a b s t r a c t

Purpose: Accidental drowning can cause out-of-hospital cardiac arrest (OHCA). We investigated the effect of drowning location on outcomes of individuals who experienced drowning-OHCA.

Methods: All cases of emergency medical service-treated drowning-OHCA in South Korea from January 2006 to December 2013 were analyzed. Cases were excluded if there was a preceding injury, no information on event location, or suicide. Cases were divided into 4 groups: recreational water with mandatory safety regulations (group 1, public swimming pool; group 2, beach) and nonrecreational water without mandatory safety regulations (group 3, natural freshwater; group 4, seawater). The main outcome was survival to hospital discharge. multiple logistic regression analysis was conducted using natural freshwater as the reference location. Results: We analyzed 1691 drowning-OHCAs (public swimming pools, 3.4%; public beaches, 5.2%; unsupervised seawater, 33.8%; and unsupervised open freshwater, 57.6%). The rate of survival to discharge was 4.6% for all cases, 17.5% for cases in public swimming pools, 9.1% for cases in public beaches, 4.9% for cases in unsupervised seawater, and 3.3% for cases in unsupervised open freshwater (p b 0.01). The adjusted odds ratios (95% confidence intervals [CIs]) for survival relative to natural freshwater were 3.97 (95% CI, 1.77-8.89) for public swimming pools, 2.81 (95% CI, 1.22-6.45) for public beaches, and 1.54 (95% CI, 0.88-2.70) for unsupervised seawater. Conclusion: Individuals who experience drowning-OHCA in public locations with safety regulations had a better rate of survival. There should be improved public awareness of the significantly greater risk of drowning-OHCA in locations that have no safety regulations.

(C) 2016

Introduction

Drowning is the third leading cause of death due to unintentional in- jury, and an estimated 372 000 people worldwide died by drowning in 2012 [1]. Globally, approximately 1.3 million disability-adjusted life years were lost due to premature death or disability by drowning, and this accounts for about 7% of disability-adjusted life years due to all in- juries [2]. Previous research reported that drowning is a major cause

? This study was supported by the National Emergency Management Agency of Korea and the Korea Centers for Disease Control and Prevention. The study was funded by the Seoul Metropolitan City Government (2008) and the Korea Centers for Disease Control and Prevention (2008-2013).

* Corresponding author at: Department of Emergency Medicine, Seoul National Univer- sity Boramae Medical Center, 20 Boramae-ro 5 Gil, Dongjak-Gu, Seoul, 07061, South Korea. Tel.: +82 2 870 2663; fax: +82 2 831 0207.

E-mail addresses: [email protected] (J. Jeong), [email protected] (K.J. Hong), [email protected] (S.D. Shin), [email protected] (Y.S. Ro), [email protected] (K.J. Song), [email protected] (E.J. Lee), [email protected] (Y.J. Lee), [email protected] (K.O. Ahn).

of out-of-hospital cardiac arrest (OHCA) among Young people in many developing and developed countries [3].

There are limited population-based surveillance data on drowning- induced OHCA (drowning-OHCA) and predictors of outcomes. Previous research examined the association of several predictors with outcome, such as presence of a witness, duration of the drowning event, shock- able rhythm, and time until emergency medical service (EMS) treat- ment [4-7]. The inaccessibility of a drowning victim greatly limits resuscitation efforts because of challenges related to recognition, rescue, and basic life support. The recommended guidelines for uniform reporting of data from drowning-OHCA suggest that location of the event can affect outcome [8]. Previous studies reported the locations of drowning events as baseline characteristics, but the sample sizes were too small to test the association with outcome [4,6,7,9-12].

A water safety plan that is reinforced by public policy and regula-

tions may help to prevent drowning, and some countries are developing such plans [1]. However, as far as we know, no studies have examined the effect the different public locations on drowning-OHCA outcomes.

http://dx.doi.org/10.1016/j.ajem.2016.06.008

0735-6757/(C) 2016

This study examined the association between location and drowning-OHCA outcomes in South Korea. We hypothesize that drowning-OHCA cases that occur in locations where water safety regu- lations are enforced have better outcomes than those in nonregulated locations, such as natural waters and nonregulated water spaces.

