Edited by: Hanno L. Tan, Amsterdam University Medical Center, Netherlands
Reviewed by: Albert Ariza Solé, Institut d’Investigació Biomèdica de Bellvitge (IDIBELL), Spain; Marina Cerrone, New York University, United States
This article was submitted to Cardiac Rhythmology, a section of the journal Frontiers in Cardiovascular Medicine
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
The primary aim was to investigate the association between alarm acceptance compared to no-acceptance by volunteer responders, bystander intervention, and survival in out-of-hospital cardiac arrest.
This retrospective observational study included all suspected out-of-hospital cardiac arrests (OHCAs) with activation of volunteer responders in the Capital Region of Denmark (1 November 2018 to 14 May 2019), the Central Denmark Region (1 November 2018 to 31 December 2020), and the Northern Denmark Region (14 February 2020 to 31 December 2020). All OHCAs unwitnessed by Emergency Medical Services (EMS) were analyzed on the basis on alarm acceptance and arrival before EMS. The primary outcomes were bystander cardio-pulmonary resuscitation (CPR), bystander defibrillation and secondary outcome was 30-day survival. A questionnaire sent to all volunteer responders was used with respect to their arrival status.
We identified 1,877 OHCAs with volunteer responder activation eligible for inclusion and 1,725 (91.9%) of these had at least one volunteer responder accepting the alarm (accepted). Of these, 1,355 (79%) reported arrival status whereof 883 (65%) arrived before EMS. When volunteer responders accepted the alarm and arrived before EMS, we found increased proportions and adjusted odds ratio for bystander CPR {94 vs. 83%, 4.31 [95% CI (2.43–7.67)] and bystander defibrillation [13 vs. 9%, 3.16 (1.60–6.25)]} compared to cases where no volunteer responders accepted the alarm.
We observed a fourfold increased odds ratio for bystander CPR and a threefold increased odds ratio for bystander defibrillation when volunteer responders accepted the alarm and arrived before EMS.
香京julia种子在线播放
During the last decade, a strategy of activating volunteer responders to increase bystander cardiopulmonary resuscitation (CPR) and early defibrillation with automated external defibrillators (AEDs) has been implemented world-wide with positive results and increasing interest from both the general public and professionals (
This observational study with prospective data collection included OHCAs with activation of volunteer responders in the Capital Region of Denmark (1 November 2018 to 14 May 2019), the Central Denmark Region (1 November 2018 to 31 December 2020), and the Northern Denmark Region (14 February 2020 to 31 December 2020). Due to an ongoing randomized controlled trial, data collection for the Capital Region does not include patients after 14 May 2019. All three included regions consist of both urban, suburban, and rural areas and covers approximately 23,554 km2 (≈55% of Denmark) and inhabits 3.75 million people (≈64% of the total population).
All included regions are served by a two-tiered EMS system consisting of an ambulance and a physician-staffed unit which are dispatched in case of suspected OHCA. The three regions have separate and independent dispatch centers but follow the same standardized protocol in the event of suspected OHCA. In addition, emergency dispatchers perform telephone-guided CPR and assistance with information on accessing the nearest available AED. In Denmark, a national AED registry was established in 2007 and now contains >21,000 publicly available registered AEDs. The AED registry is linked to all dispatch centers with information on opening hours, accessibility, and global position system location. All EMS dispatchers in Denmark can activate volunteer responders in case of suspected OHCA to assist with CPR and acquisition of a nearby AED. Volunteer responders can be activated simultaneously with the ambulance but usually the activation occurs 30−60 s later as the dispatcher needs to make sure that the surroundings are safe for volunteer responders to attend (
The program was first implemented in the Capital Region of Denmark in September 2017 with other regions gradually following resulting in full national coverage by May 2020. The program is based on volunteers willing to assist in case of OHCA. The purpose of the program is to improve bystander intervention prior to the arrival of EMS to ultimately increase the chances of survival (
All presumed OHCAs, assessed by emergency dispatchers, with volunteer responder activation from the Central, the Capital and North regions of Denmark were identified. Confirmed OHCA was defined as OHCA registered in the Danish Cardiac Arrest Registry, thus we excluded cases (non-OHCAs) not found in the Danish Cardiac Arrest Registry. Further, we excluded OHCAs witnessed by EMS and OHCAs with volunteer responder activation but with no one within range (<1,800 m) of the OHCA. The study population was divided into two groups: one group where at least one volunteer responder accepted the alarm (referred to as “accepted”) and one group where no volunteer responder accepted the alarm (referred to as “not-accepted”). The “not-accepted” group thus includes both rejected and unseen/unanswered alarms. Further, the primary analysis was in the group where volunteer responders reported arriving before EMS in the subsequent questionnaire sent to them. Secondary analysis included OHCAs where at least one volunteer responder accepted the alarm and arrived at the site of OHCA (irrespective of EMS arrival) and OHCAs where at least one accepted the alarm (irrespective of their reported arrival status) both compared to OHCA where no volunteer responders accepted the alarm. This was done in order to compare our data with different volunteer responder programs where arrival status of the volunteer responders is unavailable for scientific reporting. Finally, as Supplementary data we provided a comparison of patients according to initial shockable rhythm.
