Predicting high‐risk novel benzodiazepines: A surveillance study of novel benzodiazepine detections and a case series of bromazolam exposures in New South Wales, Australia

Abstract Background and aims Novel benzodiazepines are increasingly detected in poisonings. Their relative toxicities remain poorly defined, complicating public health responses. Bromazolam has emerged in recent reports of severe poisoning. We sought a method to identify higher‐risk novel benzodiazepines, and to describe the clinical effects of bromazolam exposures. Design A prospective observational surveillance study examined time trends in laboratory‐confirmed detections of benzodiazepines of interest (novels plus alprazolam) using police drug seizure data and a clinical surveillance program from January 2019 to December 2024. A case series described hospitalised patients with confirmed bromazolam exposure between January 2022 and December 2024. Setting Population‐based study in New South Wales, Australia. Cases All laboratory confirmed detections of benzodiazepines of interest in blood, urine or physical drug samples were included. Forty‐eight hospitalised patients with confirmed bromazolam exposure were included in the case series, and eight additional hospitalised patients with alprazolam exposures. Measurements Annual detections of benzodiazepines of interest in clinical and physical drug samples were recorded. Relative toxicity was estimated with a pooled risk ratio (RR) of clinical to physical detections for each benzodiazepine. A minimum clinical dataset described the demographics, clinical effects, drug co‐detections, treatments and outcomes of bromazolam and alprazolam‐related hospital presentations. Secondary analyses measured the relationship between clinical effects and bromazolam blood concentrations and opioid co‐detection. Results The median number of benzodiazepines of interest detected was 265/year in physical drug samples (range 230–301) and 16/year in clinical samples (range 11–36). Bromazolam was the most common novel benzodiazepine detected in clinical (52/118) and physical (382/1590) samples. Pooled RRs were highest for etizolam [RR = 1.84, 95% confidence interval (95% CI) = 1.39–2.43], bromonordiazepam (RR = 1.70, 95% CI = 0.74–3.92), flualprazolam (RR = 1.48, 95% CI = 0.50–4.39) and bromazolam (RR = 1.22, 95% CI = 0.85–1.73). In 2024, there were no detections of etizolam or flualprazolam in clinical cases, compared with bromazolam ( n = 29; 80.6% of cases) and bromonordiazepam ( n = 7; 19.4% of cases). Among the 48 hospital cases with bromazolam detected, severe outcomes were frequent, including coma (36/48), respiratory failure (34/48) and endotracheal intubation (25/48). Polysubstance use was common (47/48). Bromazolam blood concentrations were poorly predictive of clinical effects. Opioid co‐detection was not associated with statistically significantly worse outcomes. Summary The incidence of novel benzodiazepine detections in New South Wales, Australia, increased between January 2019 and December 2024. Bromazolam is a current concern due to increasing detections and severe clinical outcomes; however, polysubstance use likely contributed to clinical outcomes. Bromonordiazepam may be an emerging concern.