Comparative psychometric properties of the short versions of the SCL-90-R for patients with substance use disorder

Abstract

Background: The prevalence of comorbid psychopathological disorders and associated problems is high among patients with substance use disorder (SUD). The Symptom Checklist 90-R (SCL-90-R) is one of the most widely used self-report instruments to assess psychopathology. This study examined the psychometric properties of the SCL-90-R and its short-form versions for patients with SUD. Method: The values of the short versions were calculated a posteriori based on the SCL-90-R scores. Confirmatory factor analysis (CFA) was conducted using the diagonally weighted least squares estimator to assess construct validity. Reliability was assessed via the ordinal alpha coefficient, McDonald’s omega coefficient and intraclass correlation coefficient (ICC). Results: The SCL-90-R showed acceptable model fit (χ2/df = 1.916, CFI = 0.940, TLI = 0.938, RMSEA = 0.035). Among the short versions, the BSI-53, SA-45, SCL-27, HSCL-25, and BSI-18 demonstrated good to excellent fit. The depression scale showed the highest reliability across the versions. Pearson correlations and ICCs revealed strong associations between the SCL-90-R and its short forms, although the agreement declined with shorter versions. Despite the excellent fit of the SCL-6, it had lower reliability. Conclusion: These results provide evidence of the psychometric robustness of short versions of the SCL-90-R and support their use to assess comorbid psychopathology in people with SUD.