Dimensional vs. categorical diagnosis in psychosis

Objective: To compare dimensional measures vs. categorical diagnosis of psychopathology in their prediction of disability and outcome in psychotic illness. Method: A community study of 980 subjects with psychotic illness was included in the Australian National Survey of Mental Health and Wellbeing. The study instrument [including the Schedules for Clinical Assessment in Neuropsychiatry (SCAN)] yielded symptom data along with Operational Criteria for Psychotic Illness (OPCRIT) diagnoses by several diagnostic schemata. Factor analysis of symptoms yielded five dimensions of psychopathology (positive symptoms, negative symptoms, dysphoria, mania and substance use). Variance in service demand and disability explained by the diagnostic schemata was compared with the variance explained by the dimensions of psychopathology. Results: Dimensional measures of psychopathology explained more of the variance in service demand, dysfunctional behaviour, social adaptation and global occupation and function. Only use of support services and illness course were better predicted by categorical diagnosis. Dimensional scores explained significant extra variance when added to categorical diagnoses. Conclusion: Dimensional measures of psychopathology explain more variance in behaviour, disability and outcome than does categorical diagnosis in functional psychosis. Dimensions provided significant extra information not provided by diagnosis and would be a more useful basis for clinical management.