In his brief ‘refreshment’, Reference CastleCastle (2012) considers the very topical subject of ‘schizoaffective disorder’. Because of their low reliability and questionable validity, there is need for a revision of the current diagnostic concepts of schizoaffective disorder, but developments in ICD-11 and DSM-5 are as yet unclear. How we might see this disorder in future nosology is an interesting area of discussion. Psychiatric nosology can be conceptualised in terms of three models: Kraepelin’s dichotomous approach, the dimensional diagnostic approach and the multiple diagnostic approach (Fig. 1).

FIG 1 Nosological models of psychoses.

A dichotomous classification of non-organic psychoses is not compatible with recent neurobiological findings. Furthermore, various studies of psychopathological symptoms have also failed to confirm the dichotomous classification. Rather, they point to a continuous-spectrum model of functional psychoses using a more dimensional diagnostic approach. However, at this point, switching from a categorical classification to a purely dimensional model entails hazards in clinical practice. This is because current treatment guidelines are based on categorical diagnoses, and the psychopathological syndrome dimensions have low stability over the long-term course of the illness. An alternative would be subdivision of the classic categories into multiple diagnostic entities. In such a model, distinct aetiological factors, psychopathological characteristics, neuropathological findings and outcomes can be entertained simultaneously.

Further, one can classify non-organic psychoses on the basis of course and outcome, which might improve the individual treatment of psychiatric disorders. Following Kasanin’s original concept (Reference KasaninKasanin 1933), future diagnostic systems could try to establish reliable criteria for schizoaffective disorder that predict a favourable outcome and allow a differentiation from chronic schizophrenic disorders. However, it would be interesting to see the relationship of schizoaffective disorder to brief and acute psychoses. Some authors have argued against this approach because distinct categories for remitting psychoses such as schizoaffective disorders could contribute to increasing the stigma of schizophrenia.

The forthcoming ICD-11 and DSM-5 might follow a ‘triaxial’ classification similar to that proposed by Reference Essen-MollerEssen-Moller (1962). Reference Jager, Frasch and BeckerJager et al (2008) speculated that they will introduce the complementary use of dimensional and categorical concepts. A dimensional concept can be helpful in describing the cross-sectional clinical picture, whereas a categorical approach can specify course and outcome. Psychopathological course types can be considered as prototypes within a continuous biological spectrum of schizophrenic and affective disorders. An additional ‘axis’ could comprise information about aetiology. Many authors (e.g. Reference Lake and HurwitzLake 2007; Reference Malhi, Green and FagioliniMalhi 2008) propose the omission of the current concept of schizoaffective disorder from ICD-11 and DSM-5. Thus, a model of multiple diagnostic entities may be compatible with retaining the diagnostic category of schizoaffective disorder in the nosology.