[Skip to Navigation]
April 1997

The Diagnostic Validity of Melancholic Major Depression in a Population-Based Sample of Female Twins

Author Affiliations

From the Departments of Psychiatry and Human Genetics, Medical College of Virginia, Virginia Commonwealth University, Richmond.

Arch Gen Psychiatry. 1997;54(4):299-304. doi:10.1001/archpsyc.1997.01830160013002

Background:  Although the diagnosis of melancholia is among the oldest in psychiatry, the validity of the melancholic subtype of major depression (MD) is still debated. If melancholia is a valid subtype of depression, is it quantitatively more severe than or qualitatively distinct from nonmelancholic depression?

Methods:  The lifetime history of MD and melancholia, defined by DSM-IV criteria, was assessed at interview in 1902 female twins selected from a population-based register. Patterns of comorbidity and the relationship between melancholia and risk for MD in the co-twin were assessed by logistic regression and Cox proportional hazards models, respectively.

Results:  In those with a lifetime history of MD, melancholia was associated with the following: (1) increased comorbidity with anxiety disorders and nicotine dependence but not alcohol dependence or bulimia; (2) greater number of episodes, more impairment, and help seeking; (3) lower levels of neuroticism; and (4) an increased risk of MD in cotwins—greater in monozygotic than in dizygotic pairs. Within twin pairs concordant for MD, no resemblance was found for melancholia. A multiple threshold model in which melancholic MD represented a quantitatively more severe form of depressive illness fitted the data well.

Conclusions:  Melancholia, defined by DSM-IV criteria, is a valid subtype of MD and identifies a subset of affected individuals with distinct clinical features and a particularly high familial liability to depressive illness. However, from a familial perspective, the differences between melancholic and nonmelancholic MD are quantitative, not qualitative. ie, melancholic MD is more severe than, but is not etiologically distinct from, nonmelancholic MD.

Add or change institution