Cardiopulmonary resuscitation by trained responders versus lay persons and outcomes of out-of-hospital cardiac arrest: A community observational study

Accepted Manuscript Title: Cardiopulmonary Resuscitation by Trained Responders Versus Lay Persons and Outcomes of Out-of-Hospital Cardiac Arrest: A Community Observational Study Authors: Yoo Mi Park, Sang Do Shin, Yu Jin Lee, Kyoung Jun Song, Young Sun Ro, Ki Ok Ahn PII: DOI: Reference:

S0300-9572(17)30267-8 http://dx.doi.org/doi:10.1016/j.resuscitation.2017.06.024 RESUS 7228

To appear in:

Resuscitation

Received date: Revised date: Accepted date:

20-3-2017 18-5-2017 26-6-2017

Please cite this article as: Park Yoo Mi, Shin Sang Do, Lee Yu Jin, Song Kyoung Jun, Ro Young Sun, Ahn Ki Ok.Cardiopulmonary Resuscitation by Trained Responders Versus Lay Persons and Outcomes of Out-ofHospital Cardiac Arrest: A Community Observational Study.Resuscitation http://dx.doi.org/10.1016/j.resuscitation.2017.06.024 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Title Page 1. Title Cardiopulmonary Resuscitation by Trained Responders Versus Lay Persons and Outcomes of Out-of-Hospital Cardiac Arrest: A Community Observational Study 2. Running Head Title Trained Responder CPR and Outcomes in OHCA 3. Authors Yoo Mi Park, MD Department of Health Policy and Management, Seoul Metropolitan City E-mail: [email protected] Sang Do Shin, MD, PhD Department of Emergency Medicine, Seoul National University College of Medicine E-mail: [email protected] Yu Jin Lee, MD Department of Emergency Medicine, National Medical Center E-mail: [email protected] Kyoung Jun Song, MD, PhD Department of Emergency Medicine, Seoul National University Hospital Laboratory of Emergency Medical Services, Seoul National University Hospital Biomedical Research Institute E-mail: [email protected] Young Sun Ro, MD, DrPH Laboratory of Emergency Medical Services, Seoul National University Hospital Biomedical Research Institute E-mail: [email protected] Ki Ok Ahn, MD, PhD

Laboratory of Emergency Medical Services, Seoul National University Hospital Biomedical Research Institute E-mail: [email protected] Conflict of interest statement 4. Funding Acknowledgement: This study was supported by the Seoul Metropolitan Fire Department and Seoul Metropolitan Health Department of Korea. The study was funded by the Korea Centers for Disease Control and Prevention (2012-2015) (Grant Nos: 2012-E33010-00; 2013-E3301500; 2014-E33011-00; 2015-Grant for Private Support Program). The Korea CDC approved the use of OHCA data in this study. 5. Word Count Number: 3133 (excluding the abstract, references, tables, and figures) 6. Author Contributions: Dr. Park and Dr. Shin had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Drs. Shin, Ro, Ahn, and Song. Acquisition, analysis, or interpretation of the data: Drs. Park and Shin. Drafting of the manuscript: Drs. Park and Shin. Critical revision of the manuscript for important intellectual content: Drs. Shin, Lee, Ahn, and Ro. Statistical analysis: Dr. Shin. Obtainment of funding: Dr. Shin. Manuscript approval: all authors. 7. Address for reprints: Sang Do Shin, MD, PhD (Corresponding author) Address: Department of Emergency Medicine, Seoul National University College of Medicine, 101 Daehak-Ro, Jongno-Gu, Seoul 110-744, Korea

Phone: +82-2-2072-3257 FAX: +82-2-741-7855 E-mail: [email protected] 8.

ABSTRACT Objectives Keywords: Cardiac Arrest; Bystander; Cardiopulmonary Resuscitation; Defibrillation The study aims to compare bystander processes of care (cardiopulmonary resuscitation (CPR) and defibrillation) and outcomes for witnessed presumed cardiac etiology in OHCA patients in whom initial resuscitation was provided by dedicated trained responder (TR) versus lay person (LP) bystanders. Methods Data on witnessed and presumed cardiac OHCA in adults (15 years or older) from 2011 to 2015 in a metropolitan city with 10 million persons were collected, excluding cases in which the information on TRs, bystander CPR, defibrillation, and clinical outcomes was unknown. Exposure variables were TRs who were legally designated with CPR education and response and LPs who were bystanders who witnessed the OHCA by chance. The primary/secondary/tertiary outcomes were a good cerebral performance category (CPC) of 1 or 2, survival to discharge, and bystander defibrillation. A multivariable logistic regression analysis was used to calculate the adjusted odds ratio (AOR) with 95% confidence intervals (CIs), adjusting for potential confounders. Results Of 20,984 OHCA events, 6,475 cases were ultimately analyzed. The TR group constituted 6.4% of the cases, and the patients showed significantly better survival and a good CPC. From the multivariable logistic regression analysis of the outcomes, by comparing the TR group with the LP group, the AOR (95% CIs) was 1.49 (1.04-2.15) for a good CPC, 1.59 (1.202.11) for survival to discharge, and 10.02 (7.04-14.26) for bystander defibrillation. Conclusion

