Determination of the main risk factors for benzodiazepine dependence using a multivariate and multidimensional approach

Abstract

The aim of this study was to identify risk factors for benzodiazepine (BZD) dependence, such as sociodemographic variables, characteristics of BZD use, and psychiatric parameters, which to date have been found to relate inconsistently to indicators of BZD dependence such as chronic BZD use and BZD withdrawal symptoms. The Benzodiazepine Dependence Self-Report Questionnaire (Bendep-SRQ), Schedules for Clinical Assessment in Neuropsychiatry (SCAN), and Symptom Checklist-90 (SCL-90) were administered to 599 outpatients using BZDs. Regression analyses were conducted using BZD dependence diagnoses and severity scales as dependent variables. BZD dependence diagnoses were only predicted by being a self-help patient and long BZD elimination half-life (for only the DSM-III-R). The main predictors of BZD dependence severity, as measured by the ICD-10, DSM-III-R scales, and Bendep-SRQ Rasch scales, were in decreasing order: (1) being a self-help patient; (2) higher BZD dose, longer duration of BZD use, younger age; and (3) non-native cultural origin, lower level of education, being in outpatient treatment for alcohol and/or drug dependence, and the interaction of BZD dose with duration of BZD use. We conclude that a limited number of recognizable risk factors appear to predict the severity of BZD dependence. Additional administration of a specific BZD dependence instrument is recommended to confirm suspected BZD dependence and guide further clinical decision-making.