Elsevier

Resuscitation

Volume 146, 1 January 2020, Pages 5-12
Resuscitation

Simulation and education
Accuracy of automatic geolocalization of smartphone location during emergency calls — A pilot study

https://doi.org/10.1016/j.resuscitation.2019.10.030Get rights and content

Abstract

Introduction

Widespread use of smartphones allows automatic geolocalization (i.e., transmission of location data) in countless apps. Until now, this technology has not been routinely used in connection with an emergency call in which location data play a decisive role This study evaluated a new software automatically providing emergency medical service (EMS) dispatchers with a caller's geolocation. We hypothesized that this technology will provide higher accuracy, faster dispatching of EMS and a faster beginning of thoracic compressions in a cardiac arrest scenario.

Material and methods

Approval from the local Ethics Committee was obtained. 108 simulated emergency calls reporting a patient in cardiac arrest were conducted at 54 metropolitan locations, which were chosen according to a realistic pattern. At each location, a conventional emergency call, with an oral description of the location, was given first; subsequently, another call using an app with automatic geolocation was placed. Accuracy of localization, time to location, time to EMS dispatch and time to first thoracic compression were compared between both groups.

Results

The conventional emergency call was always successful (n = 54). Emergency call via app worked successfully in n = 46 cases (85.2%). Automatic geolocation was provided to EMS in all these n = 46 cases (100%). Deviation from estimated position to actual position was 1173.5 ± 4343.1 m for conventional and 65.6 ± 320.5 m for automatic geolocalization (p < 0.001). In addition, time to localization was significantly shorter using automatic geolocalization (34.7 vs. 71.7 s, p < 0.001). Time to first thoracic compression was significantly faster in the geolocalization group (83.0 vs. 122.6 s; p < 0.001).

Conclusions

This pilot study showed that automatic geolocalization leads to a significantly shorter duration of the emergency call, significantly shorter times until the beginning of thoracic compressions, and a higher precision in determining the location of an emergency.

Introduction

Emergency situations are challenging for each involved person. Bystanders who witness an emergency are often overcome by stress, concern and uncertainty.1 All following actions are affected by these emotions, and distinct rational decision-making gets blurred. This also compromises the efficacy of an emergency call, where the caller seeks and urges immediate help, whereas the emergency medicals service (EMS) dispatcher needs clear and accurate facts to make correct decisions as to where and what kind of help to dispatch.

Amongst all emergencies, sudden out-of-hospital cardiac arrest (OHCA) is one of the most extreme and also most urgent emergencies, as it leads to death within minutes if not immediately treated adequately.2 Unfortunately, it is also amongst the leading causes for death in Central Europe, the United States and all other developed parts of the world.3, 4 The annual incidence of patients suffering an out-of-hospital cardiac arrest varies between 40 and 90 of 100,000 inhabitants.5 Only 5–14% of the patients can be discharged after resuscitation from out-of-hospital cardiac arrest.6 Despite intense research, the overall outcome of OHCA patients remains poor.5

One of the main keys for survival is early and high-quality cardiopulmonary resuscitation (CPR) conducted by bystanders and emergency witnesses.7 Without early CPR through bystanders (which are often medical laypersons), any further procedure and pharmacological treatment by healthcare experts (“Advanced Life Support”, ALS) has little to no prospect of success.

In Europe, it typically takes at least 5−8 min for the EMS to arrive in urban regions – and often considerably longer in rural areas.8 Brain tissue, however, is extremely sensitive to hypoxemia, and irrecoverable brain damage starts 2−3 min after the beginning of cardiac arrest.9 Hence, every second it takes EMS to reach the site of the incident aggravates neurological impairment or even leads to death of the victim.2 Technical advances such as area-wide broad-bandwidth wireless internet coverage and the ubiquitous utilization of smartphones have opened up new possibilities which surpass those of normal audio-only telephony by far. Transmission of data of all kind from the caller’s smartphone is possible, and is used in many situations of life, e.g., sharing one’s live position data in an online map with friends. Paradoxically, although being used by most smartphone users throughout the day and data being shared with many persons and even companies, this technology has not been deployed in communication with authorities and emergency services yet.

Within the EU-funded RAMSES (Remote Access to Medical Information on Smartphones during Emergencies and Health CriseS) project, new software has therefore been developed. Amongst other very relavant features, it facilitates automatized geolocalization by submitting global positioning system (aGPS) data from the caller’s smartphone to the dispatch center.

So far, this technology has not been tested in a controlled trail yet, and scarce data exists on whether automatic geolocalization (using aGPS data) is feasible during an emergency call.

The aim of this study was to evaluate time and precision of automatic geolocalization in comparison to a conventional emergency call, and to evaluate if this technology leads to a shortening of the emergency call itself. We hypothesized that using this technology would lead to higher accuracy of the localization, faster dispatching of EMS and a faster beginning of thoracic compressions in a cardiac arrest scenario.

