Original Research

Assessing the clinical utility of the severity dependence scale for benzodiazepine use disorder

Karishma Lowton, Gaveeta Chiba
South African Journal of Psychiatry | Vol 27 | a1571 | DOI: https://doi.org/10.4102/sajpsychiatry.v27i0.1571 | © 2021 Karishma Lowton, Gaveeta Chiba | This work is licensed under CC Attribution 4.0
Submitted: 04 June 2020 | Published: 28 July 2021

About the author(s)

Karishma Lowton, Department of Psychiatry, Faculty of Health, University of the Witwatersrand, Johannesburg, South Africa
Gaveeta Chiba, Department of Psychiatry, Faculty of Health, University of the Witwatersrand, Johannesburg, South Africa


Background: Benzodiazepines are often used as a part of mental health pharmacological management; however, often when prescribed for extended periods, they increase the risk of benzodiazepine use disorder (BUD). Clinical interviews are at the centre of diagnosing this disorder. However, in addition to clinical assessment a simple, validated questionnaire conducted by any healthcare professional may aid in screening for BUD and referral for further management.

Aim: To compare the accuracy of the severity dependence scale (SDS) as a screening tool for BUD against the standard clinical interviews using the Diagnostic and Statistical Manual of Mental Disorders, edition 5, (DSM 5) checklist amongst benzodiazepine users with primary psychiatric disorders.

Setting: Outpatient psychiatric clinic in South Rand Hospital, Johannesburg, South Africa.

Methods: A cross-sectional study was conducted, once informed consent was attained, looking at demographic and clinical profiles of benzodiazepine users. Clinical interviews were conducted in 81 patients who completed the SDS. In comparing the results of the SDS and clinical interview outcomes, chi-square tests were used to determine an association between categorical variables. A receiver-operating characteristic (ROC) curve was generated in determining the cut-off score in the SDS with the highest sensitivity and specificity.

Results: This study indicated that a cutoff score of greater than or equal to six of the SDS showed 86% sensitivity and 90.3% specificity compared to a diagnosis of BUD made with clinical interview. The only categorical variables of marginal significance (p~0.06) in comparison to a BUD diagnosis were with benzodiazepine type (oxazepam) and longer duration of use (greater than 24 months).

Conclusion: This study identified the SDS as a useful screening tool for BUD with a high sensitivity and specificity compared to interview outcomes. Statistically, correlates were identified between duration and type of benzodiazepine prescribed and BUD suggesting emphasis on these factors when prescribing benzodiazepines.


benzodiazepine; benzodiazepine addiction; benzodiazepine use disorder; severity dependence scale; South Africa


Total abstract views: 2866
Total article views: 2650

Crossref Citations

No related citations found.