Scott Halden F, Sevick Carter J, Colborn Kathryn L, Deakyne Davies Sara J, Greer Christopher H, Dafoe Ashley, Dorsey Holliman Brooke, Bajaj Lalit, Schmidt Sarah K, Kempe Allison
Section of Pediatric Emergency Medicine, Department of Pediatrics, University of Colorado School of Medicine, Aurora, Colorado.
Emergency Department, Children's Hospital Colorado, Aurora, Colorado.
Pediatrics. 2025 Jul 1;156(1). doi: 10.1542/peds.2024-069478.
Delays in septic shock diagnosis cause preventable mortality in children. Evidence is limited around early recognition strategies. The hypothesis was that clinical decision support (CDS) based on machine-learning predictive models would increase the proportion of children receiving septic shock treatment prior to shock onset.
CDS was implemented in a prospective, stepped-wedge, cluster randomized trial in 4 pediatric emergency departments (EDs) over five 10-week periods. The CDS used models identifying children who did not yet have shock but were predicted to be at high risk based on electronic health record data at arrival and after 2 hours. Providers received CDS; effectiveness was evaluated in patients 60 days to 18 years with concern for sepsis. The primary outcome was antibiotic and bolus within 1 hour of sepsis suspicion. Secondary outcomes were time-to-antibiotic, hypotensive septic shock. Implementation outcomes were evaluated in qualitative interviews.
Of 200 354 ED encounters from March 16, 2022, to March 1, 2023, 1331 encounters met inclusion criteria (979 intervention, 352 control arms). Antibiotic and bolus within 1 hour occurred in 39.0% of patients in the intervention arm versus 38.9% of patients in the control arm (adjusted odds ratio [aOR]: 1.07 [0.61-1.88]). There was no difference in outcomes of shock (aOR: 1.12 [0.53-2.46]) or antibiotic timeliness (aHR: 0.85 [0.63-1.16]). Providers reported the CDS felt valuable and unobtrusive (adoption); 6 months after the trial, EDs continued to use the CDS (maintenance).
Implementing predictive CDS that infrequently alerted was feasible and acceptable. It did not change the proportion of patients with suspected sepsis who progressed to hypotensive shock.
脓毒性休克诊断延迟会导致儿童出现可预防的死亡。关于早期识别策略的证据有限。研究假设是基于机器学习预测模型的临床决策支持（CDS）将提高在休克发作前接受脓毒性休克治疗的儿童比例。
在4个儿科急诊科进行了一项前瞻性、阶梯式、整群随机试验，为期5个10周周期，实施了CDS。CDS使用模型识别尚未发生休克但根据入院时和2小时后的电子健康记录数据预测为高危的儿童。医护人员接受了CDS；对60天至18岁疑似脓毒症的患者进行了有效性评估。主要结局是在怀疑脓毒症后1小时内使用抗生素和推注药物。次要结局是抗生素使用时间、低血压性脓毒性休克。通过定性访谈评估实施效果。
在2022年3月16日至2023年3月1日期间的200354次急诊科就诊中，1331次就诊符合纳入标准（979次干预组，352次对照组）。干预组中在1小时内使用抗生素和推注药物的患者比例为39.0%，对照组为38.9%（调整优势比[aOR]：1.07[0.61 - 1.88]）。休克结局（aOR：1.12[0.53 - 2.46]）或抗生素及时性（调整风险比[aHR]：0.85[0.63 - 1.16]）没有差异。医护人员报告称CDS很有价值且不引人注意（采用情况）；试验6个月后，急诊科仍在使用CDS（维持情况）。
实施不常发出警报的预测性CDS是可行且可接受的。它并未改变疑似脓毒症进展为低血压性休克的患者比例。
