Background - There is good evidence that psychotic symptoms segregate into symptom dimensions. However, it is still unclear how these dimensions are associated with risk indicators and other clinical variables, and whether they have advantages over categorical diagnosis in clinical practice. We investigated symptom dimensions in a first-onset psychosis sample and examined their associations with risk indicators and clinical variables. We then examined the relationship of categorical diagnoses to the same variables.
Method - We recruited 536 patients as part of a population-based, incidence study of psychosis. Psychopathology was assessed using the Schedules for Clinical Assessment in Neuropsychiatry (SCAN). A principal axis factor analysis was performed on symptom scores. The relationship of dimension scores with risk indicators and with clinical variables was then examined employing regression analyses. Finally, regression models were compared to assess the contribution of dimensions versus diagnosis in explaining these variables.
Results - Factor analysis gave rise to a five-factor solution of manic, reality distortion, negative, depressive and disorganization symptom dimensions. The scores of identified dimensions were differentially associated with specific variables. The manic dimension had the highest number of significant associations; strong correlations were observed with shorter duration of untreated psychosis, acute mode of onset and compulsory admission. Adding dimensional scores to diagnostic categories significantly increased the amount of variability explained in predicting these variables; the reverse was also true but to a lesser extent.
Conclusions - Categorical and dimensional representations of psychosis are complementary. Using both appears to be a promising strategy in conceptualising psychotic illnesses.