Inching toward Bethlehem: Mapping melancholia

Background: As melancholia has resisted symptom-based definition, this report considers possible explanations and options for moving forward. Clinician-assigned melancholic and non-melancholic groups were initially compared to refine a candidate set of differentiating symptoms alone for examination against a set of non-clinical validators. Analyses then examined the capacity of both the refined symptom and validator sets to discriminate the assigned melancholic and non-melancholic subjects. Methods: Subjects completed measures assessing symptoms and correlates (putative validators) of diagnostic sub-type, and were assessed independently by two psychiatrists. Results: Analyses identified 14 severity-based symptoms as discriminating clinically-diagnosed groups with melancholic subjects differing significantly from non-melancholic subjects across a number of validators. Such symptom-based discrimination was superior to DSM-IV and Newcastle Index assignment in a study sub-set. While the refined symptom set had an overall accurate classificatory rate of 68%, use of the combined sets of refined symptoms and validators improved classification to 80%. Conclusions: Melancholia definition is improved by the use of correlates in addition to depressive symptoms, suggesting that melancholia may be mapped more precisely by use of multiple coordinates or data sources. (C) 2009 Elsevier B.V. All rights reserved.