The methodological and statistical evaluation of medical tests: the dexamethasone suppression test in psychiatry.

Presented herein is a step-by-step discussion of the statistical evaluation of a medical test, using the literature concerning the DST to provide good and bad examples of its implementations. Specific topics of concern include: (1) conceptualization issues such as the definition of a test and its purpose; (2) implementation issues such as sampling and blindness; (3) analytic issues such as choice of statistical measures of quality; (4) issues of correct and realistic interpretation of results; and, finally (5) cost-benefit issues.