Embed Size (px)
Citation preview
American Journal of Emergency Medicine 31 (2013) 788–791
Contents lists available at SciVerse ScienceDirect
American Journal of Emergency Medicine
j ourna l homepage: www.e lsev ie r .com/ locate /a jem
Original Contribution
Factors associated with use of emergency medical services in patients withacute stroke☆,☆☆
Nai-Chuan Chen MD a,1, Ming-Ju Hsieh MD, MS a,b,1, Sung-Chun Tang MD c, Wen-Chu Chiang MD, MS a,b,Kuang-Yu Huang RN c, Li-Kai Tsai MD, PhD d, Patrick Chow-In Ko MD, MS a,b,Matthew Huei-Ming Ma MD, PhD a,⁎, Jiann-Shing Jeng MD, PhD c,⁎a Department of Emergency Medicine, National Taiwan University Hospital, Taipei, Taiwanb Graduate Institute of Epidemiology and Preventive Medicine, National Taiwan University, Taipei, Taiwanc Stroke Center and Department of Neurology, National Taiwan University Hospital, Taipei, Taiwand Department of Neurology, National Taiwan University Hospital Yun-Lin Branch, Taipei, Taiwan
☆ This article was supported by Grant No. 100-niversity Hospital.☆☆ The authors report no conflicts of interest. The aue content and writing of this manuscript.⁎ Corresponding authors. Jiann-Shing Jeng, MD,epartment of Neurology, and Matthew Huei-Mingmergency Medicine, National Taiwan University Hos886 2 23123456x65338; fax: +886 2 23418395.
E-mail addresses: [email protected] (M.H.-M. MJ.-S. Jeng).
1 These authors contributed equally to the study.
0735-6757/$ – see front matter © 2013 Elsevier Inc. Alhttp://dx.doi.org/10.1016/j.ajem.2013.01.019
a b s t r a c t
a r t i c l e i n f oArticle history:
Received 18 December 2012Accepted 21 January 2013Purposes: The aim of this study was to investigate the factors associated with use of emergency medicalservices (EMS) in patients with acute stroke.Methods: Prospective data on consecutive patients with acute stroke who presented to the emergencydepartment of a university medical center from January 1, 2010, to July 31, 2011, were analyzed. Patients wereexcluded if they had an unknown residence, had onset of stroke at a nursing home or hospital, or were
transferred from another hospital. Variables for all patients with stroke and ischemic stroke who did and didnot use EMS were compared.Results: In total, 1344 patients (60% male; mean age, 68.7 years) were included. Use of EMS (n = 409; 30.4%)was significantly associated with a higher level of education (≧6 years vs b6 years; odds ratio [OR], 1.69; 95%confidence interval [CI], 1.25-2.29), a higher National Institutes of Health Stroke Scale score (OR, 1.08; 95% CI,1.05-1.11), altered consciousness (OR, 1.88; 95% CI, 1.25-2.84), and atrial fibrillation (OR, 2.43; 95% CI, 1.71-3.44) after adjustment. For patients with ischemic stroke, use of EMS was significantly higher in cases ofcardioembolism (OR, 3.04; 95% CI, 1.40-6.60) and large artery atherothrombosis (OR, 2.10; 95% CI, 1.22-3.62)than lacunar infarction.Conclusion: Patients with stroke who have altered consciousness, a higher level of education, a higher NationalInstitutes of Health Stroke Scale score, atrial fibrillation, and cardioembolic strokeweremore likely to use EMS.© 2013 Elsevier Inc. All rights reserved.
1. Introduction
In Taiwan, stroke has been the second or third leading cause ofdeath in the past 2 decades and leads to great loss of quality-adjustedlife expectancy [1,2]. Each year, there have been more than 80 000hospital admissions for stroke in Taiwan, and the cost of hospitali-zation for stroke is nearly US$170 million [3]. Acute ischemic stroke
S1870 from National Taiwan
thors alone are responsible for
PhD, is to be contacted atMa, MD, PhD, Department ofpital, Taipei 100, Taiwan. Tel.:
U
th
DE+
(
l rights reserved.