The TR group witnessed a relatively low proportion of OHCA but was associated with better survival outcomes and good neurological recovery through higher CPR rates and defibrillation of adults older than 15 years with witnessed OHCA in a metropolitan city. Keywords: Cardiac Arrest, Trained Responder, Cardiopulmonary Resuscitation, Defibrillation

BACKGROUND Out-of-hospital cardiac arrest (OHCA) is the one of the largest public health burdens due to its high incidence and low survival rates in the United States, Europe, Australia, and Asia.(13) To improve the outcomes of OHCA, evidence-based guidelines for cardiopulmonary resuscitation have been proposed through a scientific review process by the International Liaison Committee on Resuscitation.(4, 5) However, a substantial gap between scientific knowledge and clinical practice has been apparent over the last several decades, and the implementation of current recommendations in cardiopulmonary resuscitation (CPR) at the community and population levels is one of the most difficult steps in delivering these scientific findings to the community.(6, 7) Early good quality CPR and appropriate training of trained responders (TRs) with a duty to respond to an OHCA event were recommended under the concept of “formula of survival” and by the 2015 Guidelines for CPR of the European Resuscitation Council and the American Heart Association.(4, 8) The public access defibrillator (PAD) program using automatic external defibrillators (AEDs) for those TRs is one of the key strategies for improving outcomes by early defibrillation.(9, 10) The PAD program has been argued for its costeffectiveness and low utilization issues, regardless of scientific evidence on its clear benefit in OHCA patients.(11-13) An AED deployment strategy targeting TRs, who are willing or likely to run to OHCA patients, has been regarded as a more cost-effective approach than that targeting general lay person (LP) bystanders. TRs are a specific population group that is likely to witness a patient collapsing or to be called when an event occurs in daily life. However, TRs include various population groups depending on the EMS systems and communities involved, including firefighters, policemen, public transportation vehicle drivers, school teachers, sports instructors, and lifeguards who may or may not be designated by the Emergency Medical Services (EMS) Act and regulations.(9, 10) The proportions of OHCAs witnessed by TRs among all patients, the CPR

rates performed by TRs, and outcomes after TR CPR and TR defibrillation are uncertain and not fully understood. The lack of a standard definition and standard criteria for TRs and the extensive variations in the public health regulations for TRs are causing the gap between scientific recommendations and real implementations in the communities. This study aimed to compare bystander activities (CPR and defibrillation) between TRs and LPs and their subgroups and to test the association between bystander groups (TRs and LPs) and outcomes.

METHODS This was a citywide cross-sectional observational study. All data were collected and owned by the Korea Centers for Disease Control and Prevention (CDC) according to national statistics law and were approved for this study. The institutional review board of the study hospital reviewed and approved the study. Informed consent was waived because the data variables did not include personal information, and the study process posed a minimal risk for patients. Study setting The study was conducted in a metropolitan city that spans 605 km2 with a population of approximately 10 million residents. There are 25 districts, each with one health center that provides CPR training to TRs and LPs. The CPR training standards and materials were developed and distributed by the Korea CDC and were based on international and domestic recommendation guidelines. The 2010 national CPR training standards and program were developed and approved in 2011 and then disseminated to entire provinces and metropolises. The national EMS Act was revised to designate potential TRs (school teachers, sports instructors, public transportation vehicle drivers, safety guards of national parks, and policemen) in 2004 to encourage bystander CPR. A revision was made in 2008 for bystander defibrillation and then in 2011 to designate more TRs (apartment safety guards in towns with more than 500 houses), and mandatory training for TR CPR and defibrillation was added to the act. Most places where TRs work or live are mandatory sites for PAD programs designated by the EMS Act. Since 2005, trained responders have been required to complete regular CPR and AED training every year with at least one two-hour course according to the EMS Act. (See Appendix 1 for the Korean TRs designated by the national EMS Act and the mandatory sites for the PAD program.)

The metropolitan city of Seoul developed a 2nd five-year EMS agenda in 2010 and has implemented programs based on the agenda since 2011. The agenda includes the expansion of CPR training and implementation of the public AED program. Every year, 500-1000 AEDs are distributed in public spaces supported by the city health department. The total number of AEDs for bystander use was approximately 8,000 in 2015 (80 per 100,000 persons and 1,250 persons per AED). An intermediate level of EMS care is provided by the city fire department (170 ambulances and 3 crews per ambulance) and includes CPR, AED, advanced airway, and intravenous fluid resuscitation. However, the use of medications is not permitted during CPR, and emergency medical technicians are mandated to transport all cardiac arrest patients to emergency departments while continuing CPR in the ambulance if the patients are not resuscitated on the ground.(14) There are three levels of emergency departments (EDs): one level 1 ED, where 24 hour/7 day emergency care for critically ill emergency patients is served by specialized emergency physicians; 27 level 2 EDs, where emergency physicians provide emergency care for emergency patients with high acuity; and 23 level 3 EDs, where general physicians provide emergency care for patients with low acuity. The designation, evaluation, and accreditation of level of EDs are provided annually by the national government’s health department. Data source The national OHCA registry was used for the study, which included all EMS-assessed OHCAs since 2006 (15-17). The registry was constructed from four electronic databases: 1) a dispatch CPR registry recorded by dispatchers for information and pre-arrival instructions, 2) EMS run sheets on general and time-related information regarding ambulance operations, 3) an EMS CPR registry for OHCA event information recorded by EMS providers after transporting OHCA patients, and 4) hospital medical records collected by the Korea CDC on hospital care and outcomes. OHCA cases were obtained from EMS run sheets from the

