ABSTRACT

Introduction: The first opportunity for prompt identification of a stroke in the prehospital environment often occurs when people telephone for emergency medical services. A better understanding of how callers and dispatchers communicate during emergency calls may assist dispatcher identification of stroke.

Objective: To conduct a systematic review of the literature to determine communication factors associated with the identification of stroke during emergency calls.

Methods: Six databases were searched (CINAHL, Cochrane, Embase, Informit, MEDLINE and PsychInfo). To meet eligibility criteria, studies of emergency calls must have analysed communication in relation to whether the dispatcher identified the call as "stroke" or incorrectly classified the call as "other than stroke," and included a stroke diagnosis by either paramedics at the scene or physicians in the hospital.

Results: Database searching returned 696 unique citations, of which only four met the eligibility criteria. Studies included in this review considered only one type of communication factor associated with dispatcher identification of stroke, which was the caller's use of keywords. The two key findings were: (1) caller mention of a "stroke" or typical stroke symptoms such as speech disturbance, facial weakness, and arm/limb weakness (i.e., FAST symptoms) was associated with increased dispatcher identification of stroke; and (2) caller description of the stroke as a fall-related event was associated with decreased detection of stroke.

Conclusions: There is some evidence to suggest that particular keywords used by callers may help or hinder stroke identification by the dispatcher. It might be helpful for dispatchers to be aware of these common words and phrases. Future research could investigate how various factors in both the caller's and dispatcher's communication (such as accent, tone of voice, and use of grammatical structures) may affect identification of stroke by dispatchers.

INTRODUCTION

Acute stroke is a time-critical medical emergency. Definitive treatments, such as thrombolysis, are most effective when administered as early as possible after stroke.¹ The first opportunity for prompt identification of a stroke in the prehospital environment often occurs when people telephone for emergency medical services.² If stroke is identified during an emergency call, then ambulance resources can be dispatched at a high priority and time to treatment can be reduced.³ However, estimates of stroke identification by dispatchers indicate only one-third^{4, 5, 6} to one-half^{7, 8, 9, 10} of all stroke cases are identified. Efforts to increase dispatcher identification of stroke have involved the introduction of various prehospital stroke screening tests¹¹ and the use of scripted protocols to interview the caller.¹² Another way to increase stroke identification during emergency calls may be to educate dispatchers about communication factors that are important for identifying stroke.

Communication factors are broadly defined here as any factors in the dispatcher's or caller's communication that are associated with stroke identification by dispatchers. Examples of caller communication factors could be the specific words used to convey information about a patient's complaint (i.e., keywords), the caller's language proficiency (such as native or non-native language proficiency), or the way words are spoken by them (e.g., accent, voice quality, or use of grammatical structures). Dispatchers may or may not communicate using a scripted protocol of questions to interview callers, depending on the type of emergency dispatch system in place, but even when dispatchers follow a pre-specified scripted protocol, they may vary in communication factors such as their tone of voice or the way they clarify ambiguous information from the caller (e.g., direct or indirect questioning). A systematic review of the literature is a useful method to determine which communication factors are known to help or hinder dispatcher identification of stroke. This type of literature review uses systematic and transparent methods to identify, select, and critically evaluate studies in order to answer a pre-formulated research question.¹³

OBJECTIVE

To conduct a systematic review of the literature to determine communication factors associated with the identification of stroke during emergency calls.

METHODS

The systematic review method involves several steps.¹⁴ The first step is to define the research question and specify the research protocol. An exhaustive search of the literature is then undertaken to identify potentially relevant studies. The studies identified are then screened against pre-specified criteria to assess their relevancy for inclusion. Each study included in the review is then evaluated in terms of its methodological quality to determine the validity of the study's findings. Data from each study is extracted and then collated to produce a meaningful summary; this process may or may not use statistical techniques (i.e., meta-analysis). The systematic and explicit methods aim to minimize bias and allow for the study to be replicated.

The methods used in this paper were specified in advance (before the literature search commenced) and made publically available through PROSPERO (an international registry of prospective health-related studies). The registration number of this paper is (CRD42015026086).¹⁵ This systematic review is reported according to PRISMA guidelines,¹⁶ the accepted guidelines in healthcare for transparent and complete reporting of a systematic review.

A paper was eligible for inclusion if the study: 1) was an original study examining communication in authentic emergency calls (i.e., not simulated emergency calls); 2) provided information on whether the dispatcher identified the call as "stroke" or incorrectly classified the call as "other than stroke"; and 3) had a diagnosis of stroke confirmed upon patient assessment by either paramedics at the scene or physicians in a hospital. Commentaries, letters, dissertations, editorials, and conference abstracts were excluded.

Six databases were searched, including CINAHL (ESBCO), Cochrane, Embase, Informit, MEDLINE, and PsychInfo, from their earliest availability up to September 30th, 2015. The search did not have restrictions for publication year or language of publication. Two authors independently screened the titles and abstracts of all identified citations and assessed the eligibility of full-text articles against the inclusion criteria. The reference lists of included papers were also screened by one author to further locate relevant papers.

Each included study in this review was assessed for methodological quality (i.e., risk of bias) using an objective assessment tool called the Newcastle-Ottawa Scale (NOS) for accessing the quality of non-randomised studies for cohort studies.¹⁷ Two authors independently performed this task. NOS evaluates risk of bias according to three categories, including: 1) the selection of the study groups; 2) the comparability of the groups; and 3) the ascertainment of outcome of interest. Studies were rated for risk of bias per category since an overall study rating is not deemed informative.¹⁸

Data items were extracted from each study by one author and checked for accuracy by another author. Items included author(s), study location, number of participants, year of study, dispatcher qualification, whether the stroke diagnosis was made by paramedics at the scene or physicians in the hospital, study design, method of data collection, dispatcher classification of "stroke" or "other than stroke," type of stroke diagnosis (e.g., ischemic or hemorrhagic), frequency of keywords/symptoms used by callers, and type of statistical analysis conducted with corresponding p-value. The frequency of caller-reported keywords/symptoms was compared for calls identified as "stroke" against calls incorrectly classified as "other than stroke." We conducted a Chi-Square analysis on raw data to determine statistically significant differences in the frequency of caller keywords/symptoms between calls identified as "stroke" and calls incorrectly classified as "other than stroke."

