Psychometric validation of the brief psychosomatic symptom scale (BPSS) in Chinese adolescents: Screening cutoff and associated psychosocial factors.
Abstract
Objectives
The Brief Psychosomatic Symptom Scale (BPSS) has shown promising psychometric properties in adults, but its performance in adolescents remains under-examined. This study evaluated the BPSS in Chinese adolescents, estimated an empirical screening cutoff, and examined psychosocial factors associated with elevated psychosomatic symptom burden.
Methods
The BPSS was administered to 4665 school-based Chinese adolescents. Internal consistency evidence, associations with the 26-item Psychosomatic Symptoms Scale (PSSS), the Patient Health Questionnaire-15 (PHQ-15), and the Generalized Anxiety Disorder-7 (GAD-7), and construct validity via confirmatory factor analysis (CFA) were examined. Receiver operating characteristic (ROC) analysis was used to estimate a BPSS cutoff relative to a study-defined high somatic-symptom-burden criterion based on a modified PHQ-15. Multivariable logistic regression examined psychosocial correlates of BPSS-defined elevated symptom burden.
Results
The BPSS showed good internal consistency (Cronbach's α = 0.885) and strong positive associations with the PSSS, PHQ-15, and GAD-7 (all p < 0.001) providing convergent evidence while requiring caution because of scale derivation and symptom-content overlap. The correlated two-factor CFA showed excellent global fit (CFI = 0.998, RMSEA = 0.037), but the factor correlation was very high (r = 0.95), indicating limited empirical separation between the factors. A cutoff of ≥8 was selected in the total sample and in both male and female subsamples. Depression, poor family functioning, bullying, personal distress, and loneliness were significantly associated with elevated BPSS scores.
Conclusion
The BPSS showed promising internal-consistency and screening performance in this school-based sample. A score of 8 may be useful as an empirical screening threshold relative to the study-defined criterion, but it should not be interpreted as a diagnostic cutoff. The psychosocial findings represent cross-sectional associations rather than causal risk factors, and further validation in independent and more diverse samples is needed.