Screening accuracy of the German version of the benzodiazepine self-report questionnaire

Compr Psychiatry. 2025 May 18;141:152607. doi: 10.1016/j.comppsych.2025.152607. Online ahead of print.

ABSTRACT

BACKGROUND AND AIMS: We aimed to assess the accuracy for the German version of the Benzodiazepine Dependence Self-Report Questionnaire (Bendep-SRQ-GV) as a possible screening instrument for benzodiazepine or Z-drug substance use disorder.

METHOD: This study includes a convenience sample of adults who had used benzodiazepines or Z-drugs at least once in the last year. Participants were recruited through notice boards, social media, newspaper, and a digital health information platform. All participants answered the Bendep-SRQ-GV and were assessed for benzodiazepine or Z-drug substance use disorder using the Structured Clinical Interview for DSM-5 Disorders – Clinical Version. We performed receiver operating characteristics curve analyses and calculated the area under the curve (AUC).

RESULTS: 42 out of 60 participants fulfilled the criteria for benzodiazepine or Z-drug substance use disorder according to Structured Clinical Interview. AUCs varied among the 4 subscales of the Bendep-SRQ-GV: Problematic Use (0.893 [0.819-0.975]), Preoccupation (0.751 [0.609-0.892]), Lack of Compliance (0.804 [0.699-0.919]), Withdrawal (0.743 [0.622-0.864]) and Total Score (0.860 [0.762-0.963]). For Problematic Use and Total Score selected cut-offs (5.5 and 16.5) showed best results: sensitivity (83 % [71-93 %] and 79 % [67-90 %]), high specificity (83 % [61-100 %] and 83 % [67-100 %], positive predictive value (92 % [84-100 %] and 92 % [83-100 %]), and negative predictive value (68 % [54-85 %] and 63 % [50-77 %]).

DISCUSSION: The subscale Problematic Use and the Total Score seem promising for detecting benzodiazepine or Z-drug substance use disorder in adults. Precise cut-offs should be determined with a larger sample. The Bendep-SRQ-GV could be of great value as a screening questionnaire in clinical settings, counselling sites, and for scientific purposes.

PMID:40408792 | DOI:10.1016/j.comppsych.2025.152607