Abstract
Assessment of diagnostic tests in veterinary clinical pathology usually includes analytical performance characteristics and values of sensitivity and specificity estimated when comparing test results from affected and non-affected animals to a cutoff value corresponding to the upper/lower limit of the reference range. This cutoff value is but one of many to which the test results can be compared and because sensitivity and specificity varies as the cutoff value is changed, a complete picture of the test's clinical usefulness is not presented by only one pair of sensitivity and specificity results. Instead, a complete picture of the test's clinical usefulness can be appreciated by the entire spectrum of sensitivity and specificity pairs resulting from continuously varying the cutoff value over the entire range of test re suits. The receiver-operating-characteristic (ROC) curve provides a view of this whole spectrum of sensitivity/specificity pairs and it allows the investigator to examine the test's discriminative power over all possible cutoff values. This paper presents methods for preparing and analysing ROC curves and illustrates the principles presented by applying ROC-curve analysis to a set of data consisting of protein and L-lactate concentrations in peritoneal fluid from horses suffering from either surgical or medical acute abdomen.
Originalsprog | Engelsk |
---|---|
Tidsskrift | Comparative Haematology International |
Vol/bind | 6 |
Udgave nummer | 3 |
Sider (fra-til) | 176-181 |
Antal sider | 6 |
ISSN | 0938-7714 |
DOI | |
Status | Udgivet - 1996 |