Clinical Predictors of Accurate Prehospital Stroke

Recognition
J. Adam Oostema, MD; John Konen, BS; Todd Chassee, MD; Mojdeh Nasiri, MD;
Mathew J. Reeves, PhD
Background and PurposePrehospital activation of in-hospital stroke response hastens treatment but depends on accurate
emergency medical services (EMS) stroke recognition. We sought to measure EMS stroke recognition accuracy and
identify clinical factors associated with correct stroke identification.
MethodsUsing EMS and hospital records, we assembled a cohort of EMS-transported suspect, confirmed, or missed
ischemic stroke or transient ischemic attack cases. The sensitivity and positive predictive value (PPV) for EMS stroke
recognition were calculated using the hospital discharge diagnosis as the gold standard. We used multivariable logistic
regression analysis to determine the association between Cincinnati Prehospital Stroke Scale use and EMS stroke
recognition.
ResultsDuring a 12-month period, 441 EMS-transported patients were enrolled; of which, 371 (84.1%) were EMSsuspected strokes and 70 (15.9%) were EMS-missed strokes. Overall, 264 cases (59.9%) were confirmed as either
ischemic stroke (n=186) or transient ischemic attack (n=78). The sensitivity of EMS stroke recognition was 73.5% (95%
confidence interval, 67.778.7), and PPV was 52.3% (95% confidence interval, 47.157.5). Sensitivity (84.7% versus
30.9%; P<0.0001) and PPV (56.2% versus 30.4%; P=0.0004) were higher among cases with Cincinnati Prehospital Stroke
Scale documentation. In multivariate analysis, Cincinnati Prehospital Stroke Scale documentation was independently
associated with EMS sensitivity (odds ratio, 12.0; 95% confidence interval, 5.725.5) and PPV (odds ratio, 2.5; 95%
confidence interval, 1.34.7).
ConclusionsEMS providers recognized 3 quarters of the patients with ischemic stroke and transient ischemic attack;
however, half of EMS-suspected strokes were false positives. Documentation of a Cincinnati Prehospital Stroke
Scale was associated with higher EMS stroke recognition sensitivity and PPV.(Stroke. 2015;46:1513-1517.
DOI: 10.1161/STROKEAHA.115.008650.)
Key Words: emergency medical services stroke, acute

mong patients with acute ischemic stroke (IS), transport

by emergency medical services (EMS) has been associated with earlier arrival,1 faster emergency department (ED)
evaluations,24 and improved rates and speed of tissue-type
plasminogen activator (tPA) delivery.3 These benefits stem, at
least in part, from prearrival activation of stroke teams as a
result of hospital prenotification by EMS.5,6 Stroke recognition by EMS providers in the field is therefore a critical step
in the stroke chain of recovery. However, accurate stroke
identification in the field is challenging because of variable
and often nonspecific clinical presentations of patients with
stroke and transient ischemic attack (TIA), as well as the high
prevalence of stroke mimics.79 In response to this, many prehospital stroke scales, such as the Los Angeles Prehospital

Stroke Screen,10 the Melbourne Ambulance Stroke Screen,11

the Ontario Prehospital Stroke Screening Tool,12 and the
Cincinnati Prehospital Stroke Scale (CPSS),13 have been
developed to improve the accuracy of prehospital stroke
recognition. Despite endorsement by national guideline recommendations,14 validation studies have reported variable
accuracy of these toolsparticularly with respect to false positives resulting in low specificity.15 Furthermore, the degree to
which these scales are incorporated into current EMS practice
is not well documented.
We recently established a cohort study to identify and link
EMS and hospital records for patients transported by EMS
with suspected, confirmed, or missed IS or TIA to determine
the accuracy of prehospital stroke recognition. We sought to