Section snippets

Material

We chose a model of simulated cardiac arrest for evaluation of this new software. Within the metropolitan area of Cologne, a city with 1,000,000 inhabitants and the fourth-biggest city of Germany, n = 54 locations were randomly selected by the principal investigator (Fig. 1). In order to achieve a pattern that closely represents the realistic distribution of emergency sites,10 places were chosen that include residential districts (n = 32), public places (i.e. stations, airport, sport stadiums

Methods

First, our study team moved to the predefined study location, for which the dispatchers were blinded at all times. On arrival, the exact geolocation of the emergency site (where the resuscitation manikin was placed) was measured and recorded as reference, using a common GPS device (Garmin, Schaffhausen, CH).

Following this, a conventional emergency call was placed by the study team using the telephone function of the study phone. As soon as a connection was established, the first question by the

Ethics committee approval

The Ethics Committee of the University of Cologne (Head: Prof. Drzezga) approved the study on 2018-03-27 (ID 18-043).

Study registration

The study was registered at ClinicalTrials.gov (Identifier NCT03654846).

Statistical analysis

Due to the novelty of the technology and the lack of data from comparable studies, a sample size calculation was not possible. A sample size of n = 54 was considered to be adequate to reach a big (0.8) effect size (standardized difference) of the primary end points “time until localization of the caller” and “time until first chest compression” with at least 80% power and a two-sided significance level of 5% (paired t-test).

Data were described using mean ± standard deviation (SD) or count and

Results

Placing a conventional emergency call was possible in all n = 54 cases (100%), placing an emergency call with the app was successful in n = 46 of 54 cases (85.2%). Automatic geolocalization always was possible when app calls could be established (46/46 cases, 100%, Table 1).

Reasons for malfunction of the app included “waiting for connection” failure (n = 5), “EmergencyEye halted” (n = 2), and “network connection lost” (n = 1). In all cases, sufficient mobile internet with sufficient data

Discussion

In this pilot study, we compared automatic geolocalization with conventional emergency call communication concerning precision of the suspected emergency site, time until localization and EMS dispatching and time until first delivered chest compression in a simulated cardiac arrest under realistic conditions. This pilot study showed that automatic geolocalization leads to a significantly shorter duration of localizing the caller and EMS deployment, significantly shorter times until the

Limitation

This study has several limitations. First, it was performed with an early beta version of the app (version 0.5.1) and server backbone structure, which reflects and was responsible for the high amount (n = 8) of attempts the software failed in. Further studies with appropriate testing have to be performed with a more stable enhanced version. Second, automatic geolocalization results were not validated by the dispatcher asking the caller to verify the suspected location; doing this might have

Conclusions

In this pilot study, we were able to show that automatic geolocalization of a caller’s smartphone significantly increases accuracy of the assumed position, reduces the time until localization of the caller and EMS dispatch, and reduces time until the first chest compression in a simulated cardiac arrest scenario.

Conflict of interest and acknowledgements

This study was funded by a grant of the European Institute of Innovation & Technology (EIT Health; Project: 18234), a body of the European Union.

Bernd W. Böttiger is European Resuscitation Council (ERC) Board Director Science and Research; Chairman of the German Resuscitation Council (GRC); Member of the Advanced Life Support (ALS) Task Force of the International Liaison Committee on Resuscitation (ILCOR); Member of the executive committee of the German Interdisciplinary Association for

References (25)

  • C.W. Callaway

    Cardiac resuscitation in 2015: improving outcomes after OHCA—targeting the layperson

    Nat Rev Cardiol

    (2016)
  • A. Bürger et al.

    The effect of ambulance response time on survival following out-of-hospital cardiac arrest

    Dtsch Arztebl Int

    (2018)
  • Cited by (8)

    • Responding to Cardiac Arrest in the Community in the Digital Age

      2022, Canadian Journal of Cardiology
      Citation Excerpt :

      Location data can be sent to EMS dispatchers, which could lead to faster location of the SCA victim, for the earlier arrival of lay or professional rescue personnel. Ecker et al. tested the use of automatic geolocalization using a smartphone app compared with a conventional emergency call with oral description of the location, for simulated EMS calls reporting a person suffering SCA.46 Of 108 simulated SCA calls, using the smartphone app to call EMS resulted in significantly faster geolocation of the SCA victim (37 ± 36.6 seconds faster; P < 0.001), faster dispatch of EMS (41.2 ± 38.2 seconds earlier; P < 0.001), and faster initiation of CPR (39.6 ± 38.8 seconds earlier; P < 0.001) compared with a conventional call.

    • Association between hourly call volume in the emergency medical dispatch center and dispatcher-assisted cardiopulmonary resuscitation instruction time in out-of-hospital cardiac arrest

      2020, Resuscitation
      Citation Excerpt :

      There are various types of strategies and interventions reported as solutions to emergency department crowding, including public education campaigns and in-hospital interventions such as smoothing elective admissions and allowing early discharges.16,17 Strategies such as reducing nonemergent calls to EMD through a public campaign, increasing flexibility in staffing schedules according to crowding and implementing the latest technology in EMD should be considered to decrease crowding and its effects.18–20 Generally, the neurologic outcome in OHCA research is recognized as the most important.

    View all citing articles on Scopus
    View full text