Methods

Study design and setting

This cross-sectional study used a nationwide OHCA registry in South Korea. Korea has a population of 50 million people who live in 16 prov- inces. The Korean EMS system, a single-tiered response model with a basic-to-intermediate service level of ambulances, is operated by the 16 provincial headquarters of the National Fire Department. All ambu- lance crews can perform cardiopulmonary resuscitation (CPR) at a scene and during transport, and can also provide care comparable to that of an intermediate-level emergency medical technician (EMT) in the United States. This includes administration of intravenous fluids, en- dotracheal intubation, and laryngeal mask airway insertion under direct medical control [13]. However, medications for Advanced cardiac life support are not widely available, and ACLS is only available in hospitals in most areas. The EMTs cannot declare death or stop CPR in the field or during transport unless they are under direct medical super- vision from a physician via telephone in cases of apparent death, such as decapitation, Rigor mortis, decomposition, and dependent lividity. The national protocol of prehospital field management for OHCA in South Korea states that ambulance crews should perform CPR in the field for at least 5 minutes. After delivery of more than 5 minutes of chest compression, EMTs should transport an OHCA victim as soon as possible to the nearest emergency department (ED).

The water resources in Korea are 100 032 km² in area and include

about 358 beaches designated by national and local governments, 62 national rivers and 3776 local rivers according to the National Statistics Office, and 902 approved indoor and outdoor pools according to the Ministry of Homeland. The Water-related Leisure Activities Safety Act classifies activities requiring a mandatory safety program to prevent drowning [14]. All recreational water facilities including swimming pools (indoor and outdoor) and coastal beaches approved by the gov- ernment are required to abide by this Act and to establish mandatory safety programs. For example, a supervising lifeguard must be on site when a beach or swimming pool is open to the public. The Installation and Utilization of Sports Facilities Act requires a minimum number of lifeguards to be present according to water type; a swimming pool re- quires at least 2 lifeguards, whereas a coastal beach requires 5 lifeguards per 300 000 visitors [15]. The Emergency Medical Services Act mandates lifeguards to complete basic life support training, including chest com- pression and rescue ventilation [16]. All lifeguards must complete 4 hours of training every year, which includes CPR and first aid for trauma.

Data source

All data were from a nationwide, population-based observational registry of OHCA patients who were assessed and treated by EMSs in South Korea from 2006 to 2013. The OHCA cases in this registry were collected from ambulance run sheets (completed by EMS providers who assisted with cardiac arrest) and from review of hospital medical records by the Korea Centers for Disease Control and Prevention (CDC). Ambulance run sheets, which include basic information on OHCA cases, are electronically stored on a computer server managed by the National Fire Department. Trained medical record reviewers of the Korea CDC visited all hospitals (approximately 700) that received OHCA patients and reviewed individual medical records to complete the registry with clinical information on hospital care and resuscitation outcomes [17]. A Data quality management committee, which consists of emergency medicine physicians, cardiologists, epidemiologists, and

biostatisticians, reviews these data and other variables collected by the medical record reviewers to maintain data quality through monthly committee meetings. In cases when a variable cannot be determined, the committee provided decisions with medical record reviewers.

Study population

Drowning-OHCA cases from 2006 to 2013 were collected from the registry where external causes for noncardiac etiology were coded. The etiology of a cardiac arrest was determined from review of the med- ical record, mostly recorded by emergency physicians at the ED. Exter- nal causes were classified as trauma, poisoning, burn (flame, chemical, and electrical), hanging, asphyxia, and drowning. If there was any such external event, the cause of the OHCA was regarded as noncardiac. Unless there was a definite external cause, the case was presumed to be of Noncardiac etiology. Drowning-OHCA events were excluded if there was a preceding traffic accident, suicide attempt, missing data on location, missing data on clinical outcome, or if neither EMS nor ED physicians provided treatment.

Data variables

We classified the event location (the main variable) into four categories: (i) recreational public swimming pools with a mandatory regulation program as required by the Installation and Utilization of Sports Facilities Act (public swimming pools); (ii) recreational public beaches with a mandatory regulation program as required by the Installation and Utilization of Sports Facilities Act (public beaches);

(iii) natural locations with freshwater, such as lakes, rivers, creeks, or bayous, that have no mandatory safety regulation program (unsuper- vised open freshwater); and (iv) natural locations with seawater, such as the ocean, that have no mandatory safety regulation program (unsu- pervised seawater).