(IS) accounts for most cases of stroke, and timely revascularizationwith thrombolytic therapy is the most effective maneuver to salvageischemic brain tissue that is not already infarcted [4]. Intravenousthrombolytic therapy with tissue plasminogen activator (tPA) hasbeen shown to be of benefit for patients with acute IS, but it should beadministered within 4.5 hours of the onset of stroke [5,6]. Despite theevidence in favor of early treatment, the frequency of tPA adminis-tration in patients with IS remains low. In Taiwan, only 1.5% ofpatients with IS were treated with tPA between 2006 and 2008 [7],which is lower than that reported in the United States (3.4%-5.2%) [8]and in the Swedish Stroke Register (6.0%) [9]. Among several reasonsfor excluding thrombolytic therapy for patients with acute IS,prehospital delay represents a significant barrier [10,11]. Somestudies have reported that the use of emergency medical services(EMS) can significantly reduce prehospital delay [12,13].
The American Heart Association and American Stroke Associationrecommend 3 steps to decrease prehospital delay—symptom
789N.-C. Chen et al. / American Journal of Emergency Medicine 31 (2013) 788–791
recognition, calling EMS, and rapid response by EMS—and emphasizethat calling EMS ought to be the first and only response to suspectedsymptoms of stroke [14]. Data have shown that 29% to 65% of patientswith signs or symptoms of acute stroke accessed their initial medicalcare via local EMS [15–20]. However, according to a regional hospitalstudy in Taiwan, only 22% of patients with acute stroke use EMS [21].More efforts are needed to enhance the use of EMS by patients duringthe early presentation of an acute stroke among eligible hospitals inTaiwan. Therefore, the aim of the present study was to analyze thefactors associated with use of EMS in laypersons with acute stroke at amedical center in Taiwan.
2. Materials and methods
This prospective cohort study was based on the stroke registry atNational Taiwan University Hospital (NTUH) [22,23], whichwas startedin 1995 to investigate the risk factors, clinical course, prognosis, andcomplications of patients with different types of stroke, including IS,intracerebral hemorrhage (ICH), subarachnoid hemorrhage (SAH), andtransient ischemic attack (TIA). The diagnostic criteria for stroke and TIAwere in accordance with those defined by the World Health Organiza-tion [24]. Ischemic stroke was further classified based on the diagnosticcriteria adopted from the Trial of Org 10172 in Acute Stroke Treatmentclassification system [25]. If an acute stroke or TIA was diagnosed, thepatientwas invited to join the registry by 1 of the 3 trained studynurses.If the patient was unconscious, the invitation was extended to thepatient's next of kin. After written informed consent was acquired,medical and other associated information about the patient wasgathered prospectively using a structured recording form by interview-ing the patient and his or her relatives and by reviewing the medicalrecord. The representativeness of the registry was previously demon-strated in a comparison with the Database of Catastrophic Illness of theNational Health Insurance,which collected information on all registeredpatients with acute stroke for waiving copayment in Taiwan [2].
National Taiwan University Hospital is a primary and tertiary medicalcenter with approximately 2450 beds and an average of 110 000emergency department (ED) visits per year. Approximately 300 000residents of Taipei City and New Taipei City live in the catchment area ofNTUH. The emergencymedical technicians (EMTs) for the 2 cities provideEMS in thecatchmentarea. TheEMSsystemofbothcities is amixed1-and2-tier fire-based system [26]. Basic life support teams are responsible forproviding EMS to most patients with stroke. The Cincinnati PrehospitalStroke Scale is used by EMTs in both cities to identify patientswith stroke.Asking about the time of onset of symptoms and checking the patient'sbloodglucose level byfinger-stick testing are part of the standardprotocolfor the care of patients with suspected stroke.