RESULTS

Database searching retrieved a total of 774 citations, including 225 articles from Medline, 238 from CINCAHL, 69 from Embase, 25 from Informit, 70 from PyschInfo, and 147 from Cochrane. Three records were identified by hand-searching relevant journals and reference lists of included articles. After removing duplicate records, 696 unique articles remained and were screened for eligibility by title and abstract. The full text of 22 articles were assessed for eligibility. In total, four articles met the inclusion criteria. The study selection process and reasons for exclusion are shown in Figure 1. A number of studies that examined communication in emergency calls for patients with a stroke diagnosis were excluded because they did not include the dispatcher's classification of the call as either "stroke" or "other than stroke."^{19, 20, 21, 22, 23, 24, 25, 26, 27, 28}

The main characteristics of the four reviewed studies are summarized in Table 1. All studies used an observational design and all dispatchers were health professionals.^{29, 30, 31, 32} Three studies were from non-English speaking countries.^{33, 34, 35} One study used a paramedic diagnosis of stroke at the scene as the "final" stroke diagnosis,³⁶ while the other three used a physician diagnosis of stroke at some point in hospital as the "final" stroke diagnosis.^{37, 38, 39} Patients with hemorrhagic stroke were excluded in one study,⁴⁰ while another study excluded patients with Transient Ischaemic Attack.⁴¹ No study included a comparison group of patients without stroke at final diagnosis. The methods used to collect data varied across studies; two studies used written patient records,^{42, 43} and two studies transcribed emergency audio calls.^{44, 45} No study examined the contribution of the dispatcher's communication in the identification of stroke. The only aspect of caller communication studied was keywords/symptoms. Researchers categorized the caller's keywords into symptom categories but varied in the way this was done. For example, Chenaitia et al.⁴⁶ combined keywords relating to "vertigo" with keywords relating to "falls," whereas Berglund et al.⁴⁷ combined keywords related to "lying position" with "falls."

The results of the risk of bias assessment are provided in Table 2.⁴⁸ Since all studies used an observational design, there is a possibility that the association between the caller's communication and identification of stroke may be attributable to confounding factors.

Studies included in this review differed in terms of important methodological factors, such as the way communication variables were defined and measured and how stroke was diagnosed. Such variability between studies means that combining the data using statistical methods (meta-analysis) might produce an invalid result.⁴⁹ Therefore, the data were descriptively synthesized in terms of factors that were associated with increased dispatcher identification of stroke and factors associated with decreased detection of stroke (Table 4). A summary statement of the main finding(s) for each study is presented in Table 3.

Caller-reported keywords that might help dispatcher identification of stroke

Caller keywords or symptoms that were associated with increased dispatcher identification of stroke are presented in Table 4. Caller mention of "stroke" or a history of a previous stroke/cardiovascular event was associated with increased identification of stroke by dispatchers.^{50, 51, 52} Porteous et al.⁵³ found that caller report of "stroke" was more frequently reported in calls of stroke patients that were correctly identified as "stroke" by the dispatcher (79%) versus calls incorrectly classified as "other than stroke" by the dispatcher (38%). Berglund et al.⁵⁴ also reported that caller report of "stroke" was more frequently reported in calls of stroke patients that were correctly identified as "stroke" by the dispatcher (58%) versus calls incorrectly classified as "other than stroke" by the dispatcher (33%). Chenaitia et al.9 noted that caller report of a previous stroke/cardiovascular event was more frequently reported in calls of stroke patients that were correctly identified as "stroke" by the dispatcher (58%) versus calls incorrectly classified as "other than stroke" by the dispatcher (44%).

Two studies^{55, 56} reported that caller mention of typical stroke symptoms such as speech disturbance, facial weakness, and arm/limb weakness (FAST symptoms) was associated with increased dispatcher identification of stroke. Chenaitia et al.⁵⁷ found that in 87% of calls identified as "stroke," the caller reported one or more FAST symptom(s) whereas these symptoms were only reported by callers in 56% of calls incorrectly classified.⁵⁸ The frequency of particular FAST related symptoms in the study by Chenaitia et al.⁵⁹ was as follows: limb weakness (65% vs. 34%), speech disturbance (52% vs. 26%), and facial paresis (30% vs. 12%). Berglund et al.⁶⁰ found that in 80% of calls identified as "stroke," the caller reported one or more FAST symptom(s), whereas these symptoms were only reported by callers in 35% of incorrectly classified calls. The frequency of particular FAST related symptoms in the study by Berglund et al.⁶¹ was as follows: facial weakness (25% vs. 2%), arm weakness (21% vs. 5%) speech disturbance (68% vs. 31%), unilateral symptoms (23% vs. 5%), and numbness/sensory loss (13% vs. 2%).

Lastly, "confusional state" and "impaired consciousness" were associated with increased identification of stroke by dispatchers. De Luca et al.⁶² found that caller report of "confusional state" was more frequently reported in calls identified as "stroke" by the dispatcher (22%) versus calls incorrectly classified as "other than stroke" (18%). Chenaitia et al.⁶³ found that caller report of "impaired consciousness" was more frequent in calls identified as "stroke" by the dispatcher (10%) versus calls incorrectly classified as "other than stroke" (6%).