Received January 13, 2015; final revision received March 6, 2015; accepted March 27, 2015.
From the Department of Emergency Medicine, Spectrum Health, Grand Rapids, MI (J.A.O.), Department of Emergency Medicine, Michigan State
University College of Human Medicine, Grand Rapids (J.A.O., J.K., T.C.); Kent County Emergency Medical Services, MI (T.C.); and Department of
Epidemiology, Michigan State University, East Lansing (M.N., M.J.R.).
Presented in part at the International Stroke Conference, Nashville, TN, February 1113, 2015.
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.
115.008650/-/DC1.
Reprint requests to J. Adam Oostema, MD, Department of Emergency Medicine, Michigan State University College of Human Medicine, Secchia
Center, Room 425, 15 Michigan, NE, Grand Rapids, MI 49503. E-mail oostema@msu.edu
2015 American Heart Association, Inc.
Stroke is available at http://stroke.ahajournals.org

DOI: 10.1161/STROKEAHA.115.008650

Downloaded from http://stroke.ahajournals.org/

by guest on May 26, 2015
1513

1514StrokeJune 2015
measure the prevalence of CPSS in this cohort, analyze the
relationship between CPSS use and EMS diagnostic accuracy,
and describe errors in prehospital stroke recognition.

Methods

194

177

EMS
Suspected

70

78 (6386)

<60

87 (19.5)

6069

68 (15.2)

7079

85 (19.0)

8089

125 (28.0)

Figure 1. Paramedic stroke recognition among 441 emergency

medical services (EMS)transported suspect or missed ischemic
stroke or transient ischemic attack (TIA). PPV indicates positive
predictive value.

sensitivity and PPV were compared using 2 tests. To determine the

independent association between CPSS use and the sensitivity of EMS
stroke recognition, we used multivariable logistic regression to calculate the adjusted odds ratio (OR) for accurate prehospital stroke recognition given CPSS documentation among confirmed stroke or TIA
cases. We adjusted for potential confounders, including age, National
Institute of Health Stroke Scale (NIHSS), sex, dispatch reason (stroke
versus others), and time from symptom onset. We then applied the same
model among EMS-suspected stroke cases to measure the relationship
between CPSS documentation and the PPV of EMS stroke suspicion.
Finally, we examined errors in EMS stroke recognition. To evaluate
EMS-missed IS cases, we abstracted clinical characteristics and patient
history obtained from the initial ED evaluation of cases that were subsequently confirmed as having IS. The documented NIHSS was used to
assess for the presence of specific neurological deficits because physical examination descriptions were highly variable in the ED notes.
Because patients with TIA are not candidates for acute intervention
and >55% of confirmed TIA cases in our data set lacked a documented
NIHSS or had an NIHSS of 0 by the time of presentation in the ED, we
excluded patients with TIA from this part of the analysis. We compared
the clinical characteristics of EMS-recognized cases with EMS-missed
cases using 2 tests for categorical variables, MannWhitney U tests for
ordinal variables, and Student t-tests for continuous variables. We then
compared the prevalence of stroke symptoms and signs among EMSrecognized and missed strokes. We also identified the most common
EMS transport impressions among missed stroke and TIA cases and
the most common final discharge diagnoses among the patients without
stroke transported by EMS as suspected stroke.

Results
During a 1-year period, 371 cases were transported by EMS as
suspected stroke or TIA, whereas another 70 stroke cases were
transported by EMS for other reasons and so were designated
as missed cases. Characteristics of all 441 cases are summarized in Table1. The median age was 78 years, and 59%
were women. A total of 264 cases (59.9%) were confirmed as

76 (17.0)

Sex, female

263 (58.8)

Dispatcher-suspected stroke

318 (72.1)

EMS-suspected IS/TIA

371 (84.1)

Table 2. Multivariable Logistic Regression Model Results

of Accurate EMS Stroke Recognition Among 264 EMSTransported Subjects With Confirmed Ischemic Stroke or TIA

70 (15.9)

Effects

Confirmed TIA

78 (17.7)

CPSS documentation (yes vs no)

Confirmed IS

186 (42.2)

Onset-to-door, 120 min

NIHSS, median, IQR

73.5 (67.7 to 78.7)

52.3 (47.1 to 57.5)

