ObjectiveThe purpose of the study is to assess if daily use of hypnotics increases mortality, aspiration pneumonia and hip fracture among relatively healthy individuals aged 75 years or older who lead independent lives in the community.Method and PatientsOf the adults aged 75 years or older residing in Hokkaido prefecture of Japan (n = 705,538), those who did not meet several exclusion criteria were eligible for generating propensity score‐matched cohorts (n = 214,723). Exclusion criteria included co‐prescribed medications acting on the central nervous system, diagnoses of malignant neoplasm, dementia, depression, etc. We compared 33,095 participants who were prescribed hypnotics for daily use (hypnotic group) with a propensity score‐matched cohort without a prescription (control group). Participants were followed for more than 42 months.ResultsDuring the 42‐month follow‐up period, the incidence of the three outcome measures in the hypnotics group was significantly higher than that in the control group (aspiration pneumonia p < 0.001, hip fracture p = 0.007, and all‐cause mortality p < 0.001). Sensitivity analyses utilizing inverse probability weighting demonstrated hazard ratios of 1.083 [1.023–1.146] for mortality, 1.117 [1.014–1.230] for aspiration pneumonia, and 1.720 [1.559–1.897] for hip fracture. Meanwhile, the attribute risk differences were 2.7, 1.5, and 1.0 per 1000 patient‐years, respectively.ConclusionsAlthough daily use of hypnotics increased the risk of three events, their attribute risk differences were fewer than 3.0 per 1000 patient‐years. The results will help provide guidance on whether it is reasonable to prescribe hypnotics to geriatric population aged 75 or older leading independent lives in the community.Clinical trial registrationUMIN‐CTR UMIN000048398.

中文翻译：

---

安眠药或镇静剂对日本 75 岁或以上社区老年人的安全性概况：一项回顾性倾向匹配队列研究

目的本研究的目的是评估每天使用安眠药是否会增加在社区独立生活的 75 岁或以上相对健康的个体的死亡率、吸入性肺炎和髋部骨折。方法和患者居住在 75 岁或以上的成年人日本北海道县（n= 705,538），那些不符合多项排除标准的人有资格生成倾向评分匹配队列（n= 214,723）。排除标准包括作用于中枢神经系统的共同处方药物、恶性肿瘤、痴呆、抑郁症等的诊断。我们将 33,095 名每日服用催眠药的参与者（催眠组）与没有处方的倾向评分匹配队列进行了比较（控制组）。对参与者进行了超过42个月的随访。结果在42个月的随访期间，安眠药组的三项结局指标的发生率均显着高于对照组（吸入性肺炎）p< 0.001，髋部骨折p= 0.007，全因死亡率p< 0.001）。利用逆概率加权的敏感性分析表明，死亡率的风险比为 1.083 [1.023–1.146]，吸入性肺炎的风险比为 1.117 [1.014–1.230]，髋部骨折的风险比为 1.720 [1.559–1.897]。同时，属性风险差异分别为每 1000 患者年 2.7、1.5 和 1.0。 结论 虽然每天使用催眠药会增加三种事件的风险，但它们的属性风险差异小于每 1000 患者年 3.0。研究结果将有助于为75岁或以上在社区独立生活的老年人群开安眠药是否合理提供指导。临床试验注册UMIN-CTR UMIN000048398。