Melancholia disappeared as a distinct diagnosis since 1980’s and is now only a specifier for major depressive disorder.Many still argue that given the biological abnormalities and response to tricyclics and ECT, it should be reinstated as a specific diagnosis. Endogenous depression/ psychotic depression/ vital depression are the different names/ shades of melancholia. A North Wales study that examined melancholia between 1875 and 1924 reported that psychotic symptoms were present in 70% and that these episodes had a sudden onset and were not perceived as being precipitated by external events, and in average lasted for 6 months (Harris et al., 2011). This group also reported that incidence of psychotic depression is declining.
Margaret Harris,Fiona Farquhar, David Healy, Joanna C Le Noury, Stefanie C Linden, J Andrew Hughes and Anthony P Roberts from Wales used historical cohorts ( melancholia 1874-1924 & severe depressive disorder (1995-2005), both hospitalised patients) to compare the morbidity and mortality linked to melancholia between the two periods.
A panel of clinicians reviewed records from all admissions ( 1875-1924) for each patient and made retrospective diagnoses according to ICD-10 criteria.Data ( all with severe depression with or without psychosis) for 1995-2005 period was collected from an ongoing study of the incidence of service utilization and outcomes in the same geographical area.
597 patients were admitted for melancholia in the period 1875–1924.68% had psychotic symptoms. 203 patients were admitted in the modern era of which 38% had depression with psychosis. 28% in historical cohort were deemed to be in a poor or feeble state. 27% of that sample had an additional medical diagnosis. (7% had cardiovascular disease,6% had respiratory disease including TB,0.2% had cancer).50% of the modern sample had at least one hospital admission for physical disorder prior to psychiatric admission.63% have at least one medical condition.25% of modern sample had cardio vascular disorders, 19% musculo skeletal disorders,11% had endocrine and 8% had cancer.
The crude death rates in the historical sample for the year of admission for men are 111 per 1000, and 155 per 1000 women.The crude death rates in the contemporary sample for year of admission for men are 91 per 1000, and for women 70 per 1000.Death in historical sample was highest in the first 6 months.These deaths were from respiratory causes (33% of deaths) or exhaustion (24% of deaths) .Of the 597 patients, 136 were considered suicidal on admission or at some point during their stay and there were 6 suicides within a year of admission. In the modern sample the commonest cause of death in the year after first admission was suicide (5 of 16 deaths).
This study indicate that in the late nineteenth and early twentieth centuries melancholia had a clearly defined course with patients typically recovering within 6 months with no/less recurrences; which is not true of severe depressive disorders today . Suicide rates in contemporary are 2.5 times higher for the first year of service contact .20–25% of deaths in the contemporary sample were deaths from suicide. Briefer lengths of stay associated with contemporary admissions might be a factor in increased suicide risk. Contemporary patients with melancholia almost tripled their rates of re-admission compared with historical patients.
There were only few cancers in the historical sample, even in those patients who die after decades in hospital.The contemporary sample shows a high co-incident cancer rate at the time of diagnosis. Death rates for melancholia is high in both time frames.
The severely depressed patients in the contemporary sample represent less than a third of admissions for ‘depression’ today. Changes in social supports/ expectations/ risk perception etc would explain some of this.
Comparing historical samples are always fraught with problems. This study is interesting in that it ask a key question: With the modern psychiatric services, have we made real progress? Authors think that a case can be made that patients did better a century ago, if the lack of recurrences in the historical sample is taken into account.
Summary of the article:
The morbidity and mortality linked to melancholia: two cohorts compared, 1875− 1924 and 1995−2005
Margaret Harris, Fiona Farquhar, David Healy, Joanna C Le Noury, Stefanie C Linden, J Andrew Hughes and Anthony P Roberts. History of Psychiatry 2013 24: 3