) assessment of the overall performance of a diagnostic test independently of the ultimately chosen cut-off point, (b) comparison of the overall performance of two or more diagnostic tests, and (c) selection of the optimal cut-off point of the test depending on the diagnostic value of the test, the costs and benefits of the test results and the disease prevalence.

Key words: Cut-off point, Operating characteristics, ROC curve.

© Archives of Hellenic Medicine