Diagnosis of endogenous depression: Comparison of clinical, research and neuroendocrine criteria

Abstract

Eighty-nine depressed outpatients were studied by clinical criteria, Research Diagnostic Criteria (RDC), and the dexamethasone suppression test (DST) of neuroendocrine regulation. A simple outpatient version of the DST, requiring only one blood sample, correctly identified 40% of patients diagnosed clinically as endogenous depression (ED), with a specificity of 98% and a diagnostic confidence of 95%. Differences in age, sex, or severity of symptoms between endogenous and non-endogenous depressives did not account for these results. By comparison, the diagnostic performance of the DST was weaker for the RDC categories Major Depressive Disorder (MDD) and primary MDD. These were less selective and more heterogenous than the clinical category ED. The clinical diagnoses of ED were supported in 98% of cases by the RDC, but 22% of RDC endogenous MDD diagnoses were not supported by the clinical diagnoses. Abnormal DST results were found only in patients with both the clinical diagnosis of ED and the RDC diagnosis of endogenous MDD. Patients with definite endogenous MDD had a significantly higher frequency of abnormal DST results (42%) than those with probable endogenous MDD (14%), or those with other RDC diagnoses (3%). A significant association was found between positive DST results and a positive family history of depression. These results support other evidence for use of a positive DST result as an external validating criterion for ED. The category MDD contained all cases diagnosed clinically as ED, but was diluted by cases diagnosed clinically as non-endogenous depression who had no neuroendocrine disturbance. The results also confirmed that the endogenous/nonendogenous and primary/secondary classifications of depression are not identical. We conclude: (1) that the DST can be used in the differential diagnosis of depressed outpatients as well as inpatients; (2) that the RDC category primary MDD and the Washington University category primary depression are more heterogenous and probably less valid than the clinical category ED; (3) that the RDC for endogenous MDD have only moderate validity; (4) that RDC diagnoses cannot substitute for careful clinical diagnoses in research studies, (5) that the best use of the RDC is to support clinical diagnoses, but not to generate diagnoses independently as a free-standing system; (6) that the concept of endogenous or endogenomorphic depression has validity and should be retained in research studies of depression.