national fire department’s server and merged with the dispatch CPR registry and EMS CPR registry. The OHCA cases were collected and sent to the Korea CDC to obtain hospital medical records, which are created by trained medical record reviewers who go to hospitals and review all hospital records considering the care in the ED, intensive care unit, and wards, as well as the outcomes at discharge. The data quality management team consisting of EMS physicians, epidemiologists, biostatistics experts, and cardiologists maintains the data quality through regular monthly education and providing feedback to medical record reviewers about undetermined variables during medical record reviews. The OHCA statistics from this registry were approved by the National Statistical Office as one set of national health statistics. Study population Adults aged 15 years or older who suffered from OHCA and had a presumed cardiac etiology from January 1, 2011, to December 31, 2015, were enrolled. A cardiac etiology was presumed in the absence of any other obvious cause such as trauma, drowning, hanging, overdose, or asphyxiation, as well as by using clinical information gathered in some cases from the medical record of physicians. Patients who were not treated, who were witnessed by EMS providers, who were unwitnessed in general, who had collapsed in an unknown place, who had unknown outcomes, and who had unknown information regarding bystander CPR and AED use were excluded. Data variables The general and demographic variables included the patients’ age, sex, address, and Utstein factors such as location, witness status, first recorded rhythm (shockable versus nonshockable), bystander CPR and defibrillation with/without dispatch assistance (DA). Variables also included EMS factors, such as time intervals (including the response time from call to arrival at the scene, the scene time from arrival at the scene to departure to the ED, and the transport time from departure from the scene to arrival at the ED), airway

management, and EMS defibrillation. Hospital factors were also included, such as hospital level (level 1 to level 4 designated by the national health department according to the hospital’s performance level and capacity of emergency care), ED CPR and defibrillation, post-resuscitation care (targeted temperature management, primary coronary intervention, and implantable cardiac defibrillator), co-morbidities, and outcomes at discharge. The above variables in the EMS registry were defined in the data dictionary, and EMS providers were educated and trained by the EMS education core program and quality assurance program led by the national fire department. Exposure variables were the LP (lay person) group versus the TR (trained responder) group, which were classified according to the EMS Act of Korea. The TR group included 1) professionals working at nursing facilities, 2) police officers, 3) health teachers, 4) transportation vehicle drivers, 5) sports facility employees, 6) lifeguards, 7) workplace safety employees, and 8) travel business employees. The LP group included 1) off-duty EMS providers, 2) off-duty medical professionals, 3) family, 4) nearby bystanders, 5) friends or colleagues, and 6) other bystanders. The information on the TR or LP groups was collected and recorded by EMS providers who had attended to the scene and provided CPR. The EMS CPR registry contains the variables regarding who provides CPR or defibrillation before EMS arrives to the scene. Outcome measure The primary outcome was good neurological recovery (cerebral performance scale (CPC) of 1 or 2), and the secondary outcome was survival to discharge. The outcome measures were collected by the Korea CDC medical record review. The tertiary outcome was bystander AED defibrillation, which was defined as when a bystander (TR or LP) provided AED defibrillation prior to the arrival of EMS providers. The quaternary outcome was the bystander CPR rate. The application of AED and bystander CPR was recorded in the EMS

CPR registry by EMS providers when they attended to the scene. All records related with bystander defibrillation are recorded by EMTs to review the rhythm and outcomes. Statistical analysis Demographic variables were described by the distribution of potential risk factors for outcomes between the LP and TR groups. Categorical variables were compared with the Chisquare test, and continuous variables were compared using the Wilcoxon rank sum test. The trends of bystander CPR rate, defibrillation rate, survival to discharge rate and good neurological recovery rate were tested using the Cochran-Armitage trend test according to the study period (year). A multivariable logistic regression analysis was performed to test the association between TRs and LPs for the outcomes, adjusting for potential confounders such as age, sex, time of the event, weekend of the event, season, place of the event, and first recorded ECG rhythm on the PAD or EMS defibrillator. From the models, adjusted odds ratios (ORs) and 95% confidence intervals (CIs) were calculated. The other significant factors such as response time interval were not adjusted because the factors were followed after the exposure factor in terms of the time process between the risk factors and exposure factor. We performed an interaction analysis to compare the effect size of the bystander group (TR versus LP), according to the OHCA location (public versus private), on the outcomes adding the interaction term (bystander group*location of arrest) to the final multivariable logistic models. We tested the association between expanded trained responders including medical professionals and lay person responders using the same multivariable logistic regression analysis to remove the bias by the off-duty medical professionals (sensitivity analysis).

