Reliability, validity, and screening performance of the Chinese version of the McLean Screening Instrument for Borderline Personality Disorder in a psychiatric clinical sample.

Introduction: Evidence regarding the psychometric properties of the Chinese version of the McLean Screening Instrument for Borderline Personality Disorder (MSI-BPD) in psychiatric clinical populations remains limited. This study evaluated the reliability, validity, and screening performance of the Chinese MSI-BPD in a psychiatric clinical sample and examined a theory-driven four-factor model. Methods: The previously translated Chinese MSI-BPD was reviewed for comprehensibility and clinical appropriateness through expert review, cognitive debriefing, and pilot testing. In this cross-sectional study, 375 outpatients and inpatients from a tertiary psychiatric hospital in Zhuhai, China, completed the Chinese MSI-BPD, with the Revised Diagnostic Interview for Borderlines (DIB-R) used as the reference standard. Confirmatory factor analysis was conducted using the weighted least squares mean- and variance-adjusted estimator. Five CFA models were compared: a one-factor model, a classical four-factor model, a literature-derived modified four-factor model, the theory-driven four-factor model, and a higher-order model. Reliability, criterion-related validity, and screening performance were evaluated. Results: The theory-driven four-factor model showed the most favorable fit indices among the admissible models (χ²/df = 1.409, CFI = 0.995, TLI = 0.992, RMSEA = 0.033, SRMR = 0.043). The higher-order model also showed good fit, supporting a general BPD factor underlying the four symptom domains. In the theory-driven model, standardized factor loadings ranged from 0.622 to 0.911. The Chinese MSI-BPD total score was strongly correlated with the DIB-R total score (r = 0.829, p < 0.01). McDonald's omega, Guttman split-half reliability, and test-retest reliability for the total scale were 0.884, 0.894, and 0.851, respectively. The Youden-based optimal cutoff was 6.5, corresponding to a clinically practical threshold of ≥7 in the present sample, with an AUC of 0.884, sensitivity of 88.83%, specificity of 71.91%, and Cohen's kappa of 0.612. Conclusions: The Chinese MSI-BPD total score demonstrated good reliability, criterion-related validity, and screening performance in this psychiatric clinical sample. The theory-driven four-factor model may provide a clinically useful but provisional interpretive framework; however, the total score should remain the primary screening index. The instrument may serve as a brief preliminary screening tool for BPD in tertiary psychiatric clinical settings in China, but not in general psychiatric or community populations without dedicated validation.