Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Incident benzodiazepine and Z-drug use and subsequent risk of alcohol- and drug-related problems : A nationwide matched cohort study with co-twin comparison

Wang, Xinchen ; Chang, Zheng ; Molero, Yasmina ; Isomura, Kayoko ; Fernández de la Cruz, Lorena ; Lichtenstein, Paul ; Kuja-Halkola, Ralf ; D’Onofrio, Brian M. ; Quinn, Patrick D. and Larsson, Henrik , et al. (2025) In Journal of Psychopharmacology
Abstract

Background: Despite considerable interest in the consequences of benzodiazepine and benzodiazepine-related Z-drug (BZDR) use, little is known about whether and how initiation of BZDR treatment relates to the development of alcohol- and drug-related problems. Aims: This study aimed to examine the association of incident BZDR dispensing with subsequent alcohol- and drug-related problems. Methods: This nationwide register-based study included demographically matched and co-twin control cohorts. Among all Swedish residents aged older than 10 years and BZDR-naïve by 2007, 960,430 BZDR-recipients with incident dispensation in 2007–2019 and without any recorded pre-existing substance-related conditions were identified and matched (1:1) to... (More)

Background: Despite considerable interest in the consequences of benzodiazepine and benzodiazepine-related Z-drug (BZDR) use, little is known about whether and how initiation of BZDR treatment relates to the development of alcohol- and drug-related problems. Aims: This study aimed to examine the association of incident BZDR dispensing with subsequent alcohol- and drug-related problems. Methods: This nationwide register-based study included demographically matched and co-twin control cohorts. Among all Swedish residents aged older than 10 years and BZDR-naïve by 2007, 960,430 BZDR-recipients with incident dispensation in 2007–2019 and without any recorded pre-existing substance-related conditions were identified and matched (1:1) to non-recipients from the general population. Twin BZDR-recipients (n = 12,048) were linked to 12,579 unexposed co-twins. Outcomes included alcohol and drug use disorders, poisoning, deaths, and related suspected criminal offences. Flexible parametric survival models estimated outcome risks across up to 14 years of follow-up. Results: In the demographically matched cohort (60% women, median age at BZDR initiation 51 years), incidence rates in BZDR-recipients and non-recipients (per 1000 person-years) were 5.60 versus 2.79 for alcohol-related and 4.15 versus 1.23 for drug-related problems, respectively. In fully adjusted models, relative risks were increased for alcohol- and drug-related problems (adjusted hazard ratio (95% confidence interval): 1.56 (1.53–1.59) and 2.11 (2.05–2.17), respectively). The risks persisted within the co-twin comparison, different follow-ups, and all additional sensitivity analyses. Conclusions: BZDR initiation was associated with a small but robust increase in absolute and relative risks of developing alcohol- and drug-related problems. The findings contribute to evidence base for making decisions on BZDR treatment initiation.

(Less)
Please use this url to cite or link to this publication:
@article{3bda3b36-b255-4eee-b757-f0cfe9bb1360,
  abstract     = {{<p>Background: Despite considerable interest in the consequences of benzodiazepine and benzodiazepine-related Z-drug (BZDR) use, little is known about whether and how initiation of BZDR treatment relates to the development of alcohol- and drug-related problems. Aims: This study aimed to examine the association of incident BZDR dispensing with subsequent alcohol- and drug-related problems. Methods: This nationwide register-based study included demographically matched and co-twin control cohorts. Among all Swedish residents aged older than 10 years and BZDR-naïve by 2007, 960,430 BZDR-recipients with incident dispensation in 2007–2019 and without any recorded pre-existing substance-related conditions were identified and matched (1:1) to non-recipients from the general population. Twin BZDR-recipients (n = 12,048) were linked to 12,579 unexposed co-twins. Outcomes included alcohol and drug use disorders, poisoning, deaths, and related suspected criminal offences. Flexible parametric survival models estimated outcome risks across up to 14 years of follow-up. Results: In the demographically matched cohort (60% women, median age at BZDR initiation 51 years), incidence rates in BZDR-recipients and non-recipients (per 1000 person-years) were 5.60 versus 2.79 for alcohol-related and 4.15 versus 1.23 for drug-related problems, respectively. In fully adjusted models, relative risks were increased for alcohol- and drug-related problems (adjusted hazard ratio (95% confidence interval): 1.56 (1.53–1.59) and 2.11 (2.05–2.17), respectively). The risks persisted within the co-twin comparison, different follow-ups, and all additional sensitivity analyses. Conclusions: BZDR initiation was associated with a small but robust increase in absolute and relative risks of developing alcohol- and drug-related problems. The findings contribute to evidence base for making decisions on BZDR treatment initiation.</p>}},
  author       = {{Wang, Xinchen and Chang, Zheng and Molero, Yasmina and Isomura, Kayoko and Fernández de la Cruz, Lorena and Lichtenstein, Paul and Kuja-Halkola, Ralf and D’Onofrio, Brian M. and Quinn, Patrick D. and Larsson, Henrik and Brikell, Isabell and Hellner, Clara and Hasselström, Jan and Jayaram-Lindström, Nitya and Mataix-Cols, David and Sidorchuk, Anna}},
  issn         = {{0269-8811}},
  keywords     = {{alcohol use disorders; benzodiazepines; drug use disorders; Z-drugs}},
  language     = {{eng}},
  publisher    = {{SAGE Publications}},
  series       = {{Journal of Psychopharmacology}},
  title        = {{Incident benzodiazepine and Z-drug use and subsequent risk of alcohol- and drug-related problems : A nationwide matched cohort study with co-twin comparison}},
  url          = {{http://dx.doi.org/10.1177/02698811251373069}},
  doi          = {{10.1177/02698811251373069}},
  year         = {{2025}},
}