Department of Emergency Medicine, Brigham & Women's Hospital, Boston, MA.
Department of Emergency Medicine, Massachusetts General Hospital, Boston, MA.
Ann Emerg Med. 2024 Jan;83(1):24-34. doi: 10.1016/j.annemergmed.2023.08.017. Epub 2023 Sep 19.
The role of venous thromboembolism (VTE) prophylaxis among patients receiving emergency department (ED) observation unit care is unclear. We investigated an electronic health record-based clinical decision support tool aimed at increasing pharmacologic VTE prophylaxis use among at-risk patients placed in ED observation units.
We conducted an interrupted time-series study of an Epic-based best practice advisory implemented in May 2019 at a health care system comprising 2 academic medical centers and 4 community hospitals with dedicated ED observation units. The best practice advisory alerted staff at 24 hours to conduct a risk assessment and linked to a VTE prophylaxis order set. We used an interrupted time series, Bayesian structured time series, and a multivariable mixed-effect regression model to estimate the intervention effect.
Prior to the best practice advisory implementation, there were 8,895 ED observation unit patients with a length of stay more than or equal to 24 hours, and 0.9% received pharmacologic VTE prophylaxis. Afterward, there were 12,664 ED observation unit patients with a length of stay more than or equal to 24 hours, and 4.8% received pharmacologic VTE prophylaxis. The interrupted time series and causal impact analysis showed a statistically significant increase in VTE prophylaxis (eg, absolute percent difference 3.8%, 95% confidence interval 3.5 to 4.1). In a multivariable model, only the intervention was significantly associated with receiving VTE prophylaxis (odds ratio 4.56, 95% confidence interval 2.22 to 9.37).
An electronic health record-based alert helped to prompt staff caring for ED observation unit patients at risk for VTE with prolonged visits to order recommended pharmacologic prophylaxis. The best risk assessment model to use and the true incidence of VTE events in this population are unclear.
接受急诊科（ED）观察单元护理的患者中静脉血栓栓塞症（VTE）预防的作用尚不清楚。我们调查了一种基于电子健康记录的临床决策支持工具，旨在增加处于 ED 观察单元的高危患者中使用药物 VTE 预防的可能性。
我们对 2019 年 5 月在一个医疗保健系统中实施的基于 Epic 的最佳实践咨询进行了一项中断时间序列研究，该系统由 2 个学术医疗中心和 4 个设有专门 ED 观察单元的社区医院组成。最佳实践咨询会在 24 小时内向工作人员发出风险评估警报，并与 VTE 预防医嘱集相关联。我们使用中断时间序列、贝叶斯结构时间序列和多变量混合效应回归模型来估计干预效果。
在最佳实践咨询实施之前，有 8895 名 ED 观察单元患者的住院时间超过或等于 24 小时，其中 0.9%接受了药物 VTE 预防。之后，有 12664 名 ED 观察单元患者的住院时间超过或等于 24 小时，其中 4.8%接受了药物 VTE 预防。中断时间序列和因果影响分析显示，VTE 预防的比例有统计学显著增加（例如，绝对百分比差异 3.8%，95%置信区间 3.5 至 4.1）。在多变量模型中，只有干预措施与接受 VTE 预防显著相关（比值比 4.56，95%置信区间 2.22 至 9.37）。
基于电子健康记录的警报有助于提示护理 ED 观察单元患者的工作人员，对延长就诊时间有 VTE 风险的患者进行推荐的药物预防。使用最佳风险评估模型和该人群中 VTE 事件的真实发生率尚不清楚。
