a b s t r a c t

Introduction: In South Korea, on-line medical direction using voice calls has been implemented to improve the quality of the emergency medical system. However, in the same, short time span, video will be able to convey more information than by voice. The purpose of this study is to find out if videotelephony-assisted medical direc- tion (VAMD) can change the intervention of the emergency medical technician compared to using conventional voice calls.

Methods: We conducted a prospective study of 312 patients with online medical direction from November 2017 to November 2018. We assisted patients with direct medical direction using conventional voice calls from Octo- ber to November 2017, and then VAMD was implemented from October to November 2018.

Results: From the total number of conventional voice calls, 131 were used for this study, and of the total number of VAMD interventions, 181 were included. There were differences between conventional voice call and VAMD interventions in such types of medical direction as hospital selection (7.6% vs. 36.6%), ECG interpretation (0% vs. 3.4%), and advice on medical techniques (0% vs. 25.1%). The effectiveness of VAMD by survey is greater com- pared to conventional direct medical direction using voice calls (median value, 3.0 vs. 1.5).

Conclusions: The number of instances of medical direction for some interventions, such as interpretation of ECG and advice on medical techniques that did not perform well in conventional voice calls, increased in VAMD. VAMD may play an important role in the Prehospital emergency care.

(C) 2019

Introduction

In South Korea, about 1.4 million people a year avail themselves of the emergency medical service (EMS), and in terms of population share, 1 in 35 people call EMS once per year [1]. The demand for prehospital emergency care and the importance of the emergency med- ical system in this prehospital phase is increasing [2,3]. However, pa- tients sometimes don’t receive proper care due to the long distances and durations, exacerbated by such obstacles as traffic jams, they might face in reaching the hospital. Patients in need of attention for ur- gent Medical issues such as cardiac arrest, especially, require appropri- ate treatment even during the prehospital phase. To solve these problems, it is essential to improve the quality of the EMS system. Active involvement of physicians in all aspects of the EMS system helps to im- prove the quality of prehospital care [4]. The United States of America and European authorities began responding to this problem about ten

* Corresponding authors at: Department of Emergency Medicine, Soonchunhyang University Bucheon Hospital, 170 Jomaru-ro, Bucheon 14584, Republic of Korea.

E-mail addresses: [email protected] (G.W. Kim), [email protected] (Y.H. Lee).

¹ Gi Woon Kim and Young Hwan Lee contributed equally to this work.

years ago by actively holding discussions on measures that would see emergency physicians become more involved in the prehospital emer- gency system as a subspeciality [5].

In South Korea, however, because resources and costs are limited and often lacking, physicians cannot go to the field to conduct medical direction, a limitation that has led Healthcare organizations to supple- ment this lack of prehospital medical care by providing online medical direction using voice calls. Direct medical direction through voice call is effective in improving the quality of patient care, increasing patient resuscitation opportunities, and in guiding and supervising the compo- nents of the various emergency medical service systems [6,7]. The judg- ment of physicians, who conduct medical direction, depend on communication with the on-site emergency medical technician (EMT), but because the communication consists of only voice, there is a limitation in conveying important information systematically and concisely in a Short period of time [8]. To utilize limited resources effec- tively, a more efficient method than the current medical direction via voice call is needed. Like the saying, “A picture is worth a thousand words,” Video calls can more efficiently accommodate the transmission and reception of information in a short period of time. Video calls will also be of great assistance in the real-time identification of a patient’s

https://doi.org/10.1016/j.ajem.2019.06.023

0735-6757/(C) 2019

condition. Since 2011 in South Korea, video calls have become more prevalent as the fourth generation of broadband cellular network tech- nology has rapidly spread throughout the country with the use of smartphones [9]. Koreans now use video calls routinely and with great frequency. The purpose of this study is to find out if videotelephony- assisted medical direction (VAMD) can change the intervention of the emergency medical technician compared to using conventional voice calls.

Materials and methods

Study design and setting

This prospective study was conducted at the request of the National Emergency Management Agency. This study was carried out in a city of South Korea which covers 55.44 km² with the population of 845,000 people. The duration of this study lasted from November 2017 to No- vember 2018.