RESULTS The total number of eligible OHCA events for the study period was 20,984. Of these, 6,475 were ultimately analyzed, excluding pediatric events (n=302), non-cardiac events (n=4,545), EMS-witnessed events (n=1,288), unknown information on the bystander (n=679), unknown locations (n=282), unwitnessed events (n=7,124), CPR not attempted (n=279), and unknown outcomes (n=10) (Fig 1). Tables 1 and 2 show the demographic findings of the study population between the TR and LP groups (Table 1). Of the 6,475 eligible subjects, 412 (6.4%) cases in total were witnessed by trained responders. More patients witnessed by the TR group compared to the LP group were younger and female and included more public locations, shockable rhythms, occurrences in the daytime, and day of arrest (weekday versus weekend). The rate of bystander CPR, bystander defibrillation, survival to discharge, and good CPC was significantly higher in the TR group than the LP group (Table 2). The trends in bystander CPR and defibrillation, followed by the rates of survival to discharge and good neurological recovery according to the study period, were significantly changed by year. The bystander CPR rates significantly increased by year in the LP group (p for trend: <0.001), whereas they did not increase in the TR group (p for trend: 0.318). However, the bystander defibrillation rates significantly increased in the TR group (p for trend: 0.000) but not in the LP group (p for trend: 0.339) (Table 3). The rates of bystander CPR, bystander defibrillation, survival to discharge, and good CPC among the bystander groups varied according to the bystander group. The TR group showed much higher CPR rates by bystanders and defibrillation rates by medical professionals, health teachers, and lifeguards, whereas the LP group showed very low bystander defibrillation rates in most bystander groups. The average bystander defibrillation rate was 2.4% for the entire group, 14.6% for the TR group, and 1.6% for the LP group (Table 4).

Table 5 presents the multivariable logistic regression analysis on the outcomes for the TR group compared with the LP group. The adjusted ORs (95% CIs) were 1.49 (1.04-2.15) for good neurological recovery and 1.59 (1.20-2.11) for survival to discharge. The AOR (95% CIs) was significantly higher in the TR group for bystander defibrillation (10.02, 7.04-14.26) and for bystander CPR (5.31, 4.04-6.97). The TR group was associated with better outcomes (good CPC and survival) in public locations but not in private locations according to the interaction analysis: 1.76 (1.06-2.94) vs. 1.28 (0.78-2.11) for good CPC and 2.01 (1.41-2.88) vs. 1.14 (0.74-1.77) for survival to discharge (Table 6). The bystander CPR and defibrillation rates were significantly increased in the TR group compared with the LP group in both public and private locations. From the sensitivity analysis, the AORs (95% CIs by expanded TR (including off-duty medical professionals)) versus LP bystanders were 1.54 (1.11-2.14) for good CPC, 1.55 (1.202.00) for survival to discharge, 10.87 (7.72-15.30) for bystander defibrillation, and 4.53 (3.59-5.72) bystander CPR.

DISCUSSION We found that the outcomes after OHCA in adults with a presumed cardiac etiology were significantly associated with the responder groups who witnessed or found the patients. The trained responders designated by the EMS Act provided much more bystander CPR and bystander defibrillation than the lay person bystanders did. Trained responders in communities are expected to witness cardiac arrest events much more frequently. These TRs are instructed to complete regular CPR training. However, only 6.4% were in the TR group, but the bystander CPR and defibrillation rates were much higher (5 to 10 times) and resulted in better outcomes. In a previous study, bystanders and trained responders (i.e., first responders) were mainly responsible for early defibrillation within 5 min of the event and subsequently resulted in better outcomes, independent of the location of the arrest.(18) In this study, the time to the first shock by a lay person or trained responder was strongly associated with survival. The survival rate in cases where the first shock was received within 1 min was higher than 60%. The trained responders who were designated by legal regulation varied. Most medical professionals (physicians, nurses, and EMTs) are trained during their education and training and received continuing medical education for CPR in this setting. The EMS Acts do not include the professionals as trained responders because they are not employed by the relevant places and businesses. We tested the association in the sensitivity analysis between expanded trained responders (including medical professionals such as 27 EMTs and 74 physicians/nurses) and better outcomes and found that the effects were persistently significant.

Bystander CPR rates rapidly increased during the study period in the LP group. The increase in bystander CPR rate might be related to the dispatcher-assisted CPR program, which has been implemented since 2011 in this metropolitan city.(17, 19) The bystander CPR rates are

similar or higher than those of the other countries. The number of public AEDs per population in the city was not as large as those of other countries. The bystander defibrillation rate is still low due to limited numbers of public AEDs. Our data showed a relatively lower rate 2.4% (155/ 6475) of witnessed OHCAs in adults who received bystander defibrillation. In Osaka, 3.5% of OHCA patients (351/9,978) received bystander defibrillation using a public AED.(13) In the whole of Japan, of 43,762 patients with bystander-witnessed ventricular-fibrillation arrests of cardiac origin, 4499 (10.3%) received public-access defibrillation.(20) Of 13,769 out-of-hospital cardiac arrests in North America according to the Resuscitation Outcome Consortium study during 2005 to 2007, 289 (2.1%) had an AED applied before EMS arrival. The AED was applied by health care workers (32%), lay volunteers (35%), police (26%), or unknown persons (7%). (21) The other study in the U.S. reported that only 3.7% were treated by bystanders with an AED before the arrival of EMS providers.(22) A secondary analysis used data from 29 U.S. cities that participated in the Cardiac Arrest Registry to Enhance Survival (CARES) and found that the rate of bystander defibrillation by lay persons was 11.3%.(23) One strategy for using public AEDs effectively is to strategically distribute AEDs in geographically high-risk areas defined as having a previous occurrence of OHCAs, for example, 1 OHCA per 2 years (European Resuscitation Council) or per 5 years (American Heart Association) (12, 24). The other strategy is to distribute AEDs near trained responders who are working or living in high-risk areas for OHCA events using the AED network concept.(24) The public AED program in this study setting did not choose a strategic distribution. The legally designated areas were covered by government programs for public AEDs such as stations, public offices, and sports stadiums. There has been no consideration for distributing AEDs in historically high-risk areas. The second strategy will be the activation of bystanders to run to the patients’ side, to increase bystander CPR.(25) The protocol was developed for dispatchers to send information on OHCA events via a mobile phone application to lay persons who are in the accessible area. The protocol may be

