PATHOLOGY AND PROBABILITY - LIKELIHOOD RATIOS AND RECEIVER OPERATING CHARACTERISTIC CURVES IN THE INTERPRETATION OF BRONCHIAL BRUSH SPECIMENS

Citation
Ss. Raab et al., PATHOLOGY AND PROBABILITY - LIKELIHOOD RATIOS AND RECEIVER OPERATING CHARACTERISTIC CURVES IN THE INTERPRETATION OF BRONCHIAL BRUSH SPECIMENS, American journal of clinical pathology, 103(5), 1995, pp. 588-593
Citations number
26
Categorie Soggetti
Pathology
ISSN journal
00029173
Volume
103
Issue
5
Year of publication
1995
Pages
588 - 593
Database
ISI
SICI code
0002-9173(1995)103:5<588:PAP-LR>2.0.ZU;2-R
Abstract
Diagnoses in pathology often are qualitative, such as atypical or susp icious and consequently are thought to have limited clinical value. To investigate the utility of a qualitative diagnostic system, seven pat hologists retrospectively evaluated 100 bronchial brush specimens usin g the following categories: definitely benign, probably benign, possib ly malignant, probably malignant, and definitely malignant. The likeli hood ratio (LR) and receiver operating characteristic (ROC) curve, two statistical probabilistic measurements, were used to calculate diagno stic accuracy among individuals and groups. The results show: (1) the LR for individual diagnostic categories varied among observers, result ing in different clinically malignant probabilities; (2) observer expe rience did not appear to play a role in overall diagnostic accuracy, e xcept in the diagnosis of small cell carcinoma; (3) observers operate at higher levels of diagnostic accuracy with, rather than without, cli nical history. The authors conclude that qualitative diagnoses contain important information and can be interpreted effectively with LR and ROC.