Predictive Ability of an Emergency Medical Dispatch Stroke Diagnostic Tool in Identifying Hospital-Confirmed Strokes

• Published in: Journal of stroke and cerebrovascular diseases 2016-08, Vol.25 (8), p.2031-2042
• Main Authors: Clawson, Jeff J., MD, Scott, Greg, MBA, Gardett, Isabel, PhD, Youngquist, Scott, MD, Taillac, Peter, MD, Fivaz, Conrad, MD, Olola, Christopher, PhD
• Format: Article
• Language: English
• Subjects: Cardiovascular; cerebrovascular accident; cerebrovascular disorders; Emergency Medical Dispatch - methods; Emergency Medical Service Communication Systems; Female; Hospitals; Humans; Male; Neurology; Predictive Value of Tests; Retrospective Studies; Sensitivity and Specificity; Severity of Illness Index; Stroke; Stroke - diagnosis; Stroke - therapy; stroke diagnostic and identification; transient ischemic attack; Treatment Outcome
• ISSN: 1052-3057; 1532-8511
• DOI: 10.1016/j.jstrokecerebrovasdis.2016.04.021

Background Early hospital notification of a possible stroke arriving via emergency medical services (EMS) can prepare stroke center personnel for timely treatment, especially timely administration of tissue plasminogen activator. Stroke center notification from the emergency dispatch center—before responders reach the scene—may promote even earlier and faster system activation, meaning that stroke center teams may be ready to receive patients as soon as the ambulance arrives. This study evaluates the use of a Medical Priority Dispatch System (MPDS; Priority Dispatch Corp., Salt Lake City, UT) Stroke Diagnostic Tool (SDxT) to identify possible strokes early by comparing the tools' results to on-scene and hospital findings. Methods The retrospective descriptive study utilized stroke data from 3 sources: emergency medical dispatch, EMS, and emergency department/hospital. Results A total of 830 cases were collected between June 2012 and December 2013, of which 603 (72.7%) had matching dispatch records. Of the 603 cases, 304 (50.4%) were handled using MPDS Stroke Protocol 28. The SDxT had an 86.4% ability (OR [95% CI]: 2.3 [1.5, 3.5]) to effectively identify strokes among all the hospital-confirmed stroke cases (sensitivity), and a 26.6% ability to effectively identify nonstrokes among all the hospital-confirmed nonstroke cases (specificity). Conclusions The SDxT demonstrated a very high sensitivity, compared to similar tools used in the field and at dispatch. The specificity was somewhat low, but this was expected—and is intended in the creation of protocols to be used over the phone in emergency situations. The tool is a valuable method for identifying strokes early and may allow early hospital notification.