extended to the activation of trained responders, to increase the use of PAD. The third strategy will be the designated trained responder program for PAD activation during OHCA events in private locations. Our study showed much lower bystander defibrillation rates in private locations than in public locations. Apartment security guards may be potential candidates for a trained responder activation program. According to the TR category, the CPR activities such as bystander CPR and defibrillation varied (30.8% of bystander defibrillation in rescue guards versus 5.9% in safety guards in sports facilities). These findings can be considered for the program development of CPR training and PAD distribution as a policy priority setting. Family members witnessed the majority of OHCAs in private locations. Only 1.1% of patients received bystander defibrillation, but the bystander CPR rates were not low (51.4%). The findings suggest that the strategy to provide early defibrillation should be changed. A current technology encourages us to consider providing PADs by a remote controlled drone flying to locations without any PADs.(26, 27) The strategy for TR training may be considered 1) from volunteerbased to mandatory, 2) as a dedicated role for all TRs for defibrillation by law, 3) as adding educational content for TR activation followed by TR defibrillation instructed by dispatchers through real-time communication. We found a 50% survival benefit (AOR=1.49 for a good CPC) in the TR group compared with the LP group. The TR group exhibited more conditions for better outcomes in the demographic findings as well as higher bystander CPR rates and defibrillation. A possible reason for the better outcomes was rapid defibrillation and CPR in the TR group. Traditionally, the rapid dispatch of trained responders tends to lead to rapid defibrillation within a few minutes immediately after receiving an ambulance call.(28, 29) The FP group showed a significant improvement in outcomes in the public location but not in the private location in the interaction analysis. The TRs were designated and trained in accordance with the EMS Act in the country. Currently, TRs in private locations are

apartment guards who are mostly elderly and retired persons, whereas TRs in public locations are commonly younger and full-time employees in sports facilities, safety guards of buildings, or employees of transportation companies. These characteristics were not examined in the study due to lack of information. Further investigations are required to compare the characteristics of trained responders between public and private locations.

Limitations The present study had several limitations. First, the recording of exposure variables (type of TR group) and outcomes (bystander CPR and defibrillation) was performed by EMS providers after transporting patients to the ED. The EMS providers would have received the information from the CPR field, which would have been affected by measurement bias. We never tested the reliability of the measurements. Although the national fire department and provincial headquarters have quality assurance programs for data collection and registries, interrater reliability would need to be considered. Activity for maintaining the quality of data was highlighted by a strong and active feedback program by each medical director under the EMS agency (n=23) and dispatch center (n=1). Every OHCA registry was reviewed by each medical director, and feedback was provided to the EMS provider team responsible for the OHCA. This program is based on the EMS and Rescue Act of Korea. Second, the study collected outcomes information from retrospective medical records of hospital data. The process may have led to detection bias during the medical record review. Third, although a strong effort for data quality was made by the multidisciplinary expert team, the data integrity issue remains because the prehospital registry and the hospital registry were merged. Fourth, the study was conducted at an intermediate service level of the EMS, where CPR protocols were very different from those of Europe or North America. The EMS CPR protocol does not allow any advanced life support drugs such as epinephrine, invasive procedures, and manual defibrillation. Only advanced airway management under direct

medical control is allowed in the field. Thus, generalization of the study findings should be done with caution.

CONCLUSION Trained responder bystanders witnessed a relatively low proportion of OHCAs but were significantly associated with better outcomes after witnessed OHCAs in adults in a metropolitan city. Bystander CPR and defibrillation had much higher rates in the trained responder bystander group than among lay person bystanders, which resulted in better outcomes. DISCLOSURES None.This study was supported by the Seoul Metropolitan Fire Department and Health Department and financially supported by Korea and the Korea CDC (2012-2015). There are no conflicts of interest for all authors in this study.

ACKNOWLEDGEMENTS This study was supported by the Seoul Metropolitan Fire Department and Health Department and financed by Korea and the Korea CDC (2012-2015) (Grant Nos.: 2011Grant for Private Support Program; 2012-E33010-00; 2013-E33015-00; 2014-E33011-00; 2015-Grant for Private Support Program).

REFERENCES 1.

McNally B, Stokes A, Crouch A, Kellermann AL, Group CS. CARES: Cardiac Arrest Registry to

Enhance Survival. Ann Emerg Med. 2009;54(5):674-83 e2.

2.

Ong ME, Shin SD, De Souza NN, Tanaka H, Nishiuchi T, Song KJ, et al. Outcomes for out-of-

hospital cardiac arrests across 7 countries in Asia: The Pan Asian Resuscitation Outcomes Study (PAROS). Resuscitation. 2015;96:100-8. 3.

