Trajectories and antecedents of treatment response over time in early-episode psychosis

Background: Little is known about the extent of heterogeneity of symptomatology in treated early-onset psychosis. The current study aims to quantify the extent of heterogeneity in trajectories of treated symptom severity in early-episode psychosis and their antecedents. Methods: Data were from 491 persons with early-episode psychosis from a clinical trial of haloperidol and risperidone. Positive and Negative Syndrome Scale (PANSS) administrations were used to measure symptom severity trajectories for (a) rapid treatment response scores over 4 weeks and (b) medium-term course over 24 weeks. Baseline antecedents included sex, Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, diagnosis, age of onset, the Premorbid Adjustment Scale, and a cognitive test battery. Symptom severity trajectories were calculated with mixed mode latent class regression modeling from which groups were derived. Results: Five groups based on PANSS scores over time were identified. Over 4 weeks, 3 groups with varied baseline PANSS scores (54-105) did not surpass 30% PANSS improvement. Another group improved and then was stable (n = 76,15.3%), and another showed marked improvement (n = 94,18.9%). Logistic regression showed that membership in the best response trajectory was associated with not having a diagnosis of schizophrenia, good premorbid functioning, and higher cognitive functioning, whereas membership in the poor response trajectory was associated with earlier age of onset and poorer cognitive functioning. Conclusion: Amelioration generally characterizes treated symptom severity. Age of onset, diagnosis, cognitive functioning, and premorbid functioning have prognostic value in predicting treatment response trajectories.