We collected demographic variables and examined the effect of numerous potential confounders: age, sex, presence of a witness, bystander CPR, urbanization level (metropolis [N 5 million people] vs nonmetropolis), EMS response time (from call to ambulance arrival at the scene), EMS scene time (from arrival to departure), EMS transporta- tion time (from departure to arrival at the ED), primary electrocardio- gram (ECG) type at the scene (ventricular fibrillation or ventricular tachycardia, pulseless electrical activity [PEA], and asystole), prehospital defibrillation, and level of ED (1, 2, or 3). Witnesses and bystanders in- cluded laypersons and lifeguards without specific classification. The pri- mary end point was survival-to-hospital discharge and was measured by medical record review of the Korea CDC. This study was approved by the institutional review board of the Seoul National University Hospi- tal (number: 1103-153-357).

Statistical analysis

Demographic findings were compared between different event loca- tions using the ?² test for categorical variables and the independent t test for continuous variables. A P value less than .05 was regarded as sig- nificant. To estimate the effect of different event locations on study end points, multivariable logistic regression analysis was used to calculate unadjusted and adjusted odds ratios (aORs) and 95% confidence inter- vals (CIs). Two models were developed to measure the effect of drown- ing location on survival to discharge with adjustment for confounders: model 1 adjusted for confounding by variables present before rescue (age, sex, metropolis, and witness); model 2 adjusted for all potential confounders including prehospital variables and the confounders con- sidered in model 1 (primary ECG at the scene, bystander CPR, defibrilla- tion by EMS, response time interval, scene time interval, transport time interval, and level of ED). SAS 9.4 (SAS Institute, Cary, North Carolina) was used for statistical analysis.

Results

Demographic findings

There were 172 558 EMS-assessed OHCAs in South Korea during the 7-year study period. We initially excluded cases classified as nondrowning-OHCA (n = 169 579). Among the 2979 drowning- OHCA cases, we excluded cases that that were at unknown locations (n = 216), were preceded by injury (such as a car accident) near water (n = 121), were due to suicide (n = 381), who received no treat- ment by an EMS (n = 333), had unknown cerebral performance catego- ry outcomes (n = 5), and were at a nonnatural, nonrecreational location (n = 232) (Figure).

Table 1 shows the demographic variables of drowning-OHCAs ac- cording to location. Among the 1691 drowning-OHCAs, 3.4% occurred in public swimming pools, 5.2% in public beaches, 33.8% in unsupervised seawater, and 57.6% in unsupervised open freshwater. Survival-to- discharge rates were 4.6% for all cases, 17.5% for cases in public swim- ming pools, 9.1% for cases in public beaches, 4.9% for cases in unsuper- vised seawater, and 3.3% for cases in unsupervised open freshwater (P b .01). The bystander CPR rates were 7.2% for all cases, 31.6% for cases in public swimming pools, 9.1% for cases in public beaches, 6.5% for cases in unsupervised seawater, and 6.1% for cases in unsupervised open freshwater (P b .01).

Main outcomes

Table 2 shows the results of multivariable logistic regression analy- sis, using unsupervised open freshwater as the reference location. The unadjusted ORs (95% CIs) for survival at discharge were 6.26 (95% CI, 2.91-13.50) for public swimming pools, 2.94 (95% CI, 1.31-6.60) for pub- lic beaches, and 1.52 (95% CI, 0.90-2.54) for unsupervised seawater. In model 1, the adjusted odds ratios (aORs) (95% CIs) for survival were

3.97 (95% CI, 1.77-8.89) for public swimming pools, 2.81 (95% CI, 1.22- 6.45) for public beaches, and 1.54 (95% CI, 0.88-2.70) for unsupervised seawater. In model 2, the aORs (95% CIs) for survival were 1.12 (95%

CI, 0.45-2.82) for public swimming pools, 2.26 (95% CI, 0.92-5.52) for public beaches, and 1.40 (95% CI, 0.76-2.58) for unsupervised seawater.