We extracted data on the patients with acute stroke or TIA whopresented to the ED at NTUH from January 1, 2010, to July 31, 2011. If apatient visited the ED more than once during the study period, onlythe first ED visit was included for analysis. Abstracted data includedarrival route (EMS system, private vehicle, transfer from otherhospital, transfer from outpatient clinic, or private ambulance), typeof residence (own home, nursing home, other place, and unknown),site of onset of stroke (patient's home, workplace, hospital, placeexcept home, hospital and workplace, and unknown), symptoms ofstroke, findings on physical examination in the ED, risk factors,comorbidity, years of education, diagnostic tests performed, and theresults of diagnostic tests. The severity of stroke/TIA was scored usingthe National Institutes of Health Stroke Scale (NIHSS). Symptoms ofstroke were recorded as documented according to the statement ofthe patient or family members and then were collapsed into 7 groupsthought to be the most relevant to the lay public, as previouslymentioned in another study: (1) weakness of arm, leg, or face; (2)numbness of arm, leg, or face; (3) slurred speech, aphasia, or languageproblems; (4) altered consciousness; (5) headache; (6) visualabnormalities; and (7) dizziness (including vertigo and problems
with balance or coordination) [27]. Patients could have more than 1symptom category assigned; for example, a patient could have hadweakness, numbness, and slurred speech.
Because the intent of this analysis was to study the lay public'sresponse to symptoms of stroke, patients were excluded if theirresidence was a medically supervised environment. Therefore,patients who were residents of nursing homes and patients whohad onset of symptoms or were first evaluated at a medical facilityother than a hospital ED (where the patient was an outpatient or avisitor) were excluded. In addition, patients who had an unknown siteof onset or an unknown site of residence were also excluded. Thestudy received approval from the institutional review board of NTUH.
2.1. Statistical analysis
The χ2 test, Student t test, and Wilcoxon rank sum test were usedfor univariate analyses of variables associated with use of EMS.Multivariate logistic analysis was used to determine the independentvariables of P b .2 in the univariate analyses. Because it was onlypossible for patients with IS to receive thrombolytic therapy, factorsassociated with use of EMS among patients with IS were also searchedusing the previously mentioned statistical methods for univariate andmultivariate analyses. A P value less than .05 was defined asstatistically significant. The SAS software (version 9.2; SAS InstituteInc, Cary, NC) was used in statistical analysis.
3. Results
In total, there were 1764 patients with acute stroke or TIA, with1853 ED visits during the study period. Only the first ED visit of thesame patient was included. Patients were excluded for the followingreasons: unknown residence (n = 69), onset of stroke at a nursinghome (n=7), onset of strokewhile in the hospital (n=8), transferredfrom another hospital (n = 319), missing documentation of arrivalroute (n=1), and unknown site of onset (n=16). Therefore, a total of1344 patients (59.5% male) were included in the analysis. The mean(SD) age was 68.7 (13.9) years. In total, 1025 patients (76.3%) had IS,211 (15.7%) had ICH, 34 (2.5%) had SAH, and 74 (5.5%) had TIA.
Emergencymedical services were used for emergency transport tothe ED in 409 cases (30.4%). Table 1 presents the characteristics ofpatients who used EMS and those who did not. Patients who weremore likely to use EMS for emergent transportation to the ED wereolder and had a higher level of education, hypertension, history ofstroke, atrial fibrillation (AF), congestive heart failure, strokeoccurring outside the home, a higher NIHSS score on arrival in theED, and presentation symptoms such as weakness, numbness, alteredconsciousness, slurred speech/aphasia, and visual symptoms. Incontrast, patients with symptoms of dizziness/vertigo/imbalance/incoordination were less likely to use EMS. Multivariate logisticregression analysis showed that a higher level of education, a higherNIHSS score on arrival in the ED, AF, and altered consciousness at theonset of stroke were associated with use of EMS (Table 2). There wasno difference in use of EMS among the different types of stroke.
