RESUMO
OBJECTIVES: To develop and verify the Psychosomatic Symptom Scale (PSSS) among psychosomatic patients and the cut-off value of PSSS in distinguishing psychosomatic patients from health controls. METHODS: The PSSS was drafted by an expert workgroup. 996 patients and 366 controls from 14 general hospitals in China were recruited to complete PSSS, Patient Health Questionnaire-15 (PHQ-15) and Symptom Checklist-90 (SCL-90). Student's t-test, Kruskal-Wallis test, Cronbach's α, Spearman's correlation, and confirmatory factor analysis (CFA) were used to verify the PSSS. Receiver operating characteristic (ROC) analyses were used to determine the cut-off value. RESULTS: Cronbach α of PSSS was 0.907. The PSSS was significantly correlated with SCL-90 somatization subscale (r = 0.682, P < 0.001) and PHQ-15 (r = 0.724, P < 0.001). CFA supported the theoretical two-factor structure of the PSSS, with comparative fit index (CFI) = 0.979, Tucker-Lewis index (TLI) = 0.977, root mean square error of approximation (RMSEA) = 0.039 (90% CI: 0.035-0.042), and standardized root mean residual (SRMR) = 0.054. As the sum score of PSSS was significantly higher in female, cut-off values were determined as 11 in females and 10 in males respectively. CONCLUSIONS: The PSSS is a reliable and valid instrument for measuring psychosomatic symptoms.