In Korea, the EMS is a unified public transport system reached by telephone number 1-1-9 and consists of two or three EMTs in an ambu- lance, which is without a doctor [10]. To support this system, direct medical direction using voice calls is practiced by medical physicians certified as EMS directors by the Korean Council of EMS Physicians, a certification that has been offered since 2007 [11]. The city in which we conducted our research experienced the same conditions that char- acterize the EMS system in the country as a whole. The EMTs were trained to provide medical guidance in the following situations: 1) car- diac arrest; 2) major trauma; 3) acute coronary syndrome; 4) acute stroke; 5) death declaration; 6) on-site treatment or procedures re- quired by such conditions as hypoglycemia; and 7) any other situation that EMTs think is in need of medical direction.

This study monitored traditional direct medical direction using con- ventional voice calls from October to November 2017. We developed a medical direction education program using voice calls from December 2017 to May 2018, and then conducted education for EMTs beginning in September 2018. Videotelephony-assisted medical direction (VAMD), our method for employing direct medical direction through video call, was implemented between October and November of 2018. The data for the period between October and November of 2017 were compared with that of October and November of 2018 to delineate the differences between VAMD and voice call medical direction.

VAMD was requested after the EMTs checked the patient’s history. The cell phone used for the VAMD was an Android-based phone (Samsung Galaxy S7). The criteria for the use of VAMD were the same used with traditional direct medical direction through voice call. Three EMTs were sent to the field in an ambulance equipped with a mobile phone that could make video calls. If VAMD was necessary, one EMT would make a video call and hold the phone to show the patient while the other two EMTs treated the patient. If video calls were no lon- ger required, the system switched to voice.

We collected the following data through electronic case report forms: date, time, age, gender, place, floor, total time of medical direc- tion, suspected diagnosis, intervention, quality of call, and effectiveness compared to conventional voice calls. The measures of effectiveness were gauged using a 1-to-5 Likert scale rating system.

We divided the time and day according to specific criteria. Daytime was defined as 6:00 AM to 5:59 PM and nighttime as 6:00 PM to 5:59 AM, as dictated by the duration of sunshine and its absence. Week- days were defined as occurring between 7:00 AM on Monday through 10:59 PM on Friday, and weekends were defined as occurring between 11:00 PM on Friday through 6:59 AM on Monday [12]. Floors were di- vided into high, low, and basement, and those structures with fewer than three floors were defined as low floor. Call quality was divided into two categories: good and poor. Good quality was defined as when the video or voice could provide enough information to perform a direct medical direction. Poor quality was defined as a case in which the

medical direction was not properly relayed due to breaks in the video or voice transmission.

This study was conducted after gaining approval from the Institu- tional Review Board (IRB) at our hospital. Written informed consents were exempted by the IRB. We conducted the study in accordance with the provisions of the Declaration of Helsinki.

Sample size

G*Power software, Version 3.1.9.2 for Windows, was used to calcu- late the sample size. Power (type II error) was set at 0.8 with a type I error of 5% (? = 0.05). The result of the Power calculation showed that a sample size of 130 patients in each group would suffice. We as- sumed that a 5% difference in the change in efficiency of the interven- tion of EMTs would show a statistically significant difference between both groups.

Statistical analysis

IBM SPSS Statistics for Windows, Version 25.0 (IBM Corp., Armonk, NY, USA), was used for statistical analyses. The results of the categorical data were presented via number and frequency. Continuous data were expressed as mean and standard deviation or median and interquartile range. Inter-group differences were evaluated using the independent two-sample t-test, Mann-Whitney U, chi-square test, or Fisher’s exact test, as appropriate. Statistical significance was defined as a value of p b 0.05.

Results

A total of 312 patients, 131 of which were treated through conven- tional voice call and 181 of which received treatment through VAMD, were enrolled in the study period. Two cases of VAMD were excluded because they featured voice calls only, bringing the final number of study cases for VAMD to 179.

The characteristics of the patients treated through VAMD were as follows: The average age was 61.8 years old and 65.9% (n = 116) were men. Of the total number of VAMD patients, 59.2% (n = 106) were treated during the daytime and 71.5% (n = 128) during the week- day. As for the places where medical direction was requested, indoors was the most common location with 62% (n = 114), followed by the ambulance with 25.8% (n = 51) and outdoors with 7.8% (n = 11). De- pending on the height, the low floor was where the most patients were seen at a share of 89.9% (n = 161). The median time for VAMD was 2 min (IQR 1-3). As for good call quality, the video was 88.8% (n

Table 1

General characteristics of subjects.

