Exploring the borders of the schizoaffective spectrum: A categorical and dimensional approach

Background: Schizoaffective disorder has long been considered as an intermediate condition between major mood disorders and schizophrenia, however, the nature of the relationship to these diagnoses remains unclear. We aimed at examining the nature of such a relationship in a mixed sample of psychotic disorders by using a dimensional and categorical approach to psychopathology. Methods: Six-hundred and sixty psychotic inpatients were assessed for lifetime ratings of mania, depression, psychosis and incongruence, diagnosed according to Research Diagnostic Criteria, and classified as having nonaffective psychosis without mood syndromes (n=429), nonaffective psychosis with mood syndromes (n=101), schizoaffective disorder mainly schizophrenic (n=41), schizoaffective disorder mainly affective (n=42) or mood disorder with psychotic symptoms (n=47). We tested for associations of illness-related features including risk factors, premorbid, clinical and outcome variables with classes of disorders and lifetime ratings of psychopathology, and examined the relative contribution of categorical and dimensional representations of psychopathology in explaining disease characteristics. Results: While categories at the extreme end of the psychotic spectrum meaningfully differed across a number of the illness-related variables, no substantial discontinuity was apparent between adjacent categories of psychotic disorders. Risk factors, premorbid adjustment, clinical features and impairment appeared to be present in a mostly monotonic continuous fashion from nonaffective psychoses to mood disorders with psychotic features. The overall association pattern of illness-related variables with mood and psychotic syndromes was largely independent of specific diagnostic categories, and the dimensional approach was neatly superior to the traditional diagnostic approach in explaining the characteristics of the illness. Limitations: This was a cross-sectional study with retrospective assessment of illness-related-variables and lifetime psychopathology. Conclusion: The results are compatible with the notion of the schizoaffective spectrum and with a continuum model of the psychotic illness.