Sensitivity
PPV

n=441 (%)

Age, y, median, IQR

EMS-missed IS/TIA

371

264

Table 1. Characteristics of EMS-Suspected or Missed

Ischemic Stroke or TIA

>90

Not
Ischemic Stroke/TIA

EMS Missed

The methods used to establish the registry have been published previously.16 Briefly, this observational registry was conducted in a single
county in Southwest Michigan, which is served by 3 independent advanced life support transporting EMS agencies that collectively provide >50000 transports per year. Patients who were transported by
EMS with an impression of suspected stroke or who were diagnosed
with IS or TIA after hospital arrival were included, thus capturing
EMS-suspected (false positive), confirmed (true positive), and missed
(false negative) stroke transports. Patients who were transported by
EMS to either of 2 participating primary stroke center hospitals with
a primary or secondary impression of suspected stroke/TIA were
identified from electronic EMS records. We captured EMS-missed
strokes by searching hospital records for patients with a final hospital
discharge diagnosis of stroke or TIA who were transported by EMS.
Hemorrhagic strokes were excluded. All EMS and hospital medical
records were then manually linked. We abstracted data on patient demographics, prehospital care, ED diagnostic testing and treatment,
in-hospital mortality, discharge disposition, and discharge diagnosis.
Because the local stroke transport protocol directs EMS providers to
conduct a CPSS, we recorded the explicit documentation of the CPSS
in the EMS record. This study was approved by the Spectrum Health
Institutional Review Board.
The final diagnosis for all cases was based on the final hospital
discharge diagnosis. Two authors (J.A.O. and T.C.) independently
validated the final hospital discharge diagnoses based on review of
medical records. Inter-rater agreement for a stroke/TIA diagnosis was
high (=0.89). The sensitivity and positive predictive value (PPV) of
EMS stroke recognition were calculated using a final hospital discharge diagnosis as the gold standard. Because of the fact that the
number of true negatives could not be ascertained from our design,
specificity and negative predictive value could not be calculated.
To characterize the role of the CPSS in EMS stroke recognition, we
compared the accuracy of EMS stroke recognition between cohorts
of patients with and without a documented CPSS. The differences in

Characteristics

Ischemic Stroke/TIA

OR (95% CI)
12.02 (5.6625.51)

Age, y

1.00 (0.971.02)

90 (48.4)

Sex, (male vs female)

0.82 (0.411.65)

7 (317)

NIHSS, per unit score

1.09 (1.041.15)

tPA

23 (12.4)

Time from onset, <120 vs >120 min

2.22 (1.124.39)

Endovascular therapy

10 (5.4)

Dispatch as stroke, yes vs no

1.94 (0.914.12)

EMS indicates emergency medical services; IQR, interquartile range; IS,

ischemic stroke; NIHSS, National Institutes of Health Stroke Scale; TIA, transient
ischemic attack; and tPA, tissue-type plasminogen activator.

CI indicates confidence interval; CPSS, Cincinnati Prehospital Stroke Scale;

EMS, emergency medical services; NIHSS, National Institutes of Health Stroke
Scale; OR, odds ratio; and TIA, transient ischemic attack.

Downloaded from http://stroke.ahajournals.org/ by guest on May 26, 2015

Oostema et al Accuracy of Prehospital Stroke Recognition 1515

Table 3. Clinical Characteristics of Patients With Confirmed Ischemic Stroke
(n=186) by EMS Recognition
All Ischemic
Stroke, n=186

EMS Recognized,
n=141

EMS Missed,
n=45

P Value

79 (64.588)

82 (64.588)

0.936

107 (57.5)

79 (56.0)

28 (62.2)

0.464

White

163 (87.6)

123 (87.2)

40 (88.9)

0.769

Black

15 (8.1)

11 (7.8)

4 (8.9)

0.816

Hispanic

3 (1.6)

3 (2.1)

0.324

Asian

2 (1.1)

2 (1.4)

0.422

Hypertension

154 (82.8)

115 (81.6)

39 (86.7)

