[The difference between depression and melancholia: two distinct conditions that were combined into a single category in DSM-III].

In DSM-III (1980), depressive states of neurosis and those of manic-depressive illness (melancholia or endogenous depression) were combined into the single category "major depression," which is the progenitor of "major depressive disorder" in DSM-IV-TR (2000). According to Hamilton, the word "depression" is used in three different ways. In common speech, it is used to describe the state of sadness that all persons experience when they lose something of importance to them. In psychiatry, the word is used to signify an abnormal mood, analogous to the sadness, unhappiness, and misery of everyday experiences. Moreover, the depression discussed in psychiatry often has another quality that makes it distinctive, and this quality appears to be related to an inability to experience any pleasure (anhedonia) regardless of experience. Accordingly, we classify these three uses of the term "depression" into sadness, depression, and melancholia in order of appearance within this paper. According to DSM-IV-TR criteria for major depressive disorder, depression corresponds closely to A1 "depressed mood", while melancholia is roughly compatible with A2 "markedly diminished interest or pleasure." Depression and melancholia differ in terms of origin, psychopathology, and therapy. Before DSM-III, depression had not been considered as a diagnosis, but was a ubiquitous symptom that was seen in such conditions as neurasthenia, psychasthenia, nervousness, and neurosis. Melancholia has a history that reaches back to Hippocratic times. Its modern meaning was established based on Kraepelin's manic-depressive illness. Depression is a deepened or prolonged sadness in everyday life, but melancholia has a distinct quality of mood that cannot be interpreted as severe depression. In modern times, depression has been treated with a diverse range of methods, including rest, talk therapy, amphetamines (1930s), meprobamate (1950s), and benzodiazepines (1970s). Melancholia has primarily been treated with somatic therapy, such as electroconvulsive therapy, and tricyclic antidepressants. When preparing diagnostic criteria for DSM-III, Spitzer referred not to DSM-II but to Feighner's (1972) criteria as a model because Feighner's operational criteria were considered to be effective in establishing inter-rater reliability. At the outset, Spitzer established Research Diagnostic Criteria (RDC, 1975), which he revised in 1978. In the first edition of RDC, Spitzer adopted most of the Feighner criteria, including essential criteria A "dysphoric mood" and eight optional criteria (B1-B8). However, he reduced the minimal morbid duration for diagnosis. Moreover, for the purpose of excluding neurosis from the diagnostic criteria, Spitzer eliminated the distinction between primary and secondary depression, which had been used to differentiate melancholia from depression. In the revised RDC, Spitzer upgraded optional criteria B5 "loss of pleasure or interest" to one of the essential criteria A with "dysphoric mood." This revision reflects the fact that "loss of pleasure or interest" has been designated as an essential feature of Klein's concept of "endogenomorphic depression" (1974), which is equivalent to melancholia or endogenous depression. At that time, depression and melancholia were completely amalgamated into a single category. DSM-III followed almost all of the revisions in the revised RDC and accepted the bipolar-unipolar dichotomy. However, Klein's endogenomorphic depression was downgraded to the specifier "with melancholia", which has been used only rarely. Thus, as depression and melancholia were fused into major depressive disorder, we have only limited evidence of the efficacy of pharmacotherapy and psychotherapy. DSM-IV divided major depression into major depressive disorder and bipolar II disorder. Consequently, some depression and some melancholia were moved from unipolar depression to bipolar disorder, although the bipolar-unipolar dichotomy was proposed for manic-depressive illness and recurrent unipolar melancholia, but not depression. Therefore, we suspect that we will not obtain strong therapeutic evidence for bipolar II disorder as well. Our proposals are as follows: give up the unitarian view of depression and melancholia and accept the binarian view; and restrict the bipolar-unipolar dichotomy to manic-depressive illness and unipolar melancholia.