	2017 voice calls (n = 131)	2018 VAMD (n = 179)	p
Age, years	64.5 +- 20.1	61.8 +- 21.4	0.54
Male, n (%)	78 (65.0)	116 (65.9)	0.34
Daytime, n (%)	73 (55.7)	106 (59.2)	0.54
Weekday, n (%)	91 (69.5)	128 (71.5)	0.70
Place, n (%) Indoor	96 (73.3)	114 (62.0)	0.32
Outdoor	5 (3.8)	11 (7.8)
Ambulance	29 (22.1)	51 (28.5)
Unknown Floor, n (%)	1 (0.8)	3 (1.7)	0.97
High floor ? 3rd floors	6 (4.6)	10 (5.6)
Low floor	120 (91.6)	161 (89.9)
Basement	3 (2.3)	5 (2.8)
Unknown	2 (1.5)	3 (1.7)

Medical direction time, min 2(1-4) 2 (1-3) 0.71

Call quality

Good quality of video, n (%) 159 (88.8)

Good quality of voice, n (%) 129 (98.5) 176 (98.3) 0.99

= 159) and the voice was 98.3% (n = 176). There were no statistical dif- ferences between voice calls and VAMD group. Table 1 provides a break- down of the general characteristics of patients.

After medical direction through VAMD, EMTs surveyed the effective- ness of VAMD compared to conventional direct medical direction using voice calls (Fig. 1). We ascertained if the quality of video and voice changed according to the situation or location where VAMD was re- quested. Good quality calls for video and voice through VAMD were 80% lower when patients were treated in the basement than if they were treated on either high or low floors. There were, however, no sta- tistically different cases according to location, height, or daylight (Table 2).

The characteristics of the 2017 voice calls and the 2018 VAMD were compared and analyzed in the same months of October and November. The number of requests for VAMD increased significantly compared to 2017 when only voice calls were performed (1.9% vs. 2.8%, p = 0.002). There was no statistical difference in gender, daytime, weekday, or suspected diagnosis. However, there were some differences in inter- ventions, as hospital selection increased from 7.6% to 36.3% (p b 0.001). There was no request for electrocardiogram (ECG) interpretation or ad- vice on the medical technique in 2017, however, 3.4% (n = 6), 25.1% (n

= 45), respectively by VAMD in 2018 (Table 3).

Discussion

Our research has shown that the number of instances of medical di- rection, hospital selection, ECG analysis, and advice on medical tech- niques increased in VAMD as compared to conventional voice calls. Also, the effectiveness of VAMD by survey was better than conventional medical direction using voice calls.

The medical direction, or, in other words, medical oversight, is rec- ognized as one of the essential elements of EMS system [6,7,13]. In much of the previous research, the roles of medical direction in EMS have been identified as essential in terms of effectiveness [14,15]. Med- ical direction is divided into two categories: direct and indirect [16]. Most of the time in the EMS system, both are implemented in parallel. Direct medical direction is real-time, physician-directed care through an online channel, and indirect medical direction is performed offline, such as education. If any of these are done well, the quality of EMS will be improved. Lee et al. [17] have reported that well-conducted indi- rect medical direction can improve the quality of the EMS. In this study,

Fig. 1. The median value of effectiveness through survey.

Table 2

Quality of calls according to situa	tion for VAMD.
	Outdoor	Indoor	Ambulance	p
Good quality of video, n (%)	12 (85.7)	99 (89.2)	45 (88.2)	0.92
Good quality of voice, n (%)	14 (100)	109 (98.2)	50 (98)	0.87
	High floor	Low floor	Basement	p
Good quality of video, n (%)	9 (90)	143 (88.8)	4 (80)	0.82
Good quality of voice, n (%)	10 (100)	159 (98.8)	4 (80)	0.09
Daytime Nighttime p
Good quality of video, n (%)	94 (88.7)		65 (89)	0.94
Good quality of voice, n (%)	104 (98.1)		72 (98.6)	0.80

VAMD, videotelephony-assisted medical direction.

video calls were used to maximize the effectiveness of direct medical di- rection instead of conventional voice calls. Vision is the main sense, which accounts for up to 80% of perception, learning, and cognition [18]. Therefore, it is much more efficient to communicate information using both eyesight and hearing, and not solely hearing.

Some argue that direct medical direction is unnecessary because it increases the length of stay time at the site and should be implemented only in selected cases [19,20]. In this study, VAMD took only 2 min (me- dian value). Wuerz et al. [16] reported that it took about 4 min for direct medical direction using voice calls only, compared to our study, which took a shorter time. This may be because the smartphone screen simply shows situations that must be explained only in words using voice calls. Thus, VAMD can be used more quickly and efficiently with patients than conventional voice calls, and duration time on site is not significant problem.

