文献速递多重用药老年慢病共病管理衰弱、肌少症与跌倒

长期护理机构入住后慢性阿片类药物使用风险预测模型的开发

Development of a Model to Predict the Risk of Chronic Opioid Use After Entry Into Long-Term Care Facilities.
PubMed 文献 · J Am Med Dir Assoc · 2025-12-15

本研究基于澳大利亚老年登记处国家历史队列,纳入2009至2019年间入住长期护理机构的65岁及以上非原住民、非安宁疗护且入住前非慢性阿片使用者的355,208人,采用竞争风险回归开发入住1年内新发慢性阿片使用的预测模型。

要点

  • 入住长期护理机构1年内7%新发慢性阿片使用。
  • 预测模型C指数0.75,校准可接受,具有临床效用。
  • 最强预测因子含入住前非慢性阿片使用、高药物辅助评分、疼痛史、复杂医疗需求和高镇静负荷。

核心结论

  • 开发了预测长期护理机构新入住居民慢性阿片使用风险的模型,区分度良好且具有临床效用。

研究设计与样本

  • 回顾性队列分析,使用澳大利亚老年登记处国家历史队列数据。
  • 纳入2009至2019年入住南澳大利亚、维多利亚、新南威尔士或昆士兰长期护理机构的65岁及以上非原住民、非安宁疗护且入住前非慢性阿片使用者,共355,208人。
  • 采用竞争风险回归建模入住1年内慢性阿片使用时间,评估区分度、校准和临床效用。

主要发现

  • 中位年龄85岁,61%为女性,54%患有痴呆。
  • 7%(n=25,171)成为慢性阿片使用者。
  • 模型C指数为0.75(95%CI 0.74-0.76),校准可接受。

关键数值与推荐

  • 模型C指数0.75(95%CI 0.74-0.76)。

临床怎么用

  • 可在入住长期护理机构时识别慢性阿片使用高风险居民,指导针对性疼痛管理和阿片药物管理干预。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

OBJECTIVES: Chronic opioid use is common in residents of long-term care facilities (LTCFs) and is associated with an increased risk of falls, fractures, and cognitive impairment. However, predictors of chronic opioid use in this cohort remain unclear. This study developed a prediction model for new chronic opioid use within 1 year of entry into LTCFs and identifies key predictors. DESIGN: This study was a retrospective cohort analysis using data from the Registry of Senior Australians National Historical Cohort. SETTING AND PARTICIPANTS: The cohort included non-Indigenous individuals 65 years and older who entered an LTCF in South Australia, Victoria, New South Wales, or Queensland between 2009 and 2019 who were not in palliative care and not a chronic opioid user before entry. METHODS: Time to chronic opioid use within 1 year of entry was modeled using competing risk regression. Candidate predictors included individual, health care, medication, facility, and system-related factors. Model discrimination (C-index), calibration, and clinical utility were assessed. RESULTS: Among 355,208 individuals (median age 85 years, 61% women, 54% with dementia), 7% (n = 25,171) became chronic opioid users. The prediction model had a C-index of 0.75 (95% CI, 0.74-0.76) and its calibration was acceptable. Eighteen predictors were identified, the strongest were nonchronic opioid use before entry, high medication assistance rating, history of pain, complex health care needs, and high sedative load. CONCLUSION AND IMPLICATIONS: A good predictive ability and clinical utility model was developed for estimating the risk of chronic opioid use in new LTCF residents. Identifying residents at high risk of chronic opioid use at entry into long-term can guide targeted interventions for pain management and opioid stewardship initiatives.

来源:J Am Med Dir Assoc,查看原文。版权归原作者所有。

引用

  1. Thuraisingam S, Eshetie TC, Caughey GE, et al. Development of a Model to Predict the Risk of Chronic Opioid Use After Entry Into Long-Term Care Facilities. J Am Med Dir Assoc. 2025. PMID: 41314621. doi:10.1016/j.jamda.2025.106015

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。