Variables related to the OHCA originate from the Danish Cardiac Arrest Registry which includes time and date of OHCA, latitude and longitude of OHCA, home or public location, age, sex, witnessed status, initial shockable heart rhythm [ventricular fibrillation (VF) or pulseless ventricular tachycardia (pVT)], bystander CPR, bystander defibrillation, EMS response time, EMS defibrillation, ROSC, and 30-day survival. Information regarding volunteer responders originated from the Volunteer Responder Application Server (local register) and includes geographical locations, app-interactions when accepting or declining alarms, sex, and age. Population density estimates at the OHCA site were based on the municipal population density and were stratified according to the EUROSTAT degree of urbanization system (DEGURBA) producing a three-layered population density stratification (low, intermediate, and high) (
The primary outcomes for this study were bystander CPR, bystander defibrillation, and secondary outcome was 30-day survival.
Categorical variables were presented as proportions and percentages and continuous variables were presented as medians with interquartile range (IQR). A logistic regression analysis was performed to investigate association between exposure (volunteer responders accepting the alarm and their arrival status) and primary outcome variables (bystander CPR, defibrillation, and 30-day survival). Furthermore, we performed a multivariable logistic regression analysis to adjust for identified confounders. We used Direct Acyclic Graphs to determine potential confounders affecting both the exposure and outcomes,
Data were obtained and stored according to the Danish Data Protection Agency (P-2021-670 and P-2021-82). According to Danish Law register studies do not require ethical approval. The study was approved by the Danish Safety Authority (3-3013-2721/1). At registration, volunteer responders give permission to be contacted by the research team if necessary. Volunteer responders consent not to disclosure any private information in relation to OHCA alarms and resuscitation attempts. Volunteer responders can withdraw from the program at any time and simultaneously withdraw their consent.
We initially identified 3,142 presumed OHCAs, assessed by emergency dispatchers, with volunteer responder activation within the study period. Of these, 1,082 were verified non-OHCA, but presumably other genesis, and 24 were witnessed by ambulance staff and thus excluded from further analysis. This resulted in a study population of 1,877 OHCAs with volunteer responder activation,
Flowchart of included patients.
We found no difference in baseline characteristics such as sex and age and likewise, no difference was found in initial shockable heart rhythm (VF or pVT) and proportion of bystander witnessed arrests between the accepted and not-accepted groups,
Characteristics of population with out-of-hospital cardiac arrest and volunteer responder activation based on accept or no accept of alarm.
