Predictive Validity of the 5-Factor PANSS Model in Acute Schizophrenia

Background: Predicting early treatment response in acute schizophrenia is critical yet challenging. This observational study aimed to determine whether improvements in specific symptom domains after 2 weeks predict overall response at 6 weeks in patients treated with aripiprazole (ARI) or brexpiprazole (BRE). Methods: We included 65 patients (34 antipsychotic-naïve and 31 antipsychotic-free recurrent) treated with flexible doses of ARI or BRE. Benzodiazepines were used in 41 patients (64.1%), and their use did not significantly impact prediction. The Positive and Negative Syndrome Scale (PANSS) and Clinical Global Impression-Improvement were used to assess responses. Results: Receiver operating characteristic analysis revealed area under the curve values for PANSS total (PANSS-T), negative, excitement, cognitive, positive, and depressive/anxiety components of 0.788, 0.783, 0.603, 0.746, 0.738, and 0.735, respectively. Kendall’s tau correlation and Cramer’s V revealed significant predictive relationships for PANSS-T (0.413, P <0.001), negative (0.411, P <0.001), and therapeutic response dichotomized by this score (0.573, P <0.001), cognitive (0.364, P <0.001), positive (0.344, P <0.001), and depression/anxiety (0.344, P =0.001), but not for excitement (0.15, P =0.151). Benzodiazepine use did not significantly impact these predictive associations. Conclusions: This study is the first to evaluate the predictive validity of the PANSS 5-factor model in patients with acute schizophrenia treated with ARI/BRE. Early symptom improvements, particularly in negative domains, are stronger predictors of overall response, while excitement symptom improvements showed a weaker relationship. These findings underscore the importance of early, symptom-specific assessments to optimize treatment strategies for acute schizophrenia. Further studies with larger samples are necessary to validate these results.