Because acute IS is more time sensitive toward early thrombolytictreatment, further analysis focused on this subgroup specifically. Ofthe 1025 patients with IS, 290 used EMS and 735 did not (Table 3).Older age, a higher level of education, hypertension, prior stroke, AF,congestive heart failure, smoking habits, and a higher NIHSS score onarrival in the ED were associated with use of EMS. The presentingsymptoms of weakness, numbness, altered consciousness, slurredspeech/aphasia, and visual problem were also significantly related touse of EMS. Multivariate logistic regression analysis showed that ahigher level of education, altered consciousness at onset of stroke, ahigher NIHSS score on arrival in the ED, and subtypes of IS wereassociated with use of EMS (Table 4). Patients with initial NIHSSscores of 15 to 24 (odds ratio [OR], 4.73; 95% confidence interval [CI],
Table 1Comparison between patients with acute stroke with and without use of EMS(n = 1344)
Use of EMS(n = 409)
No use of EMS(n = 935)
P
Age (y), mean (SD) 71.0 (14.4) 67.7 (13.6) b .001Male 249 (60.9%) 550 (58.8%) .480Level of educationNone 60 (14.7%) 129 (13.8%) .1531-6 y 89 (21.8%) 236 (25.2%)7-12 y 132 (32.3%) 280 (29.9%)N12 y 115 (28.1%) 214 (22.9%)Unknown 13 (3.2%) 76 (8.1%)Risk factorsDiabetes mellitus 139 (34.0%) 337 (36.0%) .468Hypertension 328 (80.2%) 710 (75.9%) .087Prior stroke 139 (34.0%) 266 (28.5%) .042AF 144 (35.2%) 128 (13.7%) b .001Congestive heart failure 29 (7.1%) 38 (4.1%) .019Hypercholesterolemia 112 (27.4%) 279 (29.8%) .362Smoking habit 117 (28.6%) 246 (26.3%) .375Place where stroke occurredHome 350 (85.6%) 819 (87.6%) .084Workplace 9 (2.2%) 33 (3.5%)Other place except home, hospital,and workplace
50 (12.2%) 83 (8.9%)
NIHSS score on ED arrival, median (IQR) 12 (5-19) 3 (2-7) b .001Type of strokeIS 290 (70.9%) 735 (78.6%) b .001ICH 92 (22.5%) 119 (12.7%)SAH 11 (2.7%) 23 (2.5%)TIA 16 (3.9%) 58 (6.2%)Presentation symptoms/signsWeakness 354 (86.6%) 678 (72.5%) b .001Numbness 263 (64.3%) 462 (49.4%) b .001Altered consciousness 180 (44.0%) 113 (12.1%) b .001Slurred speech/aphasia 305 (74.6%) 508 (54.3%) b .001Headache 33 (8.1%) 84 (9.0%) .584Visual problem 169 (41.3%) 173 (18.5%) b .001Dizziness/vertigo/imbalance/incoordination
89 (21.8%) 264 (28.2%) .013
Values are expressed as number (percentage), except for age and NIHSS scores.IQR, interquartile range.
