Paramedic Diagnosis of Stroke
Examining Long-Term Use of the Melbourne Ambulance Stroke Screen (MASS) in the Field
Background and Purpose— Recent evidence suggests the Cincinnati Prehospital Stroke Scale is ineffectively used and lacks sensitivity and specificity. Melbourne (Australia) paramedics have been using the Melbourne Ambulance Stroke Screen (MASS) since 2005. The aim of this study was to review the real-world use of MASS 3 years after citywide implementation.
Methods— Two groups of consecutively admitted patients to an Australian hospital between January and May 2008 were used: (1) patients for whom paramedics performed MASS; and (2) patients with a discharge diagnosis of stroke or transient ischemic attack. Use of MASS was examined for all transports and for patients diagnosed with stroke or transient ischemic attack. The sensitivity and specificity of paramedic diagnosis, MASS, and Cincinnati Prehospital Stroke Scale were calculated. Paramedic diagnosis of stroke among patients with stroke was statistically compared with those obtained immediately post-MASS implementation in 2002.
Results— For the study period, MASS was performed for 850 (16%) of 5286 emergency transports, including 199 of 207 (96%) patients with confirmed stroke and transient ischemic attack. In patients in whom MASS was performed (n=850), the sensitivity of paramedic diagnosis of stroke (93%, 95% CI: 88% to 96%) was higher than the MASS (83%, 95% CI: 77% to 88%, P=0.003) and equivalent to Cincinnati Prehospital Stroke Scale (88%, 95% CI: 83% to 92%, P=0.120), whereas the specificity of the paramedic diagnosis of stroke (87%, 95% CI: 84% to 89%) was equivalent to MASS (86%, 95% CI: 83% to 88%, P=0.687) and higher than Cincinnati Prehospital Stroke Scale (79%, 95% CI: 75% to 82%, P<0.001). The initial improvement in stroke paramedic diagnosis seen in 2002 (94%, 95% CI: 86% to 98%) was sustained in 2008 (89%, 95% CI: 84% to 94%, P=0.19).
Conclusion— In our experience, paramedics have successfully incorporated MASS into the assessment of neurologically compromised patients. The initial improvement to the paramedics’ diagnosis of stroke with MASS was sustained 3 years after citywide implementation.
Emergency medical services (EMS) are an integral part of the acute stroke team.1 They are fundamental in maximizing the delivery of thrombolytic therapy to patients with stroke by correctly identifying stroke in the field, transporting patients with suspected stroke to acute stroke centers, and activation of Code Stroke Teams through prehospital notification.2,3
A variety of prehospital stroke screens have been developed to assist EMS to identify patients with stroke in the field (Table 1).4–7 Our previous work confirmed the value in using these screens, showing an immediate improvement in EMS diagnosis of stroke from a baseline of 78% to 94% after education and with use of the Melbourne Ambulance Stroke Screen (MASS).8 However, recent investigations of a similar screen, the Cincinnati Prehospital Stroke Scale (CPSS), suggest poor use by paramedics and low sensitivity and specificity.9,10 Additionally, no long-term evaluations of the use of prehospital screens have been conducted.
The aim of this study was to examine the use of MASS in the field 3 years after citywide education and implementation, specifically determining the use of MASS by paramedics and to calculate and compare the sensitivity and specificity of MASS, CPSS, and paramedic diagnosis of stroke to our previous findings.
Subjects and Methods
This study was a cross-sectional design of consecutive patients transported by EMS to an Australian hospital between January and May 2008. Methods are summarized in the Figure. Institutional ethical approval was received before the start of data collection.
Emergency Medical System
The Victorian EMS is described in detail elsewhere.8 In brief, Ambulance Victoria services 5 million people in the state of Victoria, Australia. Ambulance Victoria uses the Advanced Medical Priority Dispatch System with a 2-tiered response: Advanced Life Support paramedics and Mobile Intensive Care paramedics and takes approximately 387 000 emergency calls per year. The training of Advanced Life Support paramedics required 3 years of university study and a 1-year graduate program. Mobile Intensive Care paramedics have additional education and operate at an independent practitioner level.
