Mortality among critically ill patients remains unacceptably high despite advances in life–support technologies and standardized care bundles [1], [2]. Identification of modifiable risk factors present before intensive care unit (ICU) admission is crucial for improving early prognostication and informing targeted prevention strategies. In particular, exposure to central nervous system depressants such as benzodiazepines and Z–drugs—which are widely prescribed for insomnia, anxiety, and procedural sedation—may influence critical illness trajectories via respiratory, hemodynamic, and immunomodulatory effects [3], [4]. These agents primarily act as Benzodiazepine Receptor Agonists (BZRA) by potentiating γ–aminobutyric acid (GABA) activity, which is the core pathway impacting critical organ systems.

Benzodiazepines enhance GABA activity, leading to dose–dependent respiratory depression and hypoventilation, which can result in hypercapnia and prolonged mechanical ventilation in ICU populations [5], [6]. Z–drugs (e.g., zolpidem, zopiclone) share similar GABAergic mechanisms and have been linked to adverse respiratory events, especially in older adults with sleep–disordered breathing [3]. Beyond direct respiratory compromise, BZRA use can also induce hemodynamic instability by reducing systemic vascular resistance [7]. Furthermore, BZRA exposure has been associated with immunosuppressive effects—such as impaired neutrophil function and reduced cytokine production—that could exacerbate infectious processes like sepsis and multi–organ dysfunction [4], [8]. However, existing investigations have largely focused on in–ICU sedative administration, with few studies examining whether outpatient or pre–hospital BZRA exposure portends worse outcomes after ICU admission.

To our knowledge, no large–scale, population–based cohort study has evaluated the association between pre–ICU benzodiazepine or Z–drug use and subsequent short– and intermediate–term mortality in critical care settings. Clarifying this relationship is important because BZRA prescribing is common in the community, and pre–hospital medication histories are routinely available in administrative claims datasets. Furthermore, characterizing the independent prognostic impact of BZRA exposure may guide clinicians in pre–ICU risk stratification and inform deprescribing efforts in high–risk patients.

Accordingly, we conducted a nationwide, retrospective cohort study using South Korea’s National Health Insurance Service database to assess whether benzodiazepine or Z–drug exposure preceding ICU admission is associated with mortality.