Caller-reported keywords that might hinder dispatcher identification of stroke

Caller keywords or symptoms that were associated with decreased detection of stroke by dispatchers are presented in Table 4. Berglund et al.⁶⁴ reported that calls incorrectly classified as something "other than stroke," compared with calls identified as "stroke," had more frequent caller report of "fall/lying" (66% versus 22%). Chenaitia et al.⁶⁵ found calls that were incorrectly classified as "other than stroke" compared with calls identified as stroke, had more frequent caller report of: "general discomfort" (23% versus 9%), "fall/vertigo" (12% versus 5%), and "chest pain/dyspnoea" (13% versus 6%). De Luca et al.⁶⁶ noted that calls incorrectly classified compared with calls identified as "stroke," had more frequent caller report of "not breathing" (7% versus 4%). In summary, a salient finding was the agreement between two studies that caller description of the stroke as a fall-related event was associated with decreased detection of stroke.^{67, 68}

TABLES AND GRAPHS

DISCUSSION

Our systematic review of the literature identified four studies that examined communication factors associated with identification of stroke by dispatchers. Although this systematic review sought to identify any type of communication factor that might be associated with dispatcher identification of stroke, the included studies only considered one communication factor, caller's keywords. Two key findings of this review were: (1) caller mention of a "stroke" or typical stroke symptoms such as speech disturbance, facial weakness, and arm/limb weakness was associated with increased dispatcher identification of stroke; and (2) caller description of the stroke as a fall-related event was associated with decreased detection of stroke.

It was unsurprising that caller report of "stroke" was associated with increased identification of stroke in the studies reviewed.^{69, 70} This is consistent with a study by Reginella et al.⁷¹ that established that caller report of "stroke" was highly predictive of a final stroke diagnosis, with a positive predictive ratio of 2.27. A similar study by Krebes et al.⁷² found that caller report of "stroke" was frequent in calls for patients with a final stroke diagnosis (46%) but infrequent in calls for patients without a final stroke diagnosis (11%). Taken together, these findings suggest that in some calls it could be useful for dispatchers to prompt for stroke symptoms if the caller describes the medical emergency as a "stroke." There was agreement in the reviewed studies that caller report of FAST symptoms such as speech disturbance, facial weakness, and arm/limb weakness was associated with increased identification of stroke.^{73, 74} This result was anticipated, since similar studies by Handschu et al.⁷⁵ and Rosamond et al.⁷⁶ found that caller report of FAST symptoms was associated with final diagnosis of stroke. Furthermore, studies by Reginella et al.⁷⁷ and Krebes et al.⁷⁸ indicated that when the final diagnosis was not stroke, callers did not frequently mention FAST symptoms. Such findings support the validity of stroke screening tests that elicit FAST symptoms (such as the Stroke Diagnostic Tool used in the Medical Priority Dispatch System™, MPDS®).

An interesting finding of this review was that caller description of the stroke as a fall-related event was associated with decreased detection of stroke.^{79, 80} This result is similar to the finding by Deakin et al.⁸¹ that the most frequent dispatch code for stroke patients who failed to be detected as stroke at dispatch was "fall/collapse" at 34%. It is likely that callers describe a stroke event as a fall because a fall can be the motor consequence of a stroke.⁸² Despite the frequent report of "fall" by callers for stroke patients, the usefulness of this term must be evaluated against how frequently callers report this term in calls when the final diagnosis is not stroke. Krebes et al.⁸³ found caller report of a fall occurred in 51% of calls for stroke patients but also in 31% of calls when the final diagnosis was not stroke. Similarly, Leathley et al.⁸⁴ found caller report of a fall occurred in 38% of calls for stroke patients but also in 26% of calls when the final diagnosis was not stroke. Since caller report of a fall event was reported with similar occurrence in calls for stroke patients and calls for other medical conditions, this term cannot be used to discriminate stroke during an emergency call.⁸⁵ However, when the caller describes a fall-related event, it might be helpful in some instances for the dispatcher to seek information about the cause of the fall in order to determine whether the fall is a consequence of a stroke. Some EMS systems (such as the MPDS) prompt the dispatcher to ask the caller clarifying questions to determine whether the fall was mechanical or caused by a medical condition such as stroke.^{86, 87}

It was surprising that the only communication factor considered in the reviewed studies^{88, 89, 90, 91} was the keywords/symptoms used by the caller during the emergency call, and this was also generally the case for excluded studies.^{92, 93, 94, 95, 96, 97} Future studies should examine the contribution of the dispatcher's communication in the identification of stroke, since a recent qualitative study⁹⁸ found dispatcher communication skills have a decisive effect on stroke identification. In addition, it may be informative to investigate a variety of communication factors for both the caller's and the dispatcher's language, such as accent, tone of voice, and use of grammatical structures. For example, a study by Meischke, et al.⁹⁹ examining emergency calls for a variety of medical conditions found limited native language proficiency by callers was a barrier to accurate triage and increased time to dispatch ambulance resources.