0.429

Dyslipidemia

119 (64.5)

91 (64.5)

28 (62.2)

0.778

Previous stroke/TIA

73 (39.2)

59 (41.8)

14 (31.1)

0.199

Atrial fibrillation

69 (37.1)

51 (36.2)

18 (40.0)

0.643

Diabetes mellitus

59 (31.7)

44 (31.2)

15 (33.3)

0.79

Coronary artery disease

56 (30.1)

41 (29.1)

15 (33.3)

0.588

Smoking

20 (10.8)

14 (9.9)

6 (13.3)

0.521

Statin

82 (44.1)

66 (46.8)

16 (35.6)

0.186

Antiplatelet (any)

94 (50.5)

74 (52.5)

20 (44.4)

0.348

Anticoagulation (any)

29 (15.6)

22 (15.6)

7 (15.6)

0.994

Demographics
Age, y, median, IQR
Sex

79 (64.588)

Ethnicity

Past medical history

Pre-event treatment

Clinical presentation
NIHSS, median, IQR

7 (318)

10 (419)

4 (19)

<0.001

Unilateral weakness complaint

126 (67.7)

104 (73.8)

22 (48.9)

0.002

Unilateral weakness on examination

128 (68.8)

104 (73.8)

24 (53.3)

0.010

Aphasia

71 (38.2)

55 (39.0)

16 (35.6)

0.678

Dysarthria

88 (47.3)

69 (48.9)

19 (42.2)

0.432

Vision complaints

42 (22.6)

31 (22.0)

11 (25.6)

0.731

Altered mental status

36 (19.4)

28 (19.9)

8 (18.6)

0.758

Ataxia

31 (16.7)

18 (12.8)

13 (30.2)

0.011

Headache

27 (14.5)

18 (12.8)

9 (20.9)

0.230

Vertigo

15 (8.1)

8 (5.7)

7 (16.3)

0.034

Dizziness (nonvertigo)

12 (6.5)

8 (5.7)

4 (9.3)

0.445

Vomiting

11 (5.9)

6 (4.3)

5 (11.6)

0.090

ED treatment
Door-to-CT time, min

34.6

84.7

<0.001

tPA delivery

14.9

4.4

0.074

CT indicates computed tomography; ED, emergency department; EMS, emergency medical services; IQR,
Interquartile range; NIHSS, National Institutes of Health Stroke Scale; TIA, transient ischemic attack; and tPA,
tissue-type plasminogen activator.

having a final discharge diagnosis of either IS (n=186) or TIA

(n=78). Use of the CPSS was documented in the EMS record
in 347 (79%) cases.
The overall sensitivity of EMS stroke recognition was
73.5% (95% confidence interval, 67.778.7), and PPV was
52.3% (95% confidence interval, 47.157.5; Figure1).
Among the 347 cases with a CPSS documented, the sensitivity of EMS provider stroke recognition was higher than
that among the 94 cases where a CPSS was not documented
(84.7% versus 30.9%; P<0.0001). The PPV among cases
with a documented CPSS was also higher (56.2% versus
30.4%; P=0.0004).

In multivariable logistic regression analysis conducted

among the 264 subjects with confirmed IS or TIA, we found
that CPSS documentation was independently associated with
the sensitivity of EMS stroke recognition after adjustment for
patient age, sex, stroke severity (NIHSS), dispatch reason,
and time from symptom onset (OR, 12.02; 95% confidence
interval, 5.6625.51; Table2). Other factors significantly
associated with increased sensitivity of EMS recognition
included whether the subjects were evaluated within 120
minutes of symptom onset (OR, 2.22) and higher NIHSS
(OR, 1.09). CPSS documentation was also independently
associated with higher PPV of EMS stroke suspicion (2.47;

Downloaded from http://stroke.ahajournals.org/ by guest on May 26, 2015

1516StrokeJune 2015
60%
EMS Recognized

50%

EMS Missed

40%
30%
20%

10%
0%
0 to 4

5 to 9

10 to 14

15 to 19

20 or more

NIHSS

Figure 2. Stroke severity among emergency medical services

(EMS) recognized (n=141) and EMS-missed (n=45) strokes.
NIHSS indicates National Institute of Health Stroke Scale.