Discussion

This Nationwide observational study of drowning-OHCA examined the effect of drowning location on patient survival to hospital discharge.

We classified locations as public swimming pools, public beaches, unsu- pervised open freshwater, and unsupervised seawater; the law requires safety regulation for the first 2 locations, but not for the latter 2 loca- tions. Model 1 (which controlled for prerescue confounders) indicated that an individual who experienced downing-OHCA at a supervised lo- cation had a 2.81- to 3.97-fold higher chance of survival in comparison to that in an unsupervised location. However, model 2, which adjusted for all model 1 variables as well as prerescue variables and EMS rescue efforts, indicated no significant differences among the 4 locations. These results indicate that public location may have some effect in im- proving survival from drowning-OHCA. In particular, bystander CPR was significantly associated with the location of the event and may be one of the reasons for the different outcomes of model 1 and model 2. Nonrecreational waters without safety regulation had lower bystander CPR rates than locations where safety regulations were enforced. A re- cent study of OHCA in South Korea showed that the overall bystander CPR rate was 2.1%, much lower than that for public locations of our study [18].

The current Utstein guidelines for OHCA acknowledge the impor- tance of OHCA location, and there may be different subcategories for dif- ferent local communities because knowledge of the locations of cardiac arrest may help a local community to optimize its emergency medical resources [19]. Many previous studies that investigated drowning- OHCA used a simple classification of location, such as home/residence vs public place, in accordance with other studies of the etiology of OHCA [20,21]. Some studies classified drowning location using more de- tailed variables (open water, bath tub, pool, pond or quarry, etc), but the location of the specific event was not the main predictive variable [4,12]. Only a few studies examined the association of drowning location with survival outcomes, and there is limited knowledge of the most effective classification for classification of drowning location on survival from cardiac arrest. Thus, our establishment of a definition of public drown- ing location was also difficult because of lack of previous research on this topic. Cardiac arrest due to drowning always occurs in the water and is associated with water activities; therefore, the purpose and func- tion of the recreational location should be considered when categoriz- ing location. Thus, we classified drowning locations as regulated public locations or as unsupervised locations.

Previous studies of OHCA defined public locations as places with heavy pedestrian flow or gathering places, such as streets, buildings, or work places, rather than as “home.” Because of the relatively high fre- quency of cardiac arrests with Shockable rhythms in public settings, this has motivated the installment of public-access defibrillators [20,21].

Figure. Patient selection flowchart. Abbreviation: CPC, cerebral performance scale.

Table 1

Demographic characteristics of patients who experienced drowning-associated out-of-hospital cardiac arrest in different locations

Variable	Total	Public swimming	pools		Public beaches			Unsupervised seawater			Unsupervised open freshwa	ter	P
	No. %	No.	%		No.	%		No.	%		No.	%
Total	1691 100.0	57	100.0		88	100.0		572	100.0		974	100.0
Sex													b.01
Female	330 19.5	17	29.8		8	9.1		122	21.3		183	18.8
Male	1361 80.5	40	70.2		80	90.9		450	78.7		791	81.2
Age, y													b.01
0-19	332 19.6	31	54.4		20	22.7		33	5.8		248	25.5
20-39	381 22.5	9	15.8		28	31.8		97	17.0		247	25.4
40-59	602 35.6	5	8.8		27	30.7		276	48.3		294	30.2
>= 60	376 22.2	12	21.1		13	14.8		166	29.0		185	19.0
Median (IQR)	44 (24-58)	18 (5-55)			37 (22-52)			51 (41-61)			39 (19-55)
Witness of arrest	432 25.5	17	29.8		25	28.4		140	24.5		250	25.7	.73
Bystander CPR	122 7.2	18	31.6		8	9.1		37	6.5		59	6.1	b.01
Primary ECG at the scene													b.01
VF/VT	25 1.5	2	3.5		1	1.1		10	1.7		12	1.2
PEA	12 0.7	3	5.3		0	0.0		4	0.7		5	0.5
Asystole	1654 97.8	52	91.2		87	98.9		558	97.6		957	98.3
EMS defibrillation	50 3.0	2	3.5		3	3.4		23	4.0		22	2.3	.26
Response time, min													b.01
Median (IQR)	10 (6-18)	7 (4-9)			8 (4-14)			9 (5-15)			12 (7-19)
Scene time, min													b.01
Median (IQR)	10 (5-25)	5 (2-7)			7 (4-17)			10 (4-19)			13 (6-30)
Transport time, min													b.01
Median (IQR)	14 (7-22)	7 (5-12)			15 (7-23)			11 (5-22)			15 (8-23)
Metropolis	310 18.3	19	33.3		17	19.3		151	26.4		123	12.6	b.01
ED level													.19
Level 1 ED	97 5.7	4	7.0		2	2.3		34	5.9		57	5.9
Level 2 ED	453 26.8	19	33.3		28	31.8		167	29.2		239	24.5
Level 3 ED	1141 67.5	34	59.6		58	65.9		371	64.9		678	69.6
Outcome
prehospital ROSC	42 2.5	5	8.8	0		0.0	13		2.3	24		2.5	b.01
Survival to admission	160 9.5	15	26.3	13		14.8	56		9.8	76		7.8	b.01
Survival to discharge	78 4.6	10	17.5	8		9.1	28		4.9	32		3.3	b.01
good neurological recovery	29 1.7	2	3.5	1		1.1	9		1.6	17		1.7	.72