Studies in this review used different methods to diagnosis stroke (i.e. paramedic vs. physician diagnosis). Three studies^{100, 101, 102} used a hospital diagnosis of stroke and one study used a paramedic diagnosis of stroke.¹⁰³ Although paramedics do not typically have access to brain imaging technology to detect stroke (unlike hospital physicians), they still achieve high levels of stroke identification using FAST screening tools¹⁰⁴ and show good agreement with hospital physicians in recognition of FAST.¹⁰⁵

LIMITATIONS

Limitation of Reviewed Research

The four studies that considered how communication factors were associated with dispatcher identification of stroke had several limitations. All studies used an observational (mostly retrospective) design, which means that the association between the caller's communication and identification of stroke may be attributable to confounding factors. A randomized controlled trial is preferable for achieving reliable results; however, it is extremely difficult to employ this design in an emergency communication center, and therefore most dispatch research uses the observational design.¹⁰⁶ Studies that collected data from patient records,^{107, 108} rather than a transcription of the emergency call, may be less accurate in their reporting of caller keywords/symptoms. Although listening to audio recordings is time intensive, it is the preferred method in dispatch research.¹⁰⁹ An additional limitation across studies was that a consistent method for grouping caller keywords into symptom categories was not established, and this lack of a consistent metric made it difficult to compare study results. For example, Chenaitia et al.¹¹⁰ combined keywords relating to "vertigo" with keywords relating to "falls," whereas Berglund et al.¹¹¹ combined keywords related to "lying position" with "falls."

A significant limitation of all reviewed studies was that they did not compare calls for stroke patients with calls for other medical conditions (to determine the specificity of terms used by callers). It is important to consider that although caller report of particular keywords/symptoms may assist the dispatcher to identify stroke (due to reasonable sensitivity), they are not necessarily useful for discriminating between a stroke condition and a condition other than stroke (specificity).

Limitations of our Systematic Review

The strict inclusion criteria of our systematic review meant a number of studies were excluded^{112, 113, 114, 115, 116, 117, 118, 119, 120, 121} because they did not include a dispatcher classification of the call as either "stroke" (true positives) or "other than stroke" (false negatives) for patients with a "final" stroke diagnosis. Studies that did not specify dispatcher classification of calls can provide information on what caller keywords/symptoms might be useful to dispatchers. However, our eligibility criteria allowed us to determine caller keywords/symptoms were associated with identification of stroke by dispatchers by comparing true positives with false negatives.

While we cannot guarantee that our systematic review identified all published studies, we are confident that the search strategy was comprehensive.

CONCLUSION

Our systematic review of the literature identified four studies that examined caller use of keywords in relation to dispatcher identification of stroke. There were two main findings: (1) caller mention of a "stroke" or typical stroke symptoms such as speech disturbance, facial weakness, and arm/limb weakness (FAST symptoms) was associated with increased dispatcher identification of stroke; and (2) caller description of the stroke as a fall-related event was associated with decreased detection of stroke. Based on these findings, it might be helpful for dispatchers to be aware of particular keywords that callers use to describe stroke which appear to either help or hinder stroke identification. However, recommendations about how dispatchers might use caller keywords to differentiate stroke from non-stroke conditions cannot be drawn from the reviewed studies because they did not compare calls for stroke patients with calls for other medical conditions. Future studies should compare calls for stroke patients with calls for non-stroke patients in order to provide information about how dispatchers can discriminate stroke from a non-stroke condition. A surprising finding of our systematic review was that the reviewed studies only considered one type of communication factor, the caller's use of keywords. Future research could investigate how various factors in both the caller's and dispatcher's communication (such as accent, tone of voice, and use of grammatical structures) may affect identification of stroke by dispatchers.

