A dimensional approach to psychosis: identifying cognition, depression, and thought disorder factors in a clinical sample.

Traditional classification systems based on broad nosological categories do not adequately capture the high heterogeneity of mental illness. One possible solution to this is to move to a multi-dimensional model of mental illness, as has been proposed by the Research Domain Criteria and Hierarchical Taxonomy of Psychopathology frameworks. In this study, we explored the dimensional structure of psychotic disorders. We focused on the question whether combining measures of psychosis with cognitive and depression-related measures results in meaningful, clinically relevant, and valid latent dimensions in a sample of early psychosis (n = 113) and chronic schizophrenia patients (n = 43, total n = 156). We used exploratory factor analysis to identify the symptom dimensions in the Lausanne Psychosis data, a multi-modal prospective data set that includes a broad behavioral assessment of patients diagnosed with psychotic disorders. We evaluated the validity of these dimensions by regressing them to several functioning measures. Our analysis revealed three dimensions: Cognition, Depression/Negative, and Thought Disorder, explaining 49.2% of the variance. They were related to measures of functioning, the R² ranging between 0.38 and 0.42. This study advances the development of a multi-dimensional characterization of psychotic disorders by identifying three symptom dimensions with predictive validity in people with psychosis.