Ramos John, Theophanous Rebecca, Gettler Erin, Moehring Rebekah, Wrenn Rebekah, Shaheen Stephen, Krcmar Rachel, Seidelman Jessica Leigh
Duke University Medical Center, 2301 Erwin Road, Durham, NC 27710, United States.
Dartmouth Hitchcock Medical Center, 1 Medical Center Drive, Lebanon, NH 03766, United States.
Am J Emerg Med. 2025 Jul;93:135-139. doi: 10.1016/j.ajem.2025.04.006. Epub 2025 Apr 3.
Blood culture (BCx) diagnostic stewardship is essential for reducing unnecessary treatments, minimizing false-positive results, and improving patient outcomes and hospital resource utilization. The objective of this study was to compare the effectiveness of diagnostic stewardship interventions on BCx utilization in three emergency departments (ED).
We used a quasi-experimental pre-/post-intervention study to compare BCx rates (BCx/100 ED visits) between December 1, 2020, and February 29, 2024 before and after the implementation of a BCx algorithm and electronic health record (EHR) modifications at one large academic ED and level 1 trauma center, and two EDs at academic-affiliated community hospitals. A sample of visits with a BCx order were audited in one academic ED and one academic-affiliated ED, and summary data on indication appropriateness were provided to respective leadership. In the academic ED, there was weekly provider led audit and feedback on 3478 ED visits. In one academic-affiliated ED, one pharmacist reviewed five visits weekly (100 total) for appropriateness. The second academic-affiliated ED served as a control and did not receive any feedback on BCx utilization. Each ED's BCx rates were analyzed using interrupted time series models. Incidence rate ratios (IRR) compared BCx rates before and after the interventions.
A total of 211,950 BCxs over 572,776 ED visits were included in the analysis. The academic ED saw a 25 % decrease in BCx rate with IRR 0.80 (95 % CI 0.74, 0.86, p-value 0.01). The first academic-affiliated ED experienced a 0.8 % decrease in BCx rate with IRR 1.1 (95 % CI 1.01, 1.19, p-value 0.02). No change was observed in the second academic-affiliated ED.
Decreased BCx rates occurred only after direct audit and feedback and EHR modifications. Both the academic ED and first academic-affiliated ED saw a drift back towards pre-intervention BCx rates after the intervention.
血培养（BCx）诊断管理对于减少不必要的治疗、将假阳性结果降至最低以及改善患者预后和医院资源利用至关重要。本研究的目的是比较诊断管理干预措施对三个急诊科（ED）血培养利用情况的有效性。
我们采用了一项干预前后的准实验研究，比较了在一家大型学术急诊科和一级创伤中心以及两家学术附属社区医院急诊科实施血培养算法和电子健康记录（EHR）修改前后，2020年12月1日至2024年2月29日期间的血培养率（每100次急诊就诊的血培养次数）。在一家学术急诊科和一家学术附属急诊科对有血培养医嘱的就诊样本进行了审核，并将关于指征适宜性的汇总数据提供给各自的领导。在学术急诊科，每周由提供者对3478次急诊就诊进行审核和反馈。在一家学术附属急诊科，一名药剂师每周审核5次就诊（共100次）的适宜性。第二家学术附属急诊科作为对照，未收到关于血培养利用情况的任何反馈。使用中断时间序列模型分析每个急诊科的血培养率。发病率比值（IRR）比较了干预前后的血培养率。
分析纳入了572,776次急诊就诊中的211,950次血培养。学术急诊科的血培养率下降了25%，IRR为0.80（95%CI 0.74, 0.86，p值0.01）。第一家学术附属急诊科的血培养率下降了0.8%，IRR为1.1（95%CI 1.01, 1.19，p值0.02）。第二家学术附属急诊科未观察到变化。
仅在直接审核和反馈以及电子健康记录修改后血培养率才下降。干预后，学术急诊科和第一家学术附属急诊科的血培养率都出现了向干预前水平回落的趋势。