REFERENCES

1. Wardlaw J, Murray V, Berge E, del Zoppo G, Sandercock P, Lindley R, et al. Recombinant tissue plasminogen activator for acute ischaemic stroke: an updated systematic review and meta-analysis. J Lancet. 2012;379:2364-72. view
2. Caceres J, Adil M, Jadhav V, Chaudhry A, Pawar S, Rodriguez G, et al. Diagnosis of stroke by emergency medical dispatchers and its impact on prehospital care of patients. J Stroke Cerebrovasc Dis. 2013;22(8):610-4 view
3. Berglund A, Svensson L, Sjostrand C, von Arbin M, von Euler M, Wahlgren N, et al. Higher prehosptial priority level of stroke improves thrombolysis frequency and time to stroke unit: the Hyer Acute STroke Alarm (HASTA) study. Stroke. 2012;43:2666-70.
4. Caceres J, Adil M, Jadhav V, Chaudhry A, Pawar S, Rodriguez G, et al. Diagnosis of stroke by emergency medical dispatchers and its impact on prehospital care of patients. J Stroke Cerebrovasc Dis. 2013;22(8):610-4
5. Rosamond W, Evenson K, Schroeder E, Morris D, Johnson A, Brice J. Calling emergency medical services for acute stroke: a study of 911 tapes. Prehosp Emerg Care. 2005;9:19-23
6. Porteous G, Megan D, Smith W. Emergency medical services dispatcher identification of stroke and transient ischemic attack. Prehosp Emerg Care. 1999;3:211-6
7. Buck B, Starkham S, Eckstein M, Kidwell C, Haines J, Huang R, et al. Dispatcher recognition of stroke using the national academy medical priority dispatch system. Stroke. 2009;40:2027-30.
8. Handschu R, Poppe P, Raub J, Neundorfer B, Erbguth F. Emergency calls in acute stroke. Stroke. 2003;34:1005-9.
9. Krebes S, Ebinger M, Baumann A, Kellner P, Rozanski M, Doepp F, et al. Development and validation of a dispatcher identification algorithm for stroke emergencies. Stroke. 2012;43:776-81.
10. Chenaitia H, Lefevre O, Ho V, Squarcioni C, Pradel V, Fournier M, et al. Emergency medical service in the stroke chain of survival. Eur J Emerg Med. 2013;20(1):39-44.
11. Berglund A, von Euler M, Schenck-Gustafsson K, Castren M, Bohm K. Identification of stroke during the emergency call: a descriptive study of callers' presentation of stroke. BMJ Open. 2015;5(e007661):1-6
12. Buck B, Starkham S, Eckstein M, Kidwell C, Haines J, Huang R, et al. Dispatcher recognition of stroke using the national academy medical priority dispatch system. Stroke. 2009;40:2027-30
13. Liberati A, Altman D, Telzlaff J, Mulrow C, Gotzsche P, Ioannidis J, et al. The PRISMA statement for reporting systematic reviews and meta-analysis of studies that evaluate health care interventions: explanation and elaboration. PLoS Med. 2009;6(7):1-28.
14. Liberati A, Altman D, Telzlaff J, Mulrow C, Gotzsche P, Ioannidis J, et al. The PRISMA statement for reporting systematic reviews and meta-analysis of studies that evaluate health care interventions: explanation and elaboration. PLoS Med. 2009;6(7):1-28
15. Schneiker E, Finn J, Williams W, Ball S. In emergency ambulance calls for suspected stroke, what characteristics of the communication between the caller and the call-taker affect identification of stroke? A systematic review and meta-analysis.: University of York: Centre for Reviews and Dissemination 2015 [cited 2015 28th October]. Available from: http://www.crd.york.ac.uk/PROSPERO/display_record.asp?ID=CRD42015026086 view
16. Liberati A, Altman D, Telzlaff J, Mulrow C, Gotzsche P, Ioannidis J, et al. The PRISMA statement for reporting systematic reviews and meta-analysis of studies that evaluate health care interventions: explanation and elaboration. PLoS Med. 2009;6(7):1-28
17. Wells G, Shea B, O'Connell D, Peterson J, Wlch V, Loso M. The Newcastle-Ottawa Scale (NOS) for assessing the quality of non-randomised studies in meta-analyses 2015 [cited 05.11.2015]. Available from: http://www.ohri.ca/programs/clinical_epidemiology/oxford.htm view
18. Cook D, Reed D. Appraising the quality of medical education research methods: the medical education research study quality instrument and the Newcastle-Ottawa Scale. Acad Med. 2015;90(8):1067-76
19. Rosamond W, Evenson K, Schroeder E, Morris D, Johnson A, Brice J. Calling emergency medical services for acute stroke: a study of 911 tapes. Prehosp Emerg Care. 2005;9:19-23
20. Handschu R, Poppe P, Raub J, Neundorfer B, Erbguth F. Emergency calls in acute stroke. Stroke. 2003;34:1005-9
21. Krebes S, Ebinger M, Baumann A, Kellner P, Rozanski M, Doepp F, et al. Development and validation of a dispatcher identification algorithm for stroke emergencies. Stroke. 2012;43:776-81
22. Berglund A, Svensson L, Wahlgren N, von Euler M. Face Arm Speech Time Test use in prehospital setting, better in ambulance than in the emergency medical communication centre. Cerebrovasc Dis. 2014;37:212-6
23. Leathley M, Jones S, Gibson J, Ford G, McAdam J, Quinn T, et al. "Can you send an ambulance please?": a comparison of callers' requests for emergency medical dispatch in non-stroke and stroke calls. Emerg Med Journal. 2014;31:25-8
24. Jones S, Carter B, Ford G, Gibson J, Leathley M, McAdam J, et al. The identification of acute stroke: an analysis of emergency calls. Int J Stroke. 2013;8:408-12
25. Camerlingo M, Casto L, Censori B, Ferraro B, Gazzaniga G, Partziguian T, et al. Experience with a questionnaire administered by emergency medical service for pre-hosptial identification of patients with acute stroke. Neurol Sci. 2001;22:357-61
26. Malekzadeh J, Shafee H, Behnam H, Mirhaghi A. The effect of Cincinati prehosptial stroke scale on telephone triage of stroke patients: evidence-based practice in emergency medical services. Int J Evid Based Healthc. 2015:87-92
27. Reginella R, Crocco T, Tadros A, Shackleford A, Davis S. Predictors of stroke during 911 calls: opportunites for improving EMS repsonse. Prehosp Emerg Care. 2006;10:369-73