Table 3Comparison of patients with acute IS with and without use of EMS (n = 1025)
Use of EMS(n = 290)
No use of EMS(n = 735)
P
Age (y), mean (SD) 73.0 (13.2) 69.0 (12.8) b .001Male 177 (61.0%) 431 (58.6%) .482Level of educationNone 43 (14.8%) 109 (14.8%) .0641-6 y 62 (21.4%) 190 (25.9%)7-12 y 94 (32.4%) 226 (30.7%)N12 y 83 (28.6%) 150 (20.4%)Unknown 8 (2.8%) 60 (8.2%)Risk factorsDiabetes mellitus 95 (32.8%) 279 (38.0%) .119Hypertension 232 (80.0%) 558 (75.9%) .162Prior stroke 105 (36.2%) 220 (29.9%) .052AF 129 (44.5%) 114 (15.5%) b .001Congestive heart failure 26 (9.0%) 34 (4.6%) .008Hypercholesterolemia 92 (31.7%) 237 (32.2%) .872Smoking habit 90 (31.0%) 195 (26.5%) .138Place where stroke occurredHome 252 (86.9%) 633 (86.1%) .187Workplace 7 (2.4%) 19 (2.6%)Other place except home, hospital,and workplace
31 (10.7%) 53 (7.2%)
NIHSS score on ED arrival,median (IQR)
11 (5-19) 4 (2-6) b .001
Subtypes of ISLarge artery atherothrombosis 53 (18.3%) 126 (17.1%) b .001Cardioembolism 131 (45.2%) 107 (14.6%)Lacunar infarction 29 (10.0%) 234 (31.8%)Other specific causes 8 (2.8%) 38 (5.2%)Undetermined cause 69 (23.8%) 230 (31.3%)Presentation symptoms/signsWeakness 250 (86.2%) 548 (74.6%) b .001Numbness 191 (65.9%) 373 (50.8%) b .001Altered consciousness 112 (38.6%) 64 (8.7%) b .001Slurred speech/aphasia 222 (76.6%) 413 (56.2%) b .001Headache 14 (4.8%) 43 (5.9%) .520Visual problem 126 (43.5%) 131 (17.8%) b .001Dizziness/Vertigo/Imbalance/Incoordination
68 (23.5%) 215 (29.3%) .061
Values are expressed as number (percentage), except for age and NIHSS scores.IQR, interquartile range.
Table 4
790 N.-C. Chen et al. / American Journal of Emergency Medicine 31 (2013) 788–791
2.25-9.94) were more likely to use EMS. Compared with patients withlacunar infarction, those with cardioembolism (OR, 3.04; 95% CI, 1.40-6.60) and large artery atherothrombosis (OR, 2.10; 95% CI, 1.22-3.62)were more likely to use EMS.
4. Discussion
Approximately 30% of the patients in our study used EMS. If weincluded all patients with known arrival route during the studyperiod, the proportion of patients with strokewho used EMSwould besimilar to that reported in a prior study in a regional hospital inTaiwan [22] but relatively lower than some previous reports of 38% to65% [28]. Awareness of stroke with recognition of symptoms isimportant in improving prehospital delay [29–32]. In other reports,symptoms of weakness and speech problems were most associated
Table 2Multivariate analysis of factors related to use of EMS in patients with acute stroke
β OR 95% CI P
Level of education(≧6 y vs b6 y)
0.5243 1.69 1.25-2.29 b .001
Initial NIHSS score on ED arrival 0.0759 1.08 1.05-1.11 b .001AF 0.8862 2.43 1.71-3.44 b .001Altered consciousness 0.6324 1.88 1.25-2.84 .003Types of strokeIS, TIA – 1.00 – –
ICH 0.0222 1.02 0.68-1.54 .925SAH −0.5382 0.58 0.20-1.73 .332
with increased use of EMS [16,27]. Among the 7 common symptomsof stroke that we categorized, altered consciousness was the onlysymptom that resulted in a call to EMS. Some studies have shown thatpatients with stroke would have arrived at the ED earlier if they had adisturbance in consciousness [33,34]. Emergency medical servicesmay be called not because stroke is recognized or suspected, butbecause it is believed that the severe symptoms may be lifethreatening [35]. It is also possible that stroke is suspected, but thepatient would rather go to the hospital by another mode oftransportation instead of EMS. Our study reveals that more publiceducation programs regarding the identification of symptoms ofstroke and the need to call EMS immediately if stroke is suspected areneeded to decrease prehospital delay. In addition, more efforts are
Multivariate analysis of factors related to use of EMS in patients with acute IS
β OR 95% CI P
Level of education (≧6 y vs b6 y) 0.5327 1.70 1.19-2.44 .004Initial NIHSS score on ED arrivalb5 – 1.00 – –
5-14 0.8683 2.38 1.49-3.80 b .00115-24 1.5547 4.73 2.25-9.94 b .001N25 1.3237 3.76 1.31-10.78 .014Altered consciousness 0.9023 2.47 1.47-4.15 b .001Subtypes of ISLacunar infarction – 1.00 – –
Large artery atherothrombosis 0.7405 2.10 1.22-3.62 .008Cardioembolism 1.1122 3.04 1.40-6.60 .005Undetermined or other specific causes 0.4852 1.63 0.93-2.85 .090
791N.-C. Chen et al. / American Journal of Emergency Medicine 31 (2013) 788–791
required to understand why patients with stroke are reluctant to useEMS and then try to resolve the barriers to use of EMS.