95% confidence interval, 1.304.69) among the 371 EMSsuspected stroke cases.
The clinical characteristics of the 141 EMS-recognized
IS cases and the 45 missed IS cases are described in Table3.
Demographics and past medical history were similar between
the 2 groups. A complaint of unilateral weakness (73.8% versus 48.9%; P=0.002) and unilateral weakness on examination
(73.8% versus 53.3%; P=0.01) was more common among EMSrecognized than missed IS, whereas vertigo (5.7% versus 16.3%;
P=0.034) and ataxia (12.8% versus 30.2%; P=0.02) were more
common among EMS-missed strokes. The sensitivity of EMS
stroke recognition was the highest among patients who presented with symptoms and signs included in the CPSS (Table I
in the online-only Data Supplement). EMS-recognized strokes
had faster door-to-computed tomographic times (34.6 versus
84.7 minutes; P<0.001), and there was a trend toward greater
likelihood of tPA delivery (14.9% versus 4.4%; P=0.074). EMSrecognized stroke cases had higher stroke severity (median
NIHSS 10 versus 4; MannWhitney U test; P<0.001). The frequency distribution of NIHSS categories is shown in Figure2.
The most common EMS impressions among the 70 missed
stroke transports included generalized weakness (22.9%),
altered mental status (14.3%), and dizziness (10.0%; Table4).
Seven EMS-missed cases (10%) were transported for a focal
neurological complaint, such as unilateral weakness or aphasia
without explicitly identifying the patient as a suspected stroke.
The final diagnoses of the 177 cases transported by EMS as
suspected strokes who were subsequently given a nonstroke
diagnosis are also shown in Table4. Discharge diagnoses were
highly varied among EMS false-positive cases, and more than
1 quarter received a nonspecific, symptom-based discharge
diagnosis after diagnostic workup failed to identify a specific cause. The most common stroke mimics were infections
(12.4%), seizures (11.3%), and syncope (10.2%).

Discussion
Transportation by EMS is an important predictor of improved
in-hospital stroke response and use of tPA for patients with
acute IS.13 These benefits likely stem in part from earlier activation of hospital stroke code processes through prearrival
notification.5 Therefore, accurate prehospital stroke recognition is a critical link in the stroke chain of recovery. Although
prehospital stroke scales are endorsed by national guidelines,17

their real-world effect on EMS stroke recognition is unclear. A

recent meta-analysis of 3 validation studies of the CPSS found
sensitivities ranging from 79% to 95%15; however, a recently
published study conducted in New York demonstrated EMS
sensitivity of only 50% despite CPSS education and incorporation into local stroke protocols.18
In our cohort of EMS-transported cases, EMS sensitivity for
stroke recognition was 74%, slightly lower than the observed
range in previous CPSS validation studies.15 Furthermore,
PPV of EMS suspicion of stroke was only 52%, suggesting
that there is opportunity for improvement by reducing both
the over- and under-recognition of stroke by EMS providers.
Our analysis suggests that a strong relationship exists
between documentation of the CPSS and the sensitivity
(adjusted OR, 12.02) and PPV (adjusted OR, 2.47) of prehospital stroke recognition. These relationships were independent
of stroke severity, dispatch reason, and time from symptom
onset, age, and sex. Our results corroborate those of a recently
published analysis of prehospital stroke recognition, which
reported a similarly strong association between CPSS use and
sensitivity.18 To our knowledge, this is the first study to report
higher PPV among EMS cases with a documented CPSS as
opposed to no documented stroke scale. Although this supports
the hypothesis that use of CPSS improves overall diagnostic
accuracy, we suspect that paramedics may use and document a
CPSS preferentially among patients with more obvious stroke
signs who are already recognized as suspect stroke/TIA cases.
Symptoms and signs not included in the CPSS, such as vertigo (16%) and limb ataxia (30%), were more common among
missed stroke cases. Nevertheless, over half of the EMSmissed strokes demonstrated unilateral weakness in the ED
and only 30% (11/37) of those cases had a documented CPSS,
suggesting that more consistent application of the CPSS in the
prehospital setting could improve EMS sensitivity. Because
nearly half of EMS-missed strokes were transported with
EMS impressions of generalized weakness, altered mental
status, or dizziness, increased use of the CPSS among these
populations may improve EMS stroke recognition sensitivity.
Table 4. Analysis of EMS Stroke Recognition Errors: EMS
Impression Among 70 EMS-Missed Ischemic Stroke/TIA
Cases and Final Discharge Diagnosis for 177 Nonstroke Cases
Transported by EMS as Suspected Stroke