Berdowski J, Berg RA, Tijssen JG, Koster RW. Global incidences of out-of-hospital cardiac

arrest and survival rates: Systematic review of 67 prospective studies. Resuscitation. 2010;81(11):147987. 4.

Hazinski MF, Nolan JP, Aickin R, Bhanji F, Billi JE, Callaway CW, et al. Part 1: Executive

Summary: 2015 International Consensus on Cardiopulmonary Resuscitation and Emergency Cardiovascular Care Science With Treatment Recommendations. Circulation. 2015;132(16 Suppl 1):S239. 5.

Morley PT, Lang E, Aickin R, Billi JE, Eigel B, Ferrer JM, et al. Part 2: Evidence Evaluation and

Management of Conflicts of Interest: 2015 International Consensus on Cardiopulmonary Resuscitation and Emergency Cardiovascular Care Science With Treatment Recommendations. Circulation. 2015;132(16 Suppl 1):S40-50. 6.

Meaney PA, Bobrow BJ, Mancini ME, Christenson J, de Caen AR, Bhanji F, et al.

Cardiopulmonary resuscitation quality: [corrected] improving cardiac resuscitation outcomes both inside and outside the hospital: a consensus statement from the American Heart Association. Circulation. 2013;128(4):417-35. 7.

Heffner AC, Pearson DA, Nussbaum ML, Jones AE. Regionalization of post-cardiac arrest care:

implementation of a cardiac resuscitation center. Am Heart J. 2012;164(4):493-501 e2. 8.

Soreide E, Morrison L, Hillman K, Monsieurs K, Sunde K, Zideman D, et al. The formula for

survival in resuscitation. Resuscitation. 2013;84(11):1487-93. 9.

Malta Hansen C, Kragholm K, Pearson DA, Tyson C, Monk L, Myers B, et al. Association of

Bystander and First-Responder Intervention With Survival After Out-of-Hospital Cardiac Arrest in North Carolina, 2010-2013. JAMA. 2015;314(3):255-64. 10.

Nelson RD, Bozeman W, Collins G, Booe B, Baker T, Alson R. Mobile Versus Fixed Deployment

of Automated External Defibrillators in Rural EMS. Prehosp Disaster Med. 2015;30(2):152-4. 11.

Agerskov M, Nielsen AM, Hansen CM, Hansen MB, Lippert FK, Wissenberg M, et al. Public

Access Defibrillation: Great benefit and potential but infrequently used. Resuscitation. 2015;96:53-8. 12.

Folke F, Lippert FK, Nielsen SL, Gislason GH, Hansen ML, Schramm TK, et al. Location of

cardiac arrest in a city center: strategic placement of automated external defibrillators in public locations. Circulation. 2009;120(6):510-7. 13.

Kiyohara K, Kitamura T, Sakai T, Nishiyama C, Nishiuchi T, Hayashi Y, et al. Public-access AED

pad application and outcomes for out-of-hospital cardiac arrests in Osaka, Japan. Resuscitation. 2016;106:70-5. 14.

Ro YS, Shin SD, Lee YJ, Lee SC, Song KJ, Ryoo HW, et al. Effect of Dispatcher-Assisted

Cardiopulmonary Resuscitation Program and Location of Out-of-Hospital Cardiac Arrest on Survival and Neurologic Outcome. Ann Emerg Med. 2017;69(1):52-61 e1. 15.

Ro YS, Shin SD, Song KJ, Lee EJ, Kim JY, Ahn KO, et al. A trend in epidemiology and outcomes

of out-of-hospital cardiac arrest by urbanization level: a nationwide observational study from 2006 to 2010 in South Korea. Resuscitation. 2013;84(5):547-57. 16.

Ro YS, Shin SD, Song KJ, Hong SO, Kim YT, Lee DW, et al. Public awareness and self-efficacy

of cardiopulmonary resuscitation in communities and outcomes of out-of-hospital cardiac arrest: A multi-level analysis. Resuscitation. 2016;102:17-24. 17.

Ro YS, Shin SD, Lee YJ, Lee SC, Song KJ, Ryoo HW, et al. Effect of Dispatcher-Assisted

Cardiopulmonary Resuscitation Program and Location of Out-of-Hospital Cardiac Arrest on Survival and Neurologic Outcome. Ann Emerg Med. 2016. 18.

Hansen CM, Kragholm K, Granger CB, Pearson DA, Tyson C, Monk L, et al. The role of

bystanders, first responders, and emergency medical service providers in timely defibrillation and related outcomes after out-of-hospital cardiac arrest: Results from a statewide registry. Resuscitation. 2015;96:303-9. 19.

Song KJ, Shin SD, Park CB, Kim JY, Kim DK, Kim CH, et al. Dispatcher-assisted bystander

cardiopulmonary resuscitation in a metropolitan city: a before-after population-based study. Resuscitation. 2014;85(1):34-41. 20.

Kitamura T, Kiyohara K, Sakai T, Matsuyama T, Hatakeyama T, Shimamoto T, et al. Public-

Access Defibrillation and Out-of-Hospital Cardiac Arrest in Japan. N Engl J Med. 2016;375(17):164959. 21.

Weisfeldt ML, Sitlani CM, Ornato JP, Rea T, Aufderheide TP, Davis D, et al. Survival after

application of automatic external defibrillators before arrival of the emergency medical system: evaluation in the resuscitation outcomes consortium population of 21 million. J Am Coll Cardiol.