In addition to altered consciousness, our study showed that theinitial severity of stroke was related to use of EMS, especially in patientswith initial NIHSS scores of 15 to 24. It is reasonable that patients with ahigher severity of stroke tend to call for help from professional EMSbecause the condition of such patients is viewed as serious [33,36]. Inaddition, moving patients with higher NIHSS scores to the hospital viaother modes of transportation is more difficult than via EMS, whichprovides at least 2 EMTs who can provide medical assistance as well asmove the patient with a stretcher and transport the patient to the ED byambulance. We also found that AF was an independent factor for use ofEMS in patients with stroke, and patients with a cardioembolic strokewere more likely to use EMS. Approximately 15% of all strokes areembolic strokes, and the embolus suddenly blocks the recipient site sothat the onset of symptoms is abrupt and usually maximal at the start.Because of the abrupt manifestation of symptoms, these patients maybe more alert to the abnormalities and seek immediate help. In ourstudy, patients with large artery atherothrombosis were also morelikely to arrive at the ED by EMS than patients with lacunar infarction.The hypothesis is that large artery atherothrombosis may cause moretypes of symptoms of stroke at the same time than lacunar infarctionand thus is easily viewed as an emergency.
Patientswith a higher level of educationweremore likely to use EMSin our study; these patients may have more of an opportunity to learnabout stroke. Mikulík et al [37] conducted a national survey andconcluded that the decision to use EMS was influenced by theknowledge that stroke is a serious and treatable disease and not byrecognitionof symptoms. Education should therefore enhance symptomrecognition and time-sensitive treatabilitywith immediate activation ofEMS, and then stroke will not be missed or delayed in diagnosis.
A prior study concluded that hemorrhagic stroke is independentlypredictive of use of EMS despite adjustment for the severity of stroke[36], and the result is speculated to be caused by a symptom-specificeffect. The analysis in our study removed such an effect by adjustingfor the symptoms of stroke, and the results did not show that the typeof stroke was associated with use of EMS.
There are some limitations to the present study. First, we did notknow the environment where the patients lived and the prestrokestatus of the patients; these factors may affect utilization by patientsor their family members. For example, EMS may be used when apatient is bedridden and lives in a high-rise building without anelevator. Second, we did not evaluate and adjust the severity of eachsymptom individually but used only the NIHSS score on arrival in theED to measure the severity of stroke. A patient with severe headachewith a pain score of 10 may use EMS, but his or her NIHSS score maybe zero. This may cause some bias. Third, we did not evaluate theknowledge and awareness of stroke and attitudes toward EMS inpatients with stroke, which may affect the motivation to use EMS.
5. Conclusion
Emergency medical services have not yet been widely used bypatients experiencing stroke in Taiwan. We found that patients with ahigher level of education, altered consciousness, a higher NIHSS scoreon arrival in the ED, a history of AF, and cardioembolic stroke weremore likely to use EMS for emergent transportation. More efforts areneeded to increase public awareness of stroke and, most importantly,the prompt use of EMS once the symptoms of stroke are recognized.
