| pubmed-article:6393122 | pubmed:abstractText | The promise of an easily administered and highly specific test for depression has produced a rapidly growing literature, which now contains numerous exceptions to the specificity described earlier as well as misgivings about the test's performance when endogenous depression is rare. Due to its modest sensitivity and the effect of prevalence on positive predictive value, the DST is of little use as a screening test. These limitations, however, do not affect its use as a confirmatory test if the suspicion of 'endogenous' depression is strong. According to the large majority of studies, the DST is rarely abnormal in healthy controls or in schizophrenics. Some of the exceptions are probably due to the lack of appropriate exclusion criteria, recruitment bias, nonspecific assays for cortisol, the use of overly broad definitions of schizophrenia. The possibility remains that some schizophrenics, perhaps the ones in which negative symptoms predominate, are truly nonsuppressors. This is certainly true for patients with dementia and the DST now shows little promise as a diagnostic aid when that disorder is suspected. In light of family history, follow-up and sleep studies, the occurrence of abnormal DST results among patients with such diagnoses as schizophreniform disorder and borderline personality is more likely to indicate diagnostic heterogeneity than DST nonspecificity. Nonsuppression is specific to 'endogenous' depression in the large majority of reports despite considerable differences in the 4 most commonly studied definitions. The few studies which applied 2 or more definitions to the same sample found marked specificity for one and very little for another definition, however, and did not find them to be interchangeable. Despite the use of operational criteria, occult rater variance among investigators poses a serious problem for the study of affective disorder. Studies so far suggest that the DST can figure prominently in the solution. | lld:pubmed |
