Decision making in laboratory test studies.

Evaluation of laboratory tests can be made at multiple levels. Evaluation at the levels of operational and decision making usefulness may be the most important but most neglected areas of analysis. Although quantitative models for such analysis are available, their application is often hindered by lack of precise probability measurements, unavailability of sensitivity and specificity data, and a poor understanding of the decision making process by physicians. A general understanding of such models and their mathematic principles have been presented with the hope that this will improve the diagnostic usefulness and application of the laboratory tests to be discussed in the remainder of this volume.