The dexamethasone suppression test (DST) is the most extensively studied biological test in psychiatry. Despite this, its role in the diagnostic assessment of psychiatric patients remains controversial. Shortcomings of the test include limited sensitivity (45%) and limited specificity (75-80%) (1). The DST has many proposed uses, including the differentiation of endogenous from non-endogenous depressions, helping to decide when maintenance medications may be withdrawn, and as a diagnostic test for major depression. This paper is concerned with the latter use only, that is, the ability of the DST to function as a useful diagnostic test for major depression.
The significance of the low sensitivity and specificity of the DST have been discussed extensively in the psychiatric lite rature (2,3). When the DST is used as a diagnostic test for major depression its sensitivity represents the likelihood of a positive test given that the tested patient has a major depression. Specificity represents the likelihood of a negative test given that the tested patient does not have a major depression. Notably, both sensitivity and specificity represent probabilities conditional on the presence or absence of disease. However, in the usual clinical situation, the diagnosis is unknown at the time the diagnostic test is ordered. Indeed, if the diagnosis is known then the test should not be performed. Thus neither sensitivity nor specificity is directly relevant to the interpretation of the results of diagnostic tests in a clinical setting. Another parameter, the positive predictive value (+ PV), is of more clinical relevance.
Patten, M.D., Scott B.
"Enhancing the Utility of the Dexamethasone Suppression Test: A Chart Review and Application of Bayes' Theorem,"
Jefferson Journal of Psychiatry: Vol. 8
, Article 9.
Available at: http://jdc.jefferson.edu/jeffjpsychiatry/vol8/iss1/9