Good CPC is cerebral performance category 1 or 2. Abbreviations: IQR, interquartile range; ROSC, return of spontaneous circulation; VF/VT, Ventricular fibrillation/ventricular tachycardia.

However, the cardiac rhythm during drowning-OHCA is predominantly asystole or PEA, and previous studies have debated the cost- effectiveness of an automated external defibrillator in aquatic settings [3]. Therefore, the definition used for public setting in the case of drowning-OHCA should be different from medical or nontraumatic OHCA. The guidelines for drowning prevention mainly emphasize safety signs, rules, lifeguards, and supervisors [3]. The presence of lifeguards improves the survival rate from unintentional drowning in Korea and the United States [22]. There are strict regulations to ensure safety in recreational waters, such as public beach resorts and swimming pools, in South Korea. These laws require the presence of a rescue ship, lifeboat, lifeguard, or swimming instructor. Therefore, it is reasonable to define public places according to the presence or absence of safety regulations in the setting of this study.

Why did these public places have more survivors? We examined this issue by use of 2 multiple logistic regression models to control for con- founding. Model 1 controlled for variables before the event (age, sex,

metropolis, and presence of a witness); model 2 controlled for all model 1 variables and for variables after the event (bystander CPR, pri- mary ECG at scene, defibrillation by EMS, scene time, transport time, and level of ED). Model 1 showed significantly better survival in super- vised locations than unsupervised locations, but model 2 showed no sig- nificant effect of location. This difference may be explained by the presence of lifeguards in public recreational waters who can per- form CPR and are likely to have been exposed to emergency situa- tions. The presence of a drowning-OHCA in a public location may allow early recognition and activation of the emergency response system, so that earlier EMS treatment can be provided to the pa- tient. Higher proportion of older victims at unsupervised locations as compared with supervised location could also affect the survival. Comorbidity of the elderly could adversely affect the outcome. The different characteristics between OHCA victims related with drowning and cardiac etiology was reported, and this difference could also affect the survival [23].

Table 2

Multivariable analysis of the effect of location of drowning-associated OHCA on survival

						Unadjusted			Model 1a			Model 2b
	Total		Outcomes			OR	95% CI		aOR	95% CI		aOR	95% CI
	No.		No.	%
Survival to discharge	1691		78	4.6
Public swimming pools	57		10	17.5		6.26	2.91-13.50		3.97	1.77-8.89		1.12	0.45-2.82
Public beaches	88		8	9.1		2.94	1.31-6.60		2.81	1.22-6.45		2.26	0.92-5.52
Unsupervised seawater	572		28	4.9		1.52	0.90-2.54		1.54	0.88-2.70		1.40	0.76-2.58
Unsupervised open freshwater	974		32	3.3		1.00			1.00			1.00

^a Model 1: aOR with adjustment for age, sex, metropolis residence, and presence of a witness.

^b Model 2: aOR with adjustment all variables in model 1 and primary ECG at the scene, bystander CPR, defibrillation by EMS, response time, scene time, transport time, and level of ED.