References
[1] Jeng JS, Su TC. Epidemiological studies of cerebrovascular diseases and carotidatherosclerosis in Taiwan. Acta Neurol Taiwan 2007;16:190–202.
[2] Lee HY, Hwang JS, Jeng JS, et al. Quality-adjusted life expectancy (QALE) and loss ofQALE for patients with ischemic stroke and intracerebral hemorrhage: a 13-yearfollow-up. Stroke 2010;41:739–44.
[3] Lin HC, Xirasagar S, Chen CH, et al. Association between physician volume andhospitalization costs for patients with stroke in Taiwan: a nationwide population-based study. Stroke 2007;38:1565–9.
[4] Adams Jr HP, del Zoppo G, Alberts MJ, et al. Guidelines for the earlymanagement ofadults with ischemic stroke: a guideline from the American Heart Association/American Stroke Association Stroke Council. Circulation 2007;115:e478–534.
[5] The National Institute of Neurological Disorders and Stroke rt-PA Stroke StudyGroup. Tissue plasminogen activator for acute ischemic stroke. N Engl J Med 1995;333:1581–7.
[6] Hacke W, Kaste M, Bluhmki E, et al. Thrombolysis with alteplase 3 to 4.5 hoursafter acute ischemic stroke. N Engl J Med 2008;359:1317–29.
[7] Hsieh FI, Lien LM, Chen ST, et al. Get With the Guidelines–Stroke performanceindicators: surveillance of stroke care in the Taiwan Stroke Registry: Get With theGuidelines–Stroke in Taiwan. Circulation 2010;122:1116–23.
[8] Adeoye O, Hornung R, Khatri P, et al. Recombinant tissue-type plasminogenactivator use for ischemic stroke in the United States: a doubling of treatmentrates over the course of 5 years. Stroke 2011;42:1952–5.
[9] Asplund K, Glader EL, Norrving B, et al. Effects of extending the time window ofthrombolysis to 4.5 hours: observations in the Swedish Stroke Register (Riks-Stroke). Stroke 2011;42:2492–7.
[10] Evenson KR, RosamondWD, Morris DL. Prehospital and in-hospital delays in acutestroke care. Neuroepidemiology 2001;20:65–76.
[11] Johnson M, Bakas T. A review of barriers to thrombolytic therapy: implications fornursing care in the emergency department. J Neurosci Nurs 2010;42:88–94.
[12] Schroeder EB, Rosamond WD, Morris DL, et al. Determinants of use of emergencymedical services in a population with stroke symptoms: the Second Delay inAccessing Stroke Healthcare (DASH II) Study. Stroke 2000;31:2591–6.
[13] Rosamond WD, Gorton RA, Hinn AR, et al. Rapid response to stroke symptoms: theDelay in Accessing Stroke Healthcare (DASH) study. Acad EmergMed 1998;5:45–51.
[14] Jauch EC, Cucchiara B, Adeoye O, et al. Part 11: adult stroke: 2010 American HeartAssociation guidelines for cardiopulmonary resuscitation and emergency cardio-vascular care. Circulation 2010;122:S818–28.
[15] Wein TH, Staub L, Felberg R, et al. Activation of emergency medical services foracute stroke in a nonurban population: the T.L.L. Temple Foundation StrokeProject. Stroke 2000;31:1925–8.
[16] Handschu R, Poppe R, Rauss J, et al. Emergency calls in acute stroke. Stroke 2003;34:1005–9.
[17] Rossnagel K, Jungehülsing GJ, Nolte CH, et al. Out-of-hospital delays in patientswith acute stroke. Ann Emerg Med 2004;44:476–83.
[18] Lacy CR, Suh DC, Bueno M, et al. Delay in presentation and evaluation for acutestroke: Stroke Time Registry for Outcomes Knowledge and Epidemiology(S.T.R.O.K.E.). Stroke 2001;32:63–9.
