Background <p>Fracture risk associated with sleep drugs is of great concern in patients with chronic kidney disease (CKD). However, the fracture risk among sleep drugs with different mechanisms of action remains controversial. This study aimed to examine how fracture risk differs between the use of benzodiazepine receptor agonists (BDZRA) and dual orexin receptor antagonists (DORA) in patients with CKD.</p> Methods <p>A large-scale Japanese medical claims database of patient data collected between April 2008 and August 2021 was analysed. The association between composite hip and vertebral body fracture risk and continuous prescription of DORA or BDZRA was examined using a Cox proportional hazards model. Propensity score matching was employed to reduce confounding.</p> Results <p>DORA and BDZRA were prescribed to 4,504 and 22,080 patients, respectively. The BDZRA group showed a lower cumulative incidence rate of composite hip and vertebral body fractures than the DORA group (hazard ratio [HR]: 0.67, <i>p</i> = 0.06 in the unmatched cohort and HR: 0.64, <i>p</i> = 0.129 in the matched cohort). The risk of hip fracture alone was comparable for both the unmatched cohort (HR: 0.83, <i>p</i> = 0.507) and matched cohort (HR: 0.94, <i>p</i> = 0.869). The risk of vertebral body fracture alone was significantly lower in both the unmatched and matched cohorts (HR: 0.45, <i>p</i> = 0.021 for the unmatched cohort and HR: 0.23, <i>p</i> = 0.023 for the matched cohort).</p> Conclusions <p>The results suggest a possible association between DORA prescription and bone metabolism in patients with CKD. DORA should be used with caution in these patients.</p> Clinical trial number <p>Not applicable.</p>