n=70

Discharge Diagnosis
for EMS False-Positive
IS/TIA

n=177

Generalized weakness

16 (22.9)

Infection

22 (12.4)

Altered mental status

10 (14.3)

Seizure

20 (11.3)

7 (10.0)

Syncope/transient
hypotension

18 (10.2)

EMS Impression for

EMS-Missed IS/TIA

Dizziness
Focal neurological finding

7 (10.0)

Complex migraine

13 (7.3)

Cardiovascular

5 (7.1)

Hypertensive emergency

7 (4.0)

Diabetic
Other/not specified

4 (5.7)

Bell palsy

6 (3.4)

21 (30.0)

Miscellaneous specific
diagnosis

43 (24.3)

Nonspecific diagnosis

48 (27.1)

EMS indicates emergency medical services; IS, ischemic stroke; and TIA,
transient ischemic attack.

Downloaded from http://stroke.ahajournals.org/ by guest on May 26, 2015

Oostema et al Accuracy of Prehospital Stroke Recognition 1517

Other factors associated with EMS sensitivity were early
presentation (OR, 2.22) and increasing NIHSS (OR, 1.09 for
each 1 point increase). These findings have also been described
previously18 and suggest that paramedics are more likely to
recognize patients with more obvious stroke presentations or
those perceived to be possible candidates for tPA therapy. An
emergency dispatcher impression of possible stroke was also
associated with a marginally significant increased likelihood
of accurate EMS provider stroke recognition in the multivariable analysis (OR, 1.94). This finding might suggest that
dispatch reasons provide a degree of priming for paramedics
to consider stroke. If so, this may be another potential target
for intervention. Recent studies of the accuracy of emergency
dispatcher stroke recognition suggest fairly low sensitivities1921; however, incorporation of the CPSS into dispatcher
protocols may improve this.22,23 Future studies are needed to
determine whether improved dispatcher recognition translates
into improved EMS provider stroke recognition, prehospital
notification, and thus downstream in-hospital stroke care.
Historically, the prehospital links in the stroke chain of
recovery have received less attention than in-hospital care.
Substantial evidence suggests that EMS use is associated
with faster ED stroke evaluations2,4 and increased opportunity
for treatment with tPA.3 Our results identified opportunity to
improve EMS provider recognition by reducing the rate of
missed strokes and false positives through more consistent
application of the CPSS. Given the critical role EMS stroke
recognition plays in providing high-quality prehospital stroke
care,16 future studies should focus on refinement and implementation of prehospital stroke screening tools and measure
the effect of improvements in recognition on patient outcomes.

Sources of Funding
This study was supported by a Blue Cross Blue Shield of Michigan
Foundation Investigator Initiated Award.

Disclosures
None.