28. Gibson J, Bullock M, Ford G, Jones S, Leathley M, McAdam J, et al. "Is he awake?": dialogues between callers and call handlers about consciousness during emergency calls for suspected acute stroke. Emerg Med J. 2013;30:414-8
29. Porteous G, Megan D, Smith W. Emergency medical services dispatcher identification of stroke and transient ischemic attack. Prehosp Emerg Care. 1999;3:211-6
30. Chenaitia H, Lefevre O, Ho V, Squarcioni C, Pradel V, Fournier M, et al. Emergency medical service in the stroke chain of survival. Eur J Emerg Med. 2013;20(1):39-44
31. Berglund A, von Euler M, Schenck-Gustafsson K, Castren M, Bohm K. Identification of stroke during the emergency call: a descriptive study of callers' presentation of stroke. BMJ Open. 2015;5(e007661):1-6
32. De Luca A, Rossi P, Villa G. 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):1-8
33. Chenaitia H, Lefevre O, Ho V, Squarcioni C, Pradel V, Fournier M, et al. Emergency medical service in the stroke chain of survival. Eur J Emerg Med. 2013;20(1):39-44.
34. Berglund A, von Euler M, Schenck-Gustafsson K, Castren M, Bohm K. Identification of stroke during the emergency call: a descriptive study of callers' presentation of stroke. BMJ Open. 2015;5(e007661):1-6
35. De Luca A, Rossi P, Villa G. 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):1-8
36. De Luca A, Rossi P, Villa G. 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):1-8
37. Porteous G, Megan D, Smith W. Emergency medical services dispatcher identification of stroke and transient ischemic attack. Prehosp Emerg Care. 1999;3:211-6
38. Chenaitia H, Lefevre O, Ho V, Squarcioni C, Pradel V, Fournier M, et al. Emergency medical service in the stroke chain of survival. Eur J Emerg Med. 2013;20(1):39-44
39. Berglund A, von Euler M, Schenck-Gustafsson K, Castren M, Bohm K. Identification of stroke during the emergency call: a descriptive study of callers' presentation of stroke. BMJ Open. 2015;5(e007661):1-6
40. Chenaitia H, Lefevre O, Ho V, Squarcioni C, Pradel V, Fournier M, et al. Emergency medical service in the stroke chain of survival. Eur J Emerg Med. 2013;20(1):39-44.
41. Berglund A, von Euler M, Schenck-Gustafsson K, Castren M, Bohm K. Identification of stroke during the emergency call: a descriptive study of callers' presentation of stroke. BMJ Open. 2015;5(e007661):1-6
42. Chenaitia H, Lefevre O, Ho V, Squarcioni C, Pradel V, Fournier M, et al. Emergency medical service in the stroke chain of survival. Eur J Emerg Med. 2013;20(1):39-44
43. De Luca A, Rossi P, Villa G. 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):1-8
44. Porteous G, Megan D, Smith W. Emergency medical services dispatcher identification of stroke and transient ischemic attack. Prehosp Emerg Care. 1999;3:211-6
45. Berglund A, von Euler M, Schenck-Gustafsson K, Castren M, Bohm K. Identification of stroke during the emergency call: a descriptive study of callers' presentation of stroke. BMJ Open. 2015;5(e007661):1-6
46. Chenaitia H, Lefevre O, Ho V, Squarcioni C, Pradel V, Fournier M, et al. Emergency medical service in the stroke chain of survival. Eur J Emerg Med. 2013;20(1):39-44
47. Berglund A, von Euler M, Schenck-Gustafsson K, Castren M, Bohm K. Identification of stroke during the emergency call: a descriptive study of callers' presentation of stroke. BMJ Open. 2015;5(e007661):1-6
48. Wells G, Shea B, O'Connell D, Peterson J, Wlch V, Loso M. The Newcastle-Ottawa Scale (NOS) for assessing the quality of non-randomised studies in meta-analyses 2015 [cited 05.11.2015]. Available from: http://www.ohri.ca/programs/clinical_epidemiology/oxford.htm
49. Higgins JPT, Green S. Cochrane handbook for systemtic reviews of interventions v 5.1.0 2011 [cited 2016 27th May]. Available from: http://handbook.cochrane.org/chapter_9/9_5_1_what_is_heterogeneity.htm
50. Porteous G, Megan D, Smith W. Emergency medical services dispatcher identification of stroke and transient ischemic attack. Prehosp Emerg Care. 1999;3:211-6
51. Chenaitia H, Lefevre O, Ho V, Squarcioni C, Pradel V, Fournier M, et al. Emergency medical service in the stroke chain of survival. Eur J Emerg Med. 2013;20(1):39-44
52. Berglund A, von Euler M, Schenck-Gustafsson K, Castren M, Bohm K. Identification of stroke during the emergency call: a descriptive study of callers' presentation of stroke. BMJ Open. 2015;5(e007661):1-6.
53. Porteous G, Megan D, Smith W. Emergency medical services dispatcher identification of stroke and transient ischemic attack. Prehosp Emerg Care. 1999;3:211-6
54. Berglund A, von Euler M, Schenck-Gustafsson K, Castren M, Bohm K. Identification of stroke during the emergency call: a descriptive study of callers' presentation of stroke. BMJ Open. 2015;5(e007661):1-6
55. Chenaitia H, Lefevre O, Ho V, Squarcioni C, Pradel V, Fournier M, et al. Emergency medical service in the stroke chain of survival. Eur J Emerg Med. 2013;20(1):39-44
56. Berglund A, von Euler M, Schenck-Gustafsson K, Castren M, Bohm K. Identification of stroke during the emergency call: a descriptive study of callers' presentation of stroke. BMJ Open. 2015;5(e007661):1-6
57. Chenaitia H, Lefevre O, Ho V, Squarcioni C, Pradel V, Fournier M, et al. Emergency medical service in the stroke chain of survival. Eur J Emerg Med. 2013;20(1):39-44
58. Berglund A, von Euler M, Schenck-Gustafsson K, Castren M, Bohm K. Identification of stroke during the emergency call: a descriptive study of callers' presentation of stroke. BMJ Open. 2015;5(e007661):1-6
59. Chenaitia H, Lefevre O, Ho V, Squarcioni C, Pradel V, Fournier M, et al. Emergency medical service in the stroke chain of survival. Eur J Emerg Med. 2013;20(1):39-44