All paramedics received a 1-hour stroke education program and instruction on the use of MASS in 2004 to 2005. Paramedics perform the 3 MASS physical assessments (facial droop, hand grip, and speech) in conscious but neurologically compromised patients of no obvious cause such as drug overdose or trauma. If these assessments are positive for stroke, they obtain the remaining MASS history items and perform a blood sugar level to rule out stroke mimics and suitability for thrombolysis. If the MASS is still positive and the stroke is acute, paramedics transport the patient with suspected stroke to the nearest acute stroke center and activate the hospitals “Code Stroke Team” by calling the emergency department en route.
Box Hill Hospital (BHH), located in the eastern suburbs of Melbourne, admits approximately 500 patients with stroke per year. All patients with a confirmed diagnosis of stroke by neuroimaging or with a discharge diagnosis of transient ischemic attack (TIA) admitted to the hospital are entered into the Stroke/TIA registry.
Two groups of patients admitted to BHH were used in this study: (1) patients transported by EMS with documented MASS assessments of hand grip, speech, and facial weakness; and (2) patients with a discharge diagnosis of stroke or TIA included in the Stroke/TIA registry. Patients who were unconscious or asymptomatic at the time of paramedic assessment were excluded (n=49).
Data were accessed through the Victorian Ambulance Clinical Information System and BHH Stroke/TIA registry. The Victorian Ambulance Clinical Information System allows paramedics to electronically record patient information and to access protocols and prompts for additional assessments that may be completed at the scene. The patient’s name, MASS assessments, and paramedic diagnosis were retrieved. This data were crossreferenced against the BHH Stroke/TIA registry (name, date, gender, and age) to determine if the discharge diagnosis was stroke or TIA. For patients with stroke and TIA with no MASS documentation (n=8), MASS and CPSS were retrospectively applied based on the paramedic assessment.
The sensitivity, specificity, positive and negative predictive values, and 95% CIs were calculated for MASS (positive or negative), CPSS (positive or negative), and paramedic diagnosis of stroke/TIA (yes or no) using the discharge diagnosis of stroke/TIA (yes or no). These were statistically compared in SPSS (Version 17.0) using the χ2 test. A probability value <0.05 was considered statistically significant.
Use of MASS
Of the 5286 emergency transports to BHH, 1004 (18%) were conscious but neurologically compromised with no immediately obvious cause. MASS was documented for 850: 16% of all transports and 85% of conscious neurologically compromised patients.
For the same period, 199 (96%) of 207 confirmed stroke or TIA admissions transported by EMS had MASS documentation. Patients with no MASS documentation (n=8) were posterior ischemic strokes (n=4), parietal intracerebral hemorrhages (n=3), and a TIA (n=1; Table 2); 4 of these patients had documentation of confusion and 2 were non-English-speaking.
Sensitivity and Specificity Analysis for Patients With Documented MASS
For patients with documented MASS, the sensitivity of the paramedic diagnosis of stroke was higher than MASS (93% versus 83%, P=0.003) and equivalent to CPSS (93% versus 88%, P=0.120; Table 3). In contrast, the specificity of the paramedic diagnosis of stroke was equivalent to MASS (87% versus 85%, P=0.687) and higher than CPSS (87% versus 79%, P<0.001). The MASS demonstrated equivalent sensitivity to the CPSS (P=0.149) and higher specificity (P=0.001).
Paramedic Diagnosis in Patients With Stroke and TIA
The improvement seen in stroke paramedic diagnosis after the introduction of MASS in a pilot group in 2002 (n=78 of 83 [94%], 95% CI: 86% to 98%) was sustained in 2008 (n=184 of 207 [89%], 95% CI: 84% to 94%, P=0.19). Of the 22 cases in which EMS did not diagnose stroke, 63% (n=14) did not meet MASS criteria for stroke, 14% (n=3) were MASS-positive, and 23% (n=5) did not have MASS documented.
Our large study demonstrated high use of MASS in the field by paramedics and confirms the value of such screens in the identification of stroke. The excellent initial improvement in the diagnosis of stroke by paramedics after the pilot study of MASS8 was sustained 3 years after citywide education and MASS implementation.