| Accepted ( |
Not-accepted ( |
Missing ( |
|
| Age, years (IQR) | 73 (63−81) | 73 (63−80) | 47 |
| Male sex, |
1121 (66.7%) | 100 (67.1%) | 47 |
| Witnessed arrest, |
905 (52.6%) | 81 (53.3%) | 3 |
| Initial shockable rhythm (VF/pVT), |
486 (28.3%) | 36 (23.7%) | 9 |
| Median EMS response time, min. (IQR) | 7.00 (5.00−10.00) | 9.00 (6.00−12.00) |
53 |
| Distance between volunteer responders and OHCA, m (IQR) | 527 (298−855) | 720 (384−1422) | 0 |
| Volunteer responder answered question about arrival at scene, |
1,392 (80) | 0 | 485 |
| Reported arriving at OHCA site |
1388 (99.7%) | 0 | 0 |
| Volunteer responders answered question about arrival prior to EMS, |
1,355 (79) | 0 | 522 |
| Arrival prior to EMS |
883 (65.3%) | 0 | 0 |
| Public OHCA location, |
312 (18.1%) | 17 (11.2%) | 1 |
| Population density at OHCA site, |
1 | ||
| 501 (29.1%) | 74 (48.7%) | ||
| 556 (32.3%) | 56 (36.8%) | ||
| 667 (38.7%) | 22 (14.5%) | ||
| Weekend, |
474 (27.5%) | 43 (27.6%) | 0 |
| Median number of activated volunteer responders based on population density, |
0 | ||
| 12 (5−20) | 1 (1−3) | ||
| 20 (10−20) | 6 (2−13.5) | ||
| 20 (20−20) | 15 (10−20) | ||
| Time of the day, |
0 | ||
| 828 (48.0%) | 54 (35.5%) | ||
| 602 (34.9%) | 37 (24.3%) | ||
| 295 (17.1%) | 61 (40.1%) | ||
| EMS defibrillation, |
498 (28.9%) | 47 (30.9%) | 0 |
| Bystander CPR, |
1543 (89.6%) | 126 (82.9%) | 3 |
| Volunteer responders answering question about type of CPR performed, |
606 (35%) | 0 | 1,119 |
| Reported performing compressions and ventilation, |
213 (35) | 0 | |
| Reported performing compressions alone, |
362 (60) | 0 | |
| Reported performing ventilations alone, |
31 (5) | 0 | |
| Bystander defibrillation, |
227 (13.2%) | 10 (6.6%) | 1 |
| ROSC at hospital arrival, |
471 (27.4%) | 30 (19.9%) | 7 |
| 30-day survival, |
253 (15.1%) | 14 (9.4%) | 50 |
OHCA, out-of-hospital cardiac arrest; VF/pVT, ventricular fibrillation/pulseless ventricular tachycardia; EMS, emergency medical services; CPR, cardiopulmonary resuscitation; ROSC, return of spontaneous circulation. Values are median (Q1, Q3), n or n (%). *At least one volunteer responder arrives at site. **At least one volunteer responder arrives at site prior to EMS.
A longer median EMS response time (9.00 vs. 7.00 min) and longer distance from volunteer responder to OHCA site (720 vs. 527 m) were found in the not-accepted group. A larger proportion of OHCAs in the accept group occurred in public locations (18.1 vs. 11.2%) and in areas of high population density (38.7 vs. 14.5%) compared to the not-accepted group. More OHCAs where at least one volunteer responder accepted the alarm occurred during working hours (8.00 a.m.−03.59 p.m.) with no difference during evening (04.00 p.m.−11.59 p.m.) and fewer during night-time (00.00−07.59 a.m.). Still, we found more than 4 times as many incidents of accepted as not-accepted alarms during nighttime (295 vs. 61 incidents).
In our primary analysis we observed that significant more received bystander CPR [94 vs. 83%, odds ratio (OR) 3.37 95% Confidence Interval (95% CI) (2.02−5.60)] and bystander defibrillation [13 vs. 9%, 3.19 (1.64−6.19)] when a volunteer responder accepted the alarm and arrived before EMS compared to OHCAs where no volunteer responders accepted the alarm, respectively,
Odds ratio and adjusted odds ratio of association between alarm acceptance and bystander cardiopulmonary, bystander defibrillation and 30-day survival where at least one volunteer responder accepted the alarm and arrived before the ambulance compared to cases where no volunteer responders accepted the alarm.
Patients presenting with and initial shockable rhythm had higher survival rates (39%) when at least one volunteer responder accepted the alarm compared to cases where no one accepted the alarm (29%),
This prospective observational study investigated whether alarm acceptance by volunteer responders in the case of OHCA was associated with increased bystander intervention and improved patient outcome and the circumstances related to the OHCA. In 9 out of 10 OHCAs we found that at least one volunteer responder accepted the alarm, with fewest alarms accepted (29.1%) in rural areas and during nighttime (17.1%) compared to urban areas (38.7%) and during daytime (48%). The adjusted odds ratio of bystander CPR and defibrillation increased fourfold and threefold, respectively, when a volunteer responder accepted the alarm and arrived before EMS compared with cases where no volunteer responders accepted the alarm. Similar, we found increased odds ratios of bystander CPR and defibrillation when volunteer responders reported arriving at scene of OHCA. Finally, when looking at all OHCAs irrespective of the volunteer responder’s arrival status we still observed increased odds ratios of bystander CPR and defibrillation when at least one volunteer responder accepted the alarm compared to OHCAs where no one accepted the alarm.