60. Berglund A, von Euler M, Schenck-Gustafsson K, Castren M, Bohm K. Identification of stroke during the emergency call: a descriptive study of callers' presentation of stroke. BMJ Open. 2015;5(e007661):1-6
61. Berglund A, von Euler M, Schenck-Gustafsson K, Castren M, Bohm K. Identification of stroke during the emergency call: a descriptive study of callers' presentation of stroke. BMJ Open. 2015;5(e007661):1-6
62. De Luca A, Rossi P, Villa G. 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):1-8.
63. Chenaitia H, Lefevre O, Ho V, Squarcioni C, Pradel V, Fournier M, et al. Emergency medical service in the stroke chain of survival. Eur J Emerg Med. 2013;20(1):39-44.
64. Berglund A, von Euler M, Schenck-Gustafsson K, Castren M, Bohm K. Identification of stroke during the emergency call: a descriptive study of callers' presentation of stroke. BMJ Open. 2015;5(e007661):1-6
65. Chenaitia H, Lefevre O, Ho V, Squarcioni C, Pradel V, Fournier M, et al. Emergency medical service in the stroke chain of survival. Eur J Emerg Med. 2013;20(1):39-44
66. De Luca A, Rossi P, Villa G. 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):1-8
67. Chenaitia H, Lefevre O, Ho V, Squarcioni C, Pradel V, Fournier M, et al. Emergency medical service in the stroke chain of survival. Eur J Emerg Med. 2013;20(1):39-44
68. Berglund A, von Euler M, Schenck-Gustafsson K, Castren M, Bohm K. Identification of stroke during the emergency call: a descriptive study of callers' presentation of stroke. BMJ Open. 2015;5(e007661):1-6
69. Chenaitia H, Lefevre O, Ho V, Squarcioni C, Pradel V, Fournier M, et al. Emergency medical service in the stroke chain of survival. Eur J Emerg Med. 2013;20(1):39-44.
70. Berglund A, von Euler M, Schenck-Gustafsson K, Castren M, Bohm K. Identification of stroke during the emergency call: a descriptive study of callers' presentation of stroke. BMJ Open. 2015;5(e007661):1-6
71. Reginella R, Crocco T, Tadros A, Shackleford A, Davis S. Predictors of stroke during 911 calls: opportunites for improving EMS repsonse. Prehosp Emerg Care. 2006;10:369-73
72. Krebes S, Ebinger M, Baumann A, Kellner P, Rozanski M, Doepp F, et al. Development and validation of a dispatcher identification algorithm for stroke emergencies. Stroke. 2012;43:776-81
73. Porteous G, Megan D, Smith W. Emergency medical services dispatcher identification of stroke and transient ischemic attack. Prehosp Emerg Care. 1999;3:211-6
74. Berglund A, von Euler M, Schenck-Gustafsson K, Castren M, Bohm K. Identification of stroke during the emergency call: a descriptive study of callers' presentation of stroke. BMJ Open. 2015;5(e007661):1-6
75. Handschu R, Poppe P, Raub J, Neundorfer B, Erbguth F. Emergency calls in acute stroke. Stroke. 2003;34:1005-9
76. Rosamond W, Evenson K, Schroeder E, Morris D, Johnson A, Brice J. Calling emergency medical services for acute stroke: a study of 911 tapes. Prehosp Emerg Care. 2005;9:19-23
77. Reginella R, Crocco T, Tadros A, Shackleford A, Davis S. Predictors of stroke during 911 calls: opportunites for improving EMS repsonse. Prehosp Emerg Care. 2006;10:369-73
78. Krebes S, Ebinger M, Baumann A, Kellner P, Rozanski M, Doepp F, et al. Development and validation of a dispatcher identification algorithm for stroke emergencies. Stroke. 2012;43:776-81
79. Chenaitia H, Lefevre O, Ho V, Squarcioni C, Pradel V, Fournier M, et al. Emergency medical service in the stroke chain of survival. Eur J Emerg Med. 2013;20(1):39-44
80. Berglund A, von Euler M, Schenck-Gustafsson K, Castren M, Bohm K. Identification of stroke during the emergency call: a descriptive study of callers' presentation of stroke. BMJ Open. 2015;5(e007661):1-6
81. Deakin C, Alasaad M, King P, Thomspon F. Is ambulance telephone triage using advanced medical priority dispatch protocols able to identify patients with acute stroke correctly? Emerg Med J. 2009;26:442-5
82. Berglund A, von Euler M, Schenck-Gustafsson K, Castren M, Bohm K. Identification of stroke during the emergency call: a descriptive study of callers' presentation of stroke. BMJ Open. 2015;5(e007661):1-6
83. Krebes S, Ebinger M, Baumann A, Kellner P, Rozanski M, Doepp F, et al. Development and validation of a dispatcher identification algorithm for stroke emergencies. Stroke. 2012;43:776-81
84. Leathley M, Jones S, Gibson J, Ford G, McAdam J, Quinn T, et al. "Can you send an ambulance please?": a comparison of callers' requests for emergency medical dispatch in non-stroke and stroke calls. Emerg Med Journal. 2014;31:25-8.
85. Krebes S, Ebinger M, Baumann A, Kellner P, Rozanski M, Doepp F, et al. Development and validation of a dispatcher identification algorithm for stroke emergencies. Stroke. 2012;43:776-81
86. Leathley M, Jones S, Gibson J, Ford G, McAdam J, Quinn T, et al. "Can you send an ambulance please?": a comparison of callers' requests for emergency medical dispatch in non-stroke and stroke calls. Emerg Med Journal. 2014;31:25-8
87. Deakin C, Alasaad M, King P, Thomspon F. Is ambulance telephone triage using advanced medical priority dispatch protocols able to identify patients with acute stroke correctly? Emerg Med J. 2009;26:442-5
88. Porteous G, Megan D, Smith W. Emergency medical services dispatcher identification of stroke and transient ischemic attack. Prehosp Emerg Care. 1999;3:211-6
89. Chenaitia H, Lefevre O, Ho V, Squarcioni C, Pradel V, Fournier M, et al. Emergency medical service in the stroke chain of survival. Eur J Emerg Med. 2013;20(1):39-44
90. Berglund A, von Euler M, Schenck-Gustafsson K, Castren M, Bohm K. Identification of stroke during the emergency call: a descriptive study of callers' presentation of stroke. BMJ Open. 2015;5(e007661):1-6