Limitations

In South Korea, all OHCA victims are recorded by a central govern- mental fire department. The data set used in this observational study therefore included all cases that occurred in South Korea over the entire 7-year study period. All variables were collected from ambulance run sheets or hospital records using the Utstein guidelines, but these are not optimized for drowning-OHCA, which has separate reporting guide- lines [8]. The name and location of the body of water are important in- formation, but these information were not available in the national database used in this study. Thus, we determined drowning location retrospectively by examination of the coding used for the drowning scenes and the narrative in the ambulance run sheets. The location was unavailable for about 7% of drowning-OHCA patients, so we exclud- ed these patients from the analysis; however, the number of patients in- cluded in the final analysis was similar to those of previous studies. Ambulance run sheets do not classify witnesses and bystanders into dif- ferent categories, such as laypersons or lifeguards. In the near future, this will not be a limitation because more detailed documents for re- cording of OHCAs were recently introduced in South Korea.

Location was a significant predictor of poor drowning-OHCA outcome, and we assumed that there were differences in safety regulations among lo- cations according to their classification. However, we did not directly mea- sure compliance with safety regulations of the different locations. Some drowning-OHCAs may have occurred in regulated locations when supervi- sion was not present. We also did not measure other important information such as time of submersion and approximate water temperature [8]; some previous research showed that these variables were significantly associated with survival [4,12]. However, the exact duration of submersion is difficult to measure, and the Utstein guidelines recommend treatment of time of submersion and time of removal from water as supplemental information. Our study findings are from single-tiered EMS program with an intermedi- ate level of service. Therefore, generalization of these results to other areas with higher levels of service should be made with caution.

The total delivery time of CPR by the ambulance crew at the scene was relatively short than the advanced countries. In South Korea, delivery of ACLS by the EMT is not permitted by the legislation, and level of service is EMT intermediate. Therefore, the national EMS protocol recommends at least 5 minutes of field CPR at the scene and rapid transport of the victim with CPR during ground Ambulance transport. This practice could associate with poor quality of basic life support. The comorbidities of victims may also affect survival. However, we did not have this information and could not adjust for comorbidities in our multiple logistic regression models.

Conclusions

This nationwide observational study found that individuals who ex- perienced drowning-OHCAs in public locations in South Korea, such as a public swimming pool or a public beach, had better survival to discharge and were more likely to receive CPR from a bystander. Most drowning- OHCAs occur in unregulated locations, and individuals have better out- comes if they experience drowning-OHCA in regulated locations. The most probable explanation for this is that regulated areas provide close supervision, early recognition, rapid rescue, and resuscitation by lifeguards and others who focus safety.

Disclosure

None.

