Cross-validation of the benzodiazepine dependence self-report questionnaire in outpatient benzodiazepine users

Abstract

The aim of this study was to cross-validate the Benzodiazepine Dependence Self-Report Questionnaire (Bendep-SRQ), which reflects the severity of benzodiazepine (BZD) dependence. The Bendep-SRQ, Symptom Checklist-90 (SCL-90) Schedules for Clinical Assessments in Neuropsychiatry (SCAN), and Addiction Severity Index-Revised (ASI-R) were administered to 102 general practice (GP) patients and 126 psychiatric outpatients who were using BZDs. The scalability and reliability of the Bendep-SRQ scales were reassessed in terms of Rasch homogeneity, subject discriminability, item discriminability, and test-retest stability. Present and original Rasch item orders were compared to evaluate construct validity. A matrix of all measures was factor-analyzed to assess concurrent and discriminant validity. The scalability of the Bendep-SRQ scales was confirmed. The reliability results were fairly good. Present and original Rasch item orders corresponded. The Bendep-SRQ scales and concurrent measures had high loadings on one factor, the discriminant measures on two other factors. In spite of some differences with respect to sociodemographic characteristics and pattern of BZD use, the cross-validation results agreed well with the results of the original study on the Bendep-SRQ. The Bendep-SRQ has presently acquired enough support of favorable and consistent results for clinical and scientific use.