References
1. Turan TN, Hertzberg V, Weiss P, McClellan W, Presley R, Krompf K,
et al. Clinical characteristics of patients with early hospital arrival after
stroke symptom onset. J Stroke Cerebrovasc Dis. 2005;14:272277. doi:
10.1016/j.jstrokecerebrovasdis.2005.07.002.
2. Gache K, Couralet M, Nitenberg G, Leleu H, Minvielle E. The role of
calling EMS versus using private transportation in improving the management of stroke in France. Prehosp Emerg Care. 2013;17:217222.
doi: 10.3109/10903127.2012.755584.
3. Ekundayo OJ, Saver JL, Fonarow GC, Schwamm LH, Xian Y, Zhao X,
et al. Patterns of emergency medical services use and its association
with timely stroke treatment: findings from Get With the GuidelinesStroke. Circ Cardiovasc Qual Outcomes. 2013;6:262269. doi: 10.1161/
CIRCOUTCOMES.113.000089.
4. Sheppard JP, Mellor RM, Greenfield S, Mant J, Quinn T, Sandler D, et al;
CLAHRC BBC Investigators. The association between prehospital care
and in-hospital treatment decisions in acute stroke: a cohort study. Emerg
Med J. 2015;32:9399. doi: 10.1136/emermed-2013-203026.
5. Lin CB, Peterson ED, Smith EE, Saver JL, Liang L, Xian Y, et al. Emergency
medical service hospital prenotification is associated with improved
evaluation and treatment of acute ischemic stroke. Circ Cardiovasc Qual
Outcomes. 2012;5:514522. doi: 10.1161/CIRCOUTCOMES.112.965210.
6. Patel MD, Rose KM, OBrien EC, Rosamond WD. Prehospital notification by emergency medical services reduces delays in stroke evaluation:

findings from the North Carolina stroke care collaborative. Stroke.

2011;42:22632268. doi: 10.1161/STROKEAHA.110.605857.
7. Kidwell CS, Starkman S, Eckstein M, Saver JL. Prevalence of true
strokes and stroke mimics among patients transported to the emergency
department (ED) by prehospital personnel. J Stroke Cerebrovasc Dis.
1998;7:378378.
8. Kothari R, Barsan W, Brott T, Broderick J, Ashbrock S. Frequency
and accuracy of prehospital diagnosis of acute stroke. Stroke.
1995;26:937941.
9. Smith WS, Isaacs M, Corry MD. Accuracy of paramedic identification
of stroke and transient ischemic attack in the field. Prehosp Emerg Care.
1998;2:170175.
10. 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;31:7176.
11. Bray JE, Martin J, Cooper G, Barger B, Bernard S, Bladin C. Paramedic
identification of stroke: community validation of the Melbourne
ambulance stroke screen. Cerebrovasc Dis. 2005;20:2833. doi:
10.1159/000086201.
12. Chenkin J, Gladstone DJ, Verbeek PR, Lindsay P, Fang J, Black SE,
et al. Predictive value of the Ontario prehospital stroke screening tool
for the identification of patients with acute stroke. Prehosp Emerg Care.
2009;13:153159. doi: 10.1080/10903120802706146.
13. Kothari RU, Pancioli A, Liu T, Brott T, Broderick J. Cincinnati
Prehospital Stroke Scale: reproducibility and validity. Ann Emerg Med.
1999;33:373378.
14. Jauch EC, Saver JL, Adams HP Jr, Bruno A, Connors JJ, Demaerschalk
BM, et al; American Heart Association Stroke Council; Council on
Cardiovascular Nursing; Council on Peripheral Vascular Disease;
Council on Clinical Cardiology. Guidelines for the early management of
patients with acute ischemic stroke: a guideline for healthcare professionals from the American Heart Association/American Stroke Association.
Stroke. 2013;44:870947. doi: 10.1161/STR.0b013e318284056a.
15. Brandler ES, Sharma M, Sinert RH, Levine SR. Prehospital stroke scales
in urban environments: a systematic review. Neurology. 2014;82:2241
2249. doi: 10.1212/WNL.0000000000000523.
16. Oostema JA, Nasiri M, Chassee T, Reeves MJ. The quality of prehospital ischemic stroke care: compliance with guidelines and impact on inhospital stroke response. J Stroke Cerebrovasc Dis. 2014;23:27732779.
doi: 10.1016/j.jstrokecerebrovasdis.2014.06.030.
17. Acker JE 3rd, Pancioli AM, Crocco TJ, Eckstein MK, Jauch EC, Larrabee
H, et al; American Heart Association; American Stroke Association
Expert Panel on Emergency Medical Services Systems, Stroke Council.
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;38:30973115.
doi: 10.1161/STROKEAHA.107.186094.
18. Gropen TI, Gokaldas R, Poleshuck R, Spencer J, Janjua N, Szarek M,
et al. Factors related to the sensitivity of emergency medical service
impression of stroke. Prehosp Emerg Care. 2014;18:387392. doi:
10.3109/10903127.2013.864359.
19. Buck BH, Starkman S, Eckstein M, Kidwell CS, Haines J, Huang R, et
al. Dispatcher recognition of stroke using the National Academy Medical
Priority Dispatch System. Stroke. 2009;40:20272030. doi: 10.1161/
STROKEAHA.108.545574.
20. Deakin CD, Alasaad M, King P, Thompson F. Is ambulance telephone
triage using advanced medical priority dispatch protocols able to identify
patients with acute stroke correctly? Emerg Med J. 2009;26:442445.
doi: 10.1136/emj.2008.059733.
21. Caceres JA, Adil MM, Jadhav V, Chaudhry SA, Pawar S, Rodriguez
GJ, et al. Diagnosis of stroke by emergency medical dispatchers and
its impact on the prehospital care of patients. J Stroke Cerebrovasc Dis.
2013;22:e610e614. doi: 10.1016/j.jstrokecerebrovasdis.2013.07.039.
22. De Luca A, Giorgi Rossi P, Villa GF; Stroke Group Italian Society PreHospital Emergency Services. The use of Cincinnati Prehospital Stroke
Scale during telephone dispatch interview increases the accuracy in identifying stroke and transient ischemic attack symptoms. BMC Health Serv
Res. 2013;13:513. doi: 10.1186/1472-6963-13-513.
23. Watkins CL, Leathley MJ, Jones SP, Ford GA, Quinn T, Sutton CJ;
Emergency Stroke Calls: Obtaining Rapid Telephone Triage (ESCORTT)
Group. Training emergency services dispatchers to recognise stroke: an
interrupted time-series analysis. BMC Health Serv Res. 2013;13:318.
doi: 10.1186/1472-6963-13-318.

