Psychometrics of BE-PSD-V2.0 and its correspondence with PANSS and CGI-S

Aim: We aimed to examine the psychometric properties of the second version of the Brief Evaluation of Psychosis Symptom Domains (BE-PSD-V2.0) and its correspondence with the Positive and Negative Syndrome Scale (PANSS) and Clinical Global Impression–Severity scale (CGI-S) scores. Methods: This cross-sectional study was conducted at 18 institutes in Japan. We performed correlational analyses to test the convergent and divergent validity of the BE-PSD-V2.0. Inter-rater reliability was assessed using intraclass correlation coefficients (ICCs). The equipercentile equating method was employed to examine the linking of the BE-PSD-V2.0 with the PANSS and the CGI-S. Results: A total of 353 patients with schizophrenia spectrum disorders were included. Spearman's rank correlation coefficients between the BE-PSD-V2.0 item scores and their corresponding PANSS factor scores ranged from 0.64 to 0.86, indicating sufficient convergent validity. These specific correlation coefficients were consistently the highest among all item-to-factor comparisons, indicating sufficient divergent validity. In 34 patients, ICCs for the BE-PSD-V2.0 item and total scores ranged from 0.60 to 0.94, showing sufficient inter-rater reliability. Conversion tables were established between the BE-PSD-V2.0 total score and the PANSS total score, between the BE-PSD-V2.0 total score and the CGI-S score, and between the BE-PSD-V2.0 item scores and their corresponding PANSS factor scores. A correspondence between symptomatic remission criteria using the PANSS and BE-PSD-V2.0 was confirmed. Conclusions: The BE-PSD-V2.0 possesses sufficient psychometric properties and corresponds to the PANSS and CGI-S. It is a sound and reliable brief rating scale for assessing symptoms in schizophrenia spectrum disorders, and its established conversion correspondences facilitate score interpretation within existing clinical paradigms for both clinical practice and research.