91. De Luca A, Rossi P, Villa G. 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):1-8
92. Rosamond W, Evenson K, Schroeder E, Morris D, Johnson A, Brice J. Calling emergency medical services for acute stroke: a study of 911 tapes. Prehosp Emerg Care. 2005;9:19-23
93. Handschu R, Poppe P, Raub J, Neundorfer B, Erbguth F. Emergency calls in acute stroke. Stroke. 2003;34:1005-9
94. Berglund A, Svensson L, Wahlgren N, von Euler M. Face Arm Speech Time Test use in prehospital setting, better in ambulance than in the emergency medical communication centre. Cerebrovasc Dis. 2014;37:212-6
95. Jones S, Carter B, Ford G, Gibson J, Leathley M, McAdam J, et al. The identification of acute stroke: an analysis of emergency calls. Int J Stroke. 2013;8:408-12
96. Malekzadeh J, Shafee H, Behnam H, Mirhaghi A. The effect of Cincinati prehosptial stroke scale on telephone triage of stroke patients: evidence-based practice in emergency medical services. Int J Evid Based Healthc. 2015:87-92
97. Reginella R, Crocco T, Tadros A, Shackleford A, Davis S. Predictors of stroke during 911 calls: opportunites for improving EMS repsonse. Prehosp Emerg Care. 2006;10:369-73
98. Berglund A, Heikkila K, Bohm K, Schenck-Gustafsson K, von Euler M. Factors facilitating or hampering nurses identification of stroke in emergency calls. J Adv Nurs. 2015;71(11):2609-21.
99. Meischke H, Calhoun R, Yip M, Tu S, Painter I. The effect of language barriers on dispatching EMS response. Prehosp Emerg Care. 2012;17:475-80
100. Porteous G, Megan D, Smith W. Emergency medical services dispatcher identification of stroke and transient ischemic attack. Prehosp Emerg Care. 1999;3:211-6
101. Chenaitia H, Lefevre O, Ho V, Squarcioni C, Pradel V, Fournier M, et al. Emergency medical service in the stroke chain of survival. Eur J Emerg Med. 2013;20(1):39-44
102. Berglund A, von Euler M, Schenck-Gustafsson K, Castren M, Bohm K. Identification of stroke during the emergency call: a descriptive study of callers' presentation of stroke. BMJ Open. 2015;5(e007661):1-6
103. De Luca A, Rossi P, Villa G. 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):1-8
104. Harbison J, Hossain O, Jenkinson D, Davis J, Louw S, Ord G. Diagnostic accuracy of stroke referrals from primary care, emergency room physicians, and ambulance staff using the Face Arm Speech Test. Stroke. 2003;34:71-6
105. Nor A, McAllister C, Louw S, Dyker A, Davis M, Jenkinson D, et al. Agreement between ambulance paramedic and physician-recorded neurological signs with Face Arm Speech Test (FAST) in acute stroke patients. Stroke. 2004;35:1355-9
106. Gardett I, Clawson J, Scott G, Barron T, Patterson B, Oola C. Past, present, and future of emergency dispatch research: a systematic literature review. Ann Emerg Disp Resp. 2013;1(2):29-42
107. Chenaitia H, Lefevre O, Ho V, Squarcioni C, Pradel V, Fournier M, et al. Emergency medical service in the stroke chain of survival. Eur J Emerg Med. 2013;20(1):39-44
108. De Luca A, Rossi P, Villa G. 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):1-8
109. Lyon R, Bohm K, Frischknecht Christensen E, Olaveengen T, Castren M. The inaugral European mergency medical dispatch conference – a synopsis of proceedings. Scan J Trauma Resus Emerg Med. 2013;21(73):1-4
110. Chenaitia H, Lefevre O, Ho V, Squarcioni C, Pradel V, Fournier M, et al. Emergency medical service in the stroke chain of survival. Eur J Emerg Med. 2013;20(1):39-44
111. Berglund A, von Euler M, Schenck-Gustafsson K, Castren M, Bohm K. Identification of stroke during the emergency call: a descriptive study of callers' presentation of stroke. BMJ Open. 2015;5(e007661):1-6
112. Rosamond W, Evenson K, Schroeder E, Morris D, Johnson A, Brice J. Calling emergency medical services for acute stroke: a study of 911 tapes. Prehosp Emerg Care. 2005;9:19-23
113. Handschu R, Poppe P, Raub J, Neundorfer B, Erbguth F. Emergency calls in acute stroke. Stroke. 2003;34:1005-9
114. Krebes S, Ebinger M, Baumann A, Kellner P, Rozanski M, Doepp F, et al. Development and validation of a dispatcher identification algorithm for stroke emergencies. Stroke. 2012;43:776-81
115. Berglund A, Svensson L, Wahlgren N, von Euler M. Face Arm Speech Time Test use in prehospital setting, better in ambulance than in the emergency medical communication centre. Cerebrovasc Dis. 2014;37:212-6
116. Leathley M, Jones S, Gibson J, Ford G, McAdam J, Quinn T, et al. "Can you send an ambulance please?": a comparison of callers' requests for emergency medical dispatch in non-stroke and stroke calls. Emerg Med Journal. 2014;31:25-8
117. Jones S, Carter B, Ford G, Gibson J, Leathley M, McAdam J, et al. The identification of acute stroke: an analysis of emergency calls. Int J Stroke. 2013;8:408-12
118. Camerlingo M, Casto L, Censori B, Ferraro B, Gazzaniga G, Partziguian T, et al. Experience with a questionnaire administered by emergency medical service for pre-hosptial identification of patients with acute stroke. Neurol Sci. 2001;22:357-61
119. Malekzadeh J, Shafee H, Behnam H, Mirhaghi A. The effect of Cincinati prehosptial stroke scale on telephone triage of stroke patients: evidence-based practice in emergency medical services. Int J Evid Based Healthc. 2015:87-92
120. Reginella R, Crocco T, Tadros A, Shackleford A, Davis S. Predictors of stroke during 911 calls: opportunites for improving EMS repsonse. Prehosp Emerg Care. 2006;10:369-73
121. Gibson J, Bullock M, Ford G, Jones S, Leathley M, McAdam J, et al. "Is he awake?": dialogues between callers and call handlers about consciousness during emergency calls for suspected acute stroke. Emerg Med J. 2013;30:414-8