The strengths of this study are that it examined the “real-world” use of prehospital stroke screens as used by paramedics and is 1 of a few studies to examine these screening tools outside of screening tool validation studies. However, the findings of this study must be considered in light of its weaknesses. First, the final diagnosis of TIA relied on a discharge diagnosis of TIA, and not all patients with true TIA may have been correctly diagnosed at discharge. Second, positive and negative predictive values are influenced by the prevalence of disease. Because only 23% of all patients with documented MASS were diagnosed as having stroke or TIA, these figures may not reflect true values. Lastly, it is possible that MASS was only performed in patients paramedics strongly suspected had experienced a stroke. This may have artificially inflated the specificity of the screen tools by excluding false-positives and explain the better performance of CPSS compared with previous reports.9,10
Another finding that conflicts with recently published studies in this area9,10 is the improvement in paramedics’ diagnosis of stroke. This disparity may be explained by differences in the samples studied. The previous studies reporting lower paramedic identification of stroke only examined patients with a paramedic diagnosis of stroke or a positive stroke screen, whereas we extended this to include all patients with documented MASS (positive and negative stroke screen). An additional explanation could be differences in paramedic training between Australia and the United States, which may also explain differences found in the use of stroke screens.
Recent work by Frendl et al reported poor use (37.5%) of the CPSS by paramedics in patients with stroke with no improvement after paramedic education.10 However, use of the MASS in our study was high (85% of conscious neurologically impaired patients transported and 96% of patients with stroke). A review of the patients with MASS showed the majority presented with neurological problems and unexplained falls. As discussed in a previous report,11 we attribute some of our success to our feedback system. For the first 18 months after education, we provided the transporting paramedics with the outcome of patients receiving thrombolytic therapy. This allowed us the opportunity to provide a reminder about MASS and to give feedback to paramedics about their contribution to successful patient outcomes.
Previous work has demonstrated that paramedic diagnosis of stroke results in faster in-hospital times12 and improving paramedic diagnosis of stroke with the use of paramedic stroke screening tools has been linked to increasing thrombolytic therapy rates to as high as 21%.13 In our hospital, thrombolytic therapy rates improved from 5%8 to 11%14 after the combined implementation of MASS and an in-hospital code stroke system.
In summary, our large study indicates that paramedics have successfully incorporated MASS into their assessment of neurologically compromised patients and that correct paramedic diagnosis of stroke remained consistently high 3 years after citywide implementation of MASS. Paramedic diagnosis of stroke was higher than the MASS, indicative of a successful stroke education program.
We acknowledge Melissa Wright for her assistance with data collection and entry.
J.E.B. received a National Heart Foundation Research Scholarship.
- Received December 3, 2009.
- Revision received February 8, 2010.
- Accepted February 27, 2010.
Acker JE III, Pancioli AM, Crocco TJ, Eckstein MK, Jauch EC, Larrabee H, Meltzer NM, Mergendahl WC, Munn JW, Prentiss SM, Sand C, Saver JL, Eigel B, Gilpin BR, Schoeberl M, Solis P, Bailey JR, Horton KB, Stranne SK. Implementation strategies for emergency medical services within stroke systems of care: a policy statement from the American Heart Association/American Stroke Association Expert Panel on Emergency Medical Services Systems and the Stroke Council. Stroke. 2007; 38: 3097–3115.
Belvis R, Cocho D, Marti-Fabregas J, Pagonabarraga J, Aleu A, Garcia-Bargo MD, Pons J, Coma E, Garcia-Alfranca F, Jimenez-Fabrega X, Marti-Vilalta JL. Benefits of a prehospital stroke code system. Feasibility and efficacy in the first year of clinical practice in Barcelona, Spain. Cerebrovasc Dis. 2005; 19: 96–101.
Abdullah AR, Smith EE, Biddinger PD, Kalenderian D, Schwamm LH. Advance hospital notification by EMS in acute stroke is associated with shorter door-to-computed tomography time and increased likelihood of administration of tissue-plasminogen activator. Prehosp Emerg Care. 2008; 12: 426–431.
Kidwell CS, Starkman S, Eckstein M, Weems K, Saver JL. Identifying stroke in the field. Prospective validation of the Los Angeles Prehospital Stroke Screen (LAPSS). Stroke. 2000; 31: 71–76.
Frendl DM, Strauss DG, Underhill BK, Goldstein LB. Lack of impact of paramedic training and use of the Cincinnati Prehospital Stroke Scale on stroke patient identification and on-scene time. Stroke. 2009; 40: 754–756.
Bray JE, Bladin C. Success with paramedic diagnosis of stroke. Stroke. 2009; 40: e398.
Mosley I, Nicol M, Donnan G, Patrick I, Kerr F, Dewey H. The impact of ambulance practice on acute stroke care. Stroke. 2007; 38: 2765–2770.