2010;55(16):1713-20. 22.

McNally B, Robb R, Mehta M, Vellano K, Valderrama AL, Yoon PW, et al. Out-of-hospital

cardiac arrest surveillance --- Cardiac Arrest Registry to Enhance Survival (CARES), United States, October 1, 2005--December 31, 2010. MMWR Surveill Summ. 2011;60(8):1-19. 23.

Johnson MA, Grahan BJ, Haukoos JS, McNally B, Campbell R, Sasson C, et al. Demographics,

bystander CPR, and AED use in out-of-hospital pediatric arrests. Resuscitation. 2014;85(7):920-6. 24.

Hansen CM, Lippert FK, Wissenberg M, Weeke P, Zinckernagel L, Ruwald MH, et al. Temporal

trends in coverage of historical cardiac arrests using a volunteer-based network of automated external defibrillators accessible to lay persons and emergency dispatch centers. Circulation. 2014;130(21):185967. 25.

Ringh M, Rosenqvist M, Hollenberg J, Jonsson M, Fredman D, Nordberg P, et al. Mobile-phone

dispatch of lay persons for CPR in out-of-hospital cardiac arrest. N Engl J Med. 2015;372(24):2316-25. 26.

Claesson A, Fredman D, Svensson L, Ringh M, Hollenberg J, Nordberg P, et al. Unmanned

aerial vehicles (drones) in out-of-hospital-cardiac-arrest. Scand J Trauma Resusc Emerg Med. 2016;24(1):124. 27.

Pulver A, Wei R, Mann C. Locating AED Enabled Medical Drones to Enhance Cardiac Arrest

Response Times. Prehosp Emerg Care. 2016;20(3):378-89. 28.

Callaham M, Madsen CD. Relationship of timeliness of paramedic advanced life support

interventions to outcome in out-of-hospital cardiac arrest treated by first responders with defibrillators. Ann Emerg Med. 1996;27(5):638-48. 29.

MacDonald RD, Mottley JL, Weinstein C. Impact of prompt defibrillation on cardiac arrest at

a major international airport. Prehosp Emerg Care. 2002;6(1):1-5.

FIGURE LEGEND

Fig 1 Patient Enrollment EMS: emergency medical services OHCA: out-of-hospital cardiac arrest

CPR: cardiopulmonary resuscitation

Figr-1

Table 1 Demographic findings of patients between Trained responder and lay person bystander group

Variables All Age 15-39, years 40-59, years 60-79, years 79- , years Median (q1-q3) Gender Female Male Location Public Private First recorded ECG Shockable Non-shockable Hour of the event 00:00-05:59 06:00-11:59 12:00-17:59 18:00-23:59 Season MAR.-MAY JUN-AUG SEP-NOV DEC-FEB Weekend Weekday Weekend

All N 6475

LP % 100.0

N 6063

TR % 100.0

N 412

% 100.0

p-value* <0.001

380 5.9 1676 25.9 2906 44.9 1513 23.4 70 (56-79)

337 5.6 1553 25.6 2745 45.3 1428 23.6 70 (57-80)

43 10.4 123 29.9 161 39.1 85 20.6 66 (51-78)

2217 4258

2045 4018

172 240

<0.001 34.2 65.8

33.7 66.3

41.7 58.3 <0.001

1419 5056

21.9 78.1

1293 4770

21.3 78.7

126 286

30.6 69.4

1691 4784

26.1 73.9

1571 4492

25.9 74.1

120 292

29.1 70.9

1000 1901 1900 1674

15.4 29.4 29.3 25.9

971 1757 1737 1598

16.0 29.0 28.6 26.4

29 144 163 76

7.0 35.0 39.6 18.4

1627 1471 1615 1762

25.1 22.7 24.9 27.2

1504 1385 1521 1653

24.8 22.8 25.1 27.3

123 86 94 109

29.9 20.9 22.8 26.5

0.018 <0.001

0.141

<0.001 4599 1876

71.0 29.0

4272 1791

70.5 29.5

327 85

79.4 20.6

TR, Trained responder; LP, lay person bystander; CPR, cardiopulmonary resuscitation *P values for categorical variables were calculated by the chi-square test.

Table 2 Bystander cardiopulmonary resuscitation, defibrillation, prehospital return of spontaneous circulation, and hospital outcomes between trained responder and lay person bystander group

Variables

All N 6475

% 100.0

LP N 6063

% 100.0

TR N 412

P-value % 100.0

All Bystander CPR <.001 No 3046 47.0 2981 49.2 65 15.8 Yes 3429 53.0 3082 50.8 347 84.2 Bystander defibrillation <.001 No 6320 97.6 5968 98.4 352 85.4 Yes 155 2.4 95 1.6 60 14.6 Outcomes Prehospital ROSC 652 10.1 591 9.7 61 14.8 0.001 Survival to discharge 989 15.3 894 14.7 95 23.1 <.001 Good CPC 578 8.9 523 8.6 55 13.3 0.001 TR, Trained responder; LP, lay person bystander; CPR, cardiopulmonary resuscitation; Good CPC, cerebral performance category 1 or 2; ROSC, return of spontaneous circulation *P values for categorical variables were calculated by the chi-square test.