Previous studies have shown that volunteer responder programs hold the potential to increase bystander CPR and defibrillation (
A recent Dutch study by Stieglis et al. from 2020 found that 17% of all OHCAs with initial shockable rhythm were defibrillated by volunteer responders (
Unfortunately, we were not able to differentiate between whether CPR and defibrillation was performed by random bystanders at site or alerted volunteer responders. As public arrests were more frequent in the accept group this may have contributed to the higher occurrence of bystander intervention, as public location of arrest in itself is associated with bystander intervention as these arrests are more likely to be witnessed (
An OHCA alarm was more likely to be accepted in areas of high population density. This is most likely due to availability as more volunteer responders were activated in areas of high population density,
The fact that we find a longer EMS response time in the not-accepted group could also be a result of rurality of the OHCA location as OHCAs in rural areas also were more frequent in this group. A volunteer responder study by Andelius et al. from 2020 found an association between longer EMS response time and increased bystander interventions (
We also observed significant diurnal variations in alarm acceptance with significantly higher proportions of alarms being accepted during daytime/evening and higher proportions of not-accepted alarms during nighttime (
This study found a difference in 30-day survival (15.1% when at least one volunteer responder accepted the alarm vs. 9.4% where none accepted the alarm) which was statistically insignificant after adjusting for confounders [1.61 CI (0.90–2.86)]. Further, when looking only at patients presenting with an initial shockable rhythm survival increased non-significantly to 39% when at least one volunteer responder accepted the alarm compared to cases where no one accepted the alarm (29%,
The UK GoodSAM system also demonstrated a difference in 30-day survival between the alarmed and not-alarmed group [London; 17.6 vs. 10.3%, 3.15 95% CI (1.19–8.36)] but interestingly, also found a difference between the groups of alarm accepted and not-accepted [3.06 95% CI (1.0.–9.03)] (
Currently, available studies demonstrating differences in 30-day survival are all observational studies and presenting small absolute numbers of survivors. This increases the risk of confounding and misinterpretation. Furthermore, most available studies compare volunteer responder activation with no activation which is problematic as it further increases risk of inducing both bias and confounding to the analysis as several factors related to the circumstances of the OHCA differ. To fully understand the effect of volunteer responder systems, randomized controlled trials are warranted and currently being conducted in the US/Canada (PulsePoint Study; NCT04806958) and Denmark (HeartRunner Trial; NCT03835403) (
As demonstrated by the findings in this study, data reporting and selection of variables in volunteer responder programs have a big impact on results and the interpretation hereof. We found a clear tendency toward higher odds for bystander CPR and defibrillation in the cases where volunteer responders arrived at site and further with arrival before the EMS compared to only reporting data with respect to whether the volunteer responders accepted the alarm or not. This demonstrates why it is difficult to compare studies with different exposure and outcome variables (
Within the variables “arrival at site” and “arrival prior to EMS” we saw a large number of missing values as some volunteer responders have not completed the survey which should be taken into consideration. This study is limited as it is an observational study why we can only investigate associations and not causal effects. The EUROSAT degree of urbanization system (DEGURBA) (
We observed a fourfold increased odds ratio for bystander CPR and threefold increased odds ratio for bystander defibrillation in OHCAs where volunteer responders accepted the alarm and arrived before EMS compared to cases where no volunteer responders accepted the alarm. We saw no difference in 30-day survival when volunteer responders accepted the alarm.
The original contributions presented in this study are included in the article/
Ethical review and approval was not required for the study on human participants in accordance with the local legislation and institutional requirements. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.
CN: study conception and design, analysis and interpretation of data, and writing the manuscript. LA, CH, UV, EC, and CT-P: interpretation of data and revision of manuscript. AE: statistical review, analysis and interpretation of data, and revision of manuscript. FF and MG: study conception and design, analysis and interpretation of data, and revision of manuscript. All authors have read and approved the final manuscript.
This study was supported by research grants from the Danish Foundation, TrygFonden, Copenhagen, Denmark. FF received research grant from the Novo Nordisk Foundation (NNF190C0055142). MG and LA received research grants from TrygFonden. CH received research grants from TrygFonden, Helsefonden, Laerdal Foundation, and Zoll. CT-P received grants for studies from Bayer and Novo Nordisk Foundation unrelated to the current study.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
The Supplementary Material for this article can be found online at:
Directed acyclic graph showing the included variables and possible confounders for bystander cardiopulmonary resuscitation as outcome.
Directed acyclic graph showing the included variables and possible confounders for bystander defibrillation as outcome.
Directed acyclic graph showing the included variables and possible confounders for 30-day survival as outcome.