急性卒中医保受益人插管频率及其与患者结局的关联
该回顾性观察研究分析了2011至2018年65至100岁按服务付费的医保受益人急性卒中住院数据,共800467例住院、3581家医院。医院调整后插管率中位数为4.8%,插管患者30天死亡风险较未插管者高44%,最高插管率四分位医院调整后30天死亡率19.3%高于最低四分位医院16.7%。
要点
- 急性卒中住院插管率在不同医院间差异大,中位数4.8%,四分位间距2.4%至12.5%以上。
- 缺血性卒中调整后插管风险比脑出血低21.1%,血栓切除术与调整后插管风险升高19.2%相关。
- 插管患者30天死亡风险比未插管患者高44%,最高插管率四分位医院30天死亡率19.3%高于最低四分位16.7%。
核心结论
- 急性卒中插管率因患者和医院特征而异,调整后插管率较高的医院患者30天死亡率较高,但大部分差异可能源于未测量的患者严重程度,因为医疗市场层面未观察到类似关联。
研究设计与样本
- 回顾性观察研究,纳入2011至2018年65至100岁按服务付费的医保受益人急性卒中住院,共800467例住院、3581家医院。
主要发现
- 在2588家卒中住院≥25例的医院中,插管率中位数为4.8%,四分之一医院低于2.4%,10%医院高于12.5%。
- 缺血性卒中调整后插管风险比脑出血低21.1%(风险差-21.1%,95%CI -21.3%至-20.9%;p<0.001),血栓切除术与调整后插管风险升高19.2%相关(95%CI 18.8%至19.6%;p<0.001)。
- 女性和高龄患者插管率较低;大型城市医院和学术医学中心在调整后插管率最高四分位中占比过高。
- 调整可用特征后,插管患者死亡风险比未插管患者高44%(p<0.001)。
- 最高插管率四分位医院调整后30天死亡率为19.3%,高于最低四分位医院的16.7%;限制在主要教学医院时分别为22.3%和18.1%;医疗市场插管率四分位与患者30天死亡率无关联。
局限性
- 原文未报告局限性。
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
BACKGROUND AND OBJECTIVES: Intubation for acute stroke is common in the United States, with few established guidelines. METHODS: This is a retrospective observational study of acute stroke admissions from 2011 to 2018 among fee-for-service Medicare beneficiaries aged 65-100 years. Patient demographics and chronic conditions as well as hospital characteristics were identified. We identified patient intubation, stroke subtype (ischemic vs intracerebral hemorrhage), and thrombectomy. Factors associated with intubation were identified by a linear probability model with intubation as the outcome and patient characteristics, stroke subtype, and thrombectomy as predictors, adjusting for within-hospital correlation. We compared hospital characteristics between adjusted intubation rate quartiles. We specified a linear probability model with 30-day mortality as the patient-level outcome and hospital intubation rate quartile as the categorical predictor, again adjusting for patient characteristics. We specified an analogous model for quartiles of hospital referral regions. RESULTS: There were 800,467 stroke hospitalizations at 3,581 hospitals. Among 2,588 hospitals with 25 or more stroke hospitalizations, the median intubation rate was 4.8%, while a quarter had intubation rates below 2.4% and 10% had rates above 12.5%. Ischemic strokes had a 21% lower adjusted intubation risk than intracerebral hemorrhages (risk difference [RD] -21.1%, 95% CI -21.3% to -20.9%; p < 0.001), whereas thrombectomy was associated with a 19.2% higher adjusted risk (95% CI RD 18.8%-19.6%; p < 0.001). Women and older patients had lower intubation rates. Large, urban hospitals and academic medical centers were overrepresented in the top quartile of hospital adjusted intubation rates. Even after adjusting for available characteristics, intubated patients had a 44% higher mortality risk than non-intubated patients (p < 0.001). Hospitals in the highest intubation quartile had higher adjusted 30-day mortality (19.3%) than hospitals in the lowest quartile (16.7%), a finding that was similar when restricting to major teaching hospitals (22.3% vs 18.1% in the 4th vs 1st quartiles, respectively). There was no association between market quartile of intubation and patient 30-day mortality. DISCUSSION: Intubation for acute stroke varied by patient and hospital characteristics. Hospitals with higher adjusted rates of intubation had higher patient-level 30-day mortality, but much of the difference may be due to unmeasured patient severity given that no such association was observed for health care markets.
引用
- Burke LG, Fehnel CR, Burke RC, et al. Frequency of Intubation for Medicare Beneficiaries With Acute Stroke and Association With Patient Outcomes. Neurology. 2024. PMID: 38295353. doi:10.1212/WNL.0000000000208031
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。