Table 3 Trend of trained responders’ and lay persons’ bystander cardiopulmonary resuscitation and defibrillation by year

All

Year 2011

2012

2013

2014

p for 2015 trend*

6475

1078

1239

1181

1435

1542

Bystander CPR, %

53.0

39.6

46.2

54.6

59.9

60.0

<0.001

Bystander defibrillation, %

2.4

1.5

0.9

3.5

3.5

2.4

0.002

Survival to discharge, %

15.3

13.5

15.3

15.8

16.0

15.4

0.177

Good CPC, %

8.9

6.1

7.6

8.3

11.0

10.5

<0.001

6063

1000

1187

1104

1339

1433

50.8

36.4

44.5

52.5

58.3

57.9

<0.001

Bystander defibrillation, %

1.6

1.0

0.7

3.2

2.0

1.0

0.339

Survival to discharge, %

14.7

12.8

14.8

15.6

15.3

14.9

0.210

Good CPC, %

8.6

5.7

7.4

8.3

10.5

10.1

<0.001

412

78

52

77

96

109

Bystander CPR, %

84.2

80.8

84.6

84.4

83.3

87.2

0.318

Bystander defibrillation, %

14.6

7.7

5.8

7.8

24.0

20.2

<0.001

Survival to discharge, %

23.1

21.8

25.0

19.5

26.0

22.9

Good CPC, %

13.3

11.5

11.5

7.8

17.7

15.6

0.774 0.200

Outcomes Total, N

LP group, N Bystander CPR, %

TR group, N

TR, trained responder; LP, lay person bystander; CPR, cardiopulmonary resuscitation; Good CPC, cerebral performance category 1 or 2 *Ps for trend were calculated by Cochran-Armitage test

Table 4 Bystander CPR and defibrillation rate according to bystander group

Group Total (N) Bystander CPR (%) Bystander defibrillation (%) Survival to discharge (%) Good CPC (%) Total 6475 53.0 2.4 15.3 8.9 TR group, subtotal 412 84.2 14.6 23.1 13.3 Medical professional 302 86.4 14.9 19.9 9.6 Police officer 40 70.0 7.5 30.0 15.0 Health teacher 9 100.0 22.2 22.2 22.2 Transportation vehicle driver 12 58.3 8.3 0.0 0.0 Safety guard of sports facility 17 94.1 5.9 70.6 64.7 Rescue guard 13 92.3 30.8 23.1 23.1 Workplace safety guard 8 75.0 12.5 25.0 25.0 Safety guard of travel business 11 72.7 27.3 36.4 18.2 LP group, subtotal 6063 50.8 1.6 14.7 8.6 Off-duty EMS provider 27 55.6 22.2 29.6 18.5 Off-duty medical professional 74 78.4 8.1 18.9 14.9 Family member 4264 51.4 1.1 11.1 6.0 Nearby bystander 469 47.3 1.7 25.2 15.1 Friend or colleagues 626 53.5 2.6 33.4 23.6 Other non-specific bystander 603 43.0 2.2 11.9 5.5 TR, trained responder; LP, Lay person bystander; CPR, cardiopulmonary resuscitation; Good CPC, cerebral performance category 1 or 2; EMS: emergency medical services

Table 5 Multivariable logistic regression analysis for outcomes and trained responder group Outcomes

Group

Total N

Outcome n %

Model 1 Crude OR 95% CI

Model 2* Adjusted OR 95% CI

Good CPC LP TR Survival to discharge LP TR Bystander defibrillation LP TR Bystander CPR LP TR

6063 412

523 55

8.6 13.3

1.00 1.63

1.21

2.20

1.00 1.49

1.04

2.15

6063 412

894 95

14.7 23.1

1.00 1.73

1.36

2.20

1.00 1.59

1.20

2.11

6063 412

95 60

1.6 14.6

1.00 10.71

7.62

15.05

1.00 10.02

7.04

14.26

6063 412

3082 347

50.8 84.2

1.00 5.16

3.94

6.76

1.00 5.31

4.04

6.97

TR, trained responder; LP, Lay person bystander; CPR: cardiopulmonary resuscitation; Good CPC: cerebral performance category 1 or 2 OR, odds ratio; 95% CI, 95% confidence intervals *Adjusting for potential confounders such as age, sex, time of the event, weekend of the event, season, location of the event, and first recorded rhythm.

Table 6 Comparison of effect size of bystander group according to location of cardiac arrest on outcomes using interaction analysis

Outcomes

Private location AOR 95% CI

Public location AOR 95% CI

Good CPC LP 1.00 1.00 TR 1.28 0.78 2.11 1.76 1.06 2.94 Survival to discharge LP 1.00 1.00 TR 1.14 0.74 1.77 2.01 1.41 2.88 Bystander defibrillation LP 1.00 1.00 TR 6.22 3.47 11.17 13.26 8.57 20.52 Bystander CPR LP 1.00 1.00 TR 4.41 2.73 7.11 5.76 4.13 8.03 TR, trained responder; LP, Lay person bystander; CPR: cardiopulmonary resuscitation; Good CPC: cerebral performance category 1 or 2 OR, odds ratio; 95% CI, 95% confidence intervals *Adjusting for potential confounders such as age, sex, time of the event, weekend of the event, season, location of the event, first recorded rhythm, and interaction term (bystander group*location of the event)

30