Downloaded from http://stroke.ahajournals.org/ by guest on May 26, 2015

SUPPLEMENTAL MATERIAL

Supplemental Table I: EMS stroke detection rates among ischemic stroke patients with specific
symptoms and signs
Clinical Characteristic
Unilateral Weakness Complaint
Unilateral Weakness on Exam
Aphasia
Dysarthria
Vision Complaints
Altered Mental Status
Ataxia
Headache
Vertigo
Dizziness (non-vertigo)
Vomiting

Recognized
by EMS

EMS sensitivity
(%)

126
128
71
88
42
36
31
27
15
12
11

104
104
55
69
31
28
18
18
8
8
6

82.5
81.3
77.5
78.4
73.8
77.8
58.1
66.7
53.3
66.7
54.5

Clinical Predictors of Accurate Prehospital Stroke Recognition

J. Adam Oostema, John Konen, Todd Chassee, Mojdeh Nasiri and Mathew J. Reeves
Stroke. 2015;46:1513-1517; originally published online April 28, 2015;
doi: 10.1161/STROKEAHA.115.008650
Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright 2015 American Heart Association, Inc. All rights reserved.
Print ISSN: 0039-2499. Online ISSN: 1524-4628

The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://stroke.ahajournals.org/content/46/6/1513

Data Supplement (unedited) at:

http://stroke.ahajournals.org/content/suppl/2015/04/30/STROKEAHA.115.008650.DC1.html

Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Stroke can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.
Once the online version of the published article for which permission is being requested is located, click
Request Permissions in the middle column of the Web page under Services. Further information about this
process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Stroke is online at:
http://stroke.ahajournals.org//subscriptions/

Downloaded from http://stroke.ahajournals.org/ by guest on May 26, 2015