Implementation of a severity-based triage protocol improved guideline-compliant hospital selection from 62% to 71% and decreased EMS scene arrival-to-CT times by 7 minutes (95% CI: 3–11).
Does an adapted AHA Severity-Based Stroke Triage Algorithm improve destination selection and treatment times in rural suspected stroke patients?
A rural EMS severity-based triage protocol improved guideline-compliant destination selection and reduced time to CT imaging, though it did not significantly expedite mechanical thrombectomy.
Background: Timely and accurate triage of stroke patients in rural settings remains a critical challenge, especially for patients with large vessel occlusions (LVO) requiring rapid access to comprehensive stroke centers (CSCs). This project aimed to improve the efficiency and accuracy of transport decisions for EMS stroke care in rural Michigan by adapting American Heart Association (AHA) guidelines on prehospital management and EMS routing. Method: We established a novel database involving linking regional EMS records, regional hospital stroke code quality improvement data, and Get With The Guidelines–Stroke (GWTG-S) registry data for suspected and confirmed stroke cases originating from five counties within 60 minutes of an urban CSC in West Michigan. Using a stepped-wedge design, we adapted the model AHA Severity-Based Stroke Triage Algorithm and implemented it in four rural EMS agencies through a staged roll-out of educational programs delivered to EMS providers and regional ED staff. We analyzed pre- and post-intervention data to assess protocol compliance, bypass frequency, and time from EMS scene arrival to CT acquisition and reperfusion therapy delivery. Results: Among 949 (598 pre- and 351 post-intervention) stroke transports occurred between November 2021 to December 2024, guideline-compliant hospital destination selection improved from 62% to 71% (p=0.018). Stroke mimics saw the greatest improvement (72% to 88%, p<0.001) through reduced unnecessary bypass. LVO screening documentation increased (10% to 23%, p<0.001), and EMS scene arrival-to-CT time decreased by 7 minutes (95% CI: 3–11). Trends toward improved thrombolysis rates (18% to 26%, p=0.055) and faster EMS-to-needle times (99 to 87 min, p=0.059) were observed. Mechanical thrombectomy rates and transfer metrics remained unchanged. Conclusions: This project successfully implemented a rural stroke triage protocol aligned with AHA guidelines, improving EMS routing decisions and reducing time to imaging. The intervention enhanced access to thrombolysis, however did not expedite mechanical thrombectomy. Findings support the AHA protocol’s scalability and underscore the feasibility of rural stroke system optimization.
Oostema et al. (2026) studied this question. Implementation of a severity-based triage protocol improved guideline-compliant hospital selection from 62% to 71% and decreased EMS scene arrival-to-CT times by 7 minutes (95% CI: 3–11).