References

1. World Health Organization. Global report on drowning: preventing a leading killer. World Health Organization; 2014.
2. Peden M, McGee K, Sharma G. The injury chart book: a graphical overview of the global burden of injuries. Geneva: World Health Organization; 2002 p.18.
3. Szpilman D, Bierens JJ, Handley AJ, Orlowski JP. Drowning. N Engl J Med 2012; 366(22):2102-10.
4. Quan L, Mack CD, Schiff MA. Association of water temperature and submersion du- ration and drowning outcome. Resuscitation 2014;85(6):790-4.
5. Claesson A, Lindqvist J, Herlitz J. Cardiac arrest due to drowning-changes over time and factors of importance for survival. Resuscitation 2014;85(5):644-8.
6. Nitta M, Kitamura T, Iwami T, Nadkarni VM, Berg RA, Topjian AA, et al. Out-of-hos- pital cardiac arrest due to drowning among children and adults from the Utstein Osaka project. Resuscitation 2013;84(11):1568-73.
7. Youn CS, Choi SP, Yim HW, Park KN. Out-of-hospital cardiac arrest due to drowning: an Utstein style report of 10 years of experience from St. Mary’s Hospital. Resuscita- tion 2009;80(7):778-83.
8. Idris AH, Berg RA, Bierens J, Bossaert L, Branche CM, Gabrielli A, et al. Recommended guidelines for uniform reporting of data from drowning: the “Utstein style”. Circula- tion 2003;108(20):2565-74.
9. Fralick M, Denny CJ, Redelmeier DA. Drowning and the influence of hot weather. PLoS One 2013;8(8):e71689.
10. Dyson K, Morgans A, Bray J, Matthews B, Smith K. Drowning related out-of- hospital cardiac arrests: characteristics and outcomes. Resuscitation 2013;84(8): 1114-8.
11. Claesson A, Lindqvist J, Ortenwall P, Herlitz J. Characteristics of lifesaving from drowning as reported by the Swedish fire and rescue services 1996-2010. Resuscita- tion 2012;83(9):1072-7.
12. Grmec S, Strnad M, Podgorsek D. Comparison of the characteristics and outcome among patients suffering from out-of-hospital primary cardiac arrest and drowning victims in cardiac arrest. Int J Emerg Med 2009;2(1):7-12.
13. 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 ob- servational study from 2006 to 2010 in South Korea. Resuscitation 2013;84(5): 547-57.
14. Water-related leisure activities safety act, South Korea. Available from: http://www. law.go.kr/engLsSc.do?menuId=0&subMenu=5&query=%EC%88%98%EC%83%81% EB%A0%88%EC%A0%80. [Accessed at May 15, 2016].
15. Installation and utilization of sports facilities act, South Korea. Available from: http:// www.law.go.kr/engLsSc.do?menuId=0&subMenu=5&query=%EC%B2%B4%EC% 9C%A1%EC%8B%9C%EC%84%A4#liBgcolor0. [Accessed at May 15, 2016].
16. Emergency medical services act, South Korea. Available from: http://www.law.go. kr/engLsSc.do?menuId=0&subMenu=5&query=%EC%9D%91%EA%B8%89%EC%9D% 98%EB%A3%8C. [Accessed at May 15, 2016].
17. Jacobs I, Nadkarni V, Bahr J, Berg RA, Billi JE, Bossaert L, et al. Cardiac arrest and car- diopulmonary resuscitation outcome reports: update and simplification of the Utstein templates for resuscitation registries. A statement for healthcare profes- sionals from a task force of the international liaison committee on resuscitation (American Heart Association, European Resuscitation Council, Australian Resuscita- tion Council, New Zealand Resuscitation Council, Heart and Stroke Foundation of Canada, InterAmerican Heart Foundation, Resuscitation Council of Southern Africa). Resuscitation 2004;63(3):233-49.
18. Ro YS, Hwang SS, Shin SD, Han D, Kang S, Song KJ, et al. Presumed regional incidence rate of out-of-hospital cardiac arrest in Korea. J Korean Med Sci 2015;30(10): 1396-404.
19. Perkins GD, Jacobs IG, Nadkarni VM, Berg RA, Bhanji F, Biarent D, et al. Cardiac arrest and cardiopulmonary resuscitation outcome reports: update of the Utstein resusci- tation registry templates for out-of-hospital cardiac arrest: a statement for healthcare professionals from a task force of the international liaison committee on resuscitation (American Heart Association, European Resuscitation Council, Australian and New Zealand Council on Resuscitation, Heart and Stroke Foundation of Canada, InterAmerican Heart Foundation, Resuscitation Council of Southern Africa, Resuscitation Council of Asia); and the American Heart Association Emergen- cy Cardiovascular Care Committee and the Council on Cardiopulmonary, Critical Care, Perioperative and Resuscitation. Circulation 2015;132(13):1286-300.
20. Weisfeldt ML, Everson-Stewart S, Sitlani C, Rea T, Aufderheide TP, Atkins DL, et al. Ventricular tachyarrhythmias after cardiac arrest in public versus at home. N Engl J Med 2011;364(4):313-21.
21. Kitamura T, Iwami T, Kawamura T, Nagao K, Tanaka H, Hiraide A, et al. Nationwide public-access defibrillation in Japan. N Engl J Med 2010;362(11):994-1004.
22. Branche CM, Stewart S. Lifeguard effectiveness: a report of the working group. Atlanta, GA: Centers for Diseases Control and Prevention, National Center for Injury Prevention and Control; 2001.
23. Claesson A, Svensson L, Silfverstolpe J, Herlitz J. Characteristics and outcome among patients suffering out-of-hospital cardiac arrest due to drowning. Resuscitation 2008 Mar;76(3):381-7.