[19] Reeves MJ, Arora S, Broderick JP, et al. Acute stroke care in the US: results from 4pilot prototypes of the Paul Coverdell National Acute Stroke Registry. Stroke 2005;36:1232–40.
[20] CaliforniaAcute StrokePilot Registry (CASPR) Investigators. Prioritizing interventionsto improve rates of thrombolysis for ischemic stroke. Neurology 2005;64:654–9.
[21] Chang KC, Tseng MC, Tan TY. Prehospital delay after acute stroke in Kaohsiung,Taiwan. Stroke 2004;35:700–4.
[22] Jeng JS, Lee TK, Chang YC, et al. Subtypes and case-fatality rates of stroke: ahospital-based stroke registry in Taiwan (SCAN-IV). J Neurol Sci 1998;156:220–6.
[23] Yip PK, Jeng JS, Lee TK, et al. Subtypes of ischemic stroke: a hospital-based strokeregistry in Taiwan (SCAN-IV). Stroke 1997;28:2507–12.
[24] World Health Organization. Cerebrovascular disease, prevention, treatment, andrehabilitation. Geneva: WHO Tech Rep; 1971. Ser No 469.
[25] Adams Jr HP, Bendixen BH, Kappelle LJ, et al. Classification of subtype of acuteischemic stroke. Definitions for use in a multicenter clinical trial. TOAST. Trial ofOrg 10172 in Acute Stroke Treatment. Stroke 1993;24:35–41.
[26] MaMH, ChiangWC, Ko PC, et al. A randomized trial of compression first or analyzefirst strategies in patients with out-of-hospital cardiac arrest: results from anAsian community. Resuscitation 2012;83:806–12.
[27] Kleindorfer D, Lindsell CJ, Moomaw CJ, et al. Which stroke symptoms prompt a911 call? A population-based study. Am J Emerg Med 2010;28:607–12.
[28] Kwan J, Hand P, Sandercock P. A systematic review of barriers to delivery ofthrombolysis for acute stroke. Age Ageing 2004;33:116–21.
[29] Wester P, Rådberg J, Lundgren B, et al. Factors associated with delayed admission tohospital and in-hospital delays in acute stroke and TIA: a prospective, multicenterstudy. Seek-Medical-Attention-in-Time Study Group. Stroke 1999;30:40–8.
[30] Papapanagiotou P, Iacovidou N, Spengos K, et al. Temporal trends and associatedfactors for pre-hospital and in-hospital delays of stroke patients over a 16-yearperiod: the Athens study. Cerebrovasc Dis 2011;31:199–206.
[31] Iguchi Y, Wada K, Shibazaki K, et al. First impression at stroke onset plays animportant role in early hospital arrival. Intern Med 2006;45:447–51.
[32] Kim YS, Park SS, Bae HJ, et al. Stroke awareness decreases prehospital delay afteracute ischemic stroke in Korea. BMC Neurol 2011;11:2.
[33] Maestroni A, Mandelli C, Manganaro D, et al. Factors influencing delay inpresentation for acute stroke in an emergency department in Milan, Italy. EmergMed J 2008;25:340–5.
[34] Kim HJ, Ahn JH, Kim SH, et al. Factors associated with prehospital delay for acutestroke in Ulsan, Korea. J Emerg Med 2011;41:59–63.
[35] Teuschl Y, Brainin M. Stroke education: discrepancies among factors influencingprehospital delay and stroke knowledge. Int J Stroke 2010;5:187–208.
[36] Adeoye O, Lindsell C, Broderick J, et al. Emergency medical services utilization bystroke patients: a population-based study. Am J Emerg Med 2009;27:141–5.
[37] Mikulík R, Bunt L, Hrdlicka D, et al. Calling 911 in response to stroke: a nationwidestudy assessing definitive individual behavior. Stroke 2008;39:1844–9.