In our study, the effectiveness of VAMD was higher than conven- tional voice calls. Doctors are afforded greater confidence in diagnosing and treating the patient’s condition because they can see on the screen what he had to imagine with words before. Also, from the standpoint of the EMTs, it was easy to communicate by showing the patient’s condi- tions right away in the event of a situation where it was difficult to ac- curately describe the state of patients. The quality of voice and video of VAMD tends to deteriorate below ground level, even though there

Table 3

Comparison of conventional voice calls in 2017 and VAMD in 2018.

	2017 voice calls	2018 VAMD	p
Medical direction cases, n (%)	131 (1.9)	179 (2.8)	0.002
Male, n (%)	78 (65.0)	118 (65.9)	0.87
Daytime, n (%)	53 (40.5)	73 (40.8)	0.95
Weekday, n (%)	81 (62.3)	128 (71.5)	0.09
Cause, n (%) Cardiac arrest	66 (50.4)	86 (48.0)	0.99
Cardiovascular disease	14 (42.4)	19 (10.6)
Trauma	11 (8.4)	15 (8.4)
Hypoglycemia	14 (10.7)	20 (11.2)
Stroke or seizure	12 (9.2)	19 (10.6)
Pulmonary disease	8 (6.1)	12 (6.7)
Intoxication	1 (0.8)	1 (0.6)
Septic shock	1 (0.8)	2 (1.1)
Abdominal disease	1 (0.8)	2 (1.1)
Etc. Intervention, n (%)	3 (2.3)	3 (1.7)
Death declaration	46 (35.1)	53 (29.6)	0.30
Advanced airway	29 (22.1)	30 (16.8)	0.24
IV line insertion	31 (23.7)	40 (22.3)	0.79
Medication	10 (7.6)	25 (14.0)	0.08
Diagnostic assessment	24 (18.3)	18 (10.1)	0.04
Hospital selection	10 (7.6)	65 (36.3)	b0.001
Cardiopulmonary resuscitation	20 (15.3)	29 (16.2)	0.82
Refusal of transport	5 (3.8)	9 (5.0)	0.61
ECG interpretation	0 (0)	6 (3.4)	b0.001
Advice on medical technique	0 (0)	45 (25.1)	b0.001

VAMD, videotelephony-assisted medical direction.

were no statistical differences in the quality of calls. The findings indi- cate this is because the quality of both voice and video was poor in one out of five calls (20%) for VAMD below ground level. However, not only the video, but also voice deteriorated underground, which is more of an online direct medical direction problem than VAMD.

In this study, the number of VAMD requests increased at a statisti- cally significant higher rate than conventional voice calls. This is due to the burden that VAMD reduces for EMTs when attempting to explain a patient’s condition by showing rather than telling. There were no dif- ferences between VAMD and conventional voice calls in suspected diag- nosis. However, the selection of hospitals, advice on medical techniques, and ECG readings were statistically different. In particular, the interpre- tation of ECG and medical advice for procedures such as advanced air- way procedures were not possible in the past with voice calls (Fig. 2).

There are some limitations to this study. First, the sample size was relatively small. This study was conducted in a single city and for a Short duration. Also, our study results may be too limited to be general- ized. For stronger evidence, well-designed, prospective, and large popu- lation studies will be demanded in the future. Second, we have not checked patient satisfaction with their prognosis. However, Raaber et al. [21] reported that patients felt a sense of stability and high satisfac- tion with direct medical direction using the telephone. Third, VAMD is difficult to use when there are only two EMTs, only one EMT is left to manage the patient, or when the EMTs are unable to hold the mobile

Fig. 2. Analyzing electrocardiogram through VAMD.

phone due to the transfer situation of the patient via gurney. Therefore, it is deemed necessary to develop VAMD exclusive equipment in the fu- ture. Fourth, it is difficult to generalize VAMD around the world because there are no assurances regarding the internet speed and quality. Ac- cording to Akamai’s State of the Internet [22], the average connection speed in South Korea is 28.6 Mbps, which is about four times faster than the international average of 7.2 Mbps. Therefore, in countries with poor internet quality, there may be restrictions in the implementa- tion of VAMD. However, new research (e.g., VAMD) will be needed in countries with an EMS system where only the EMTs are called to the field without the presence of physicians. Despite the above limitations, this study is the first study on direct medical direction using video calls.

Conclusion

The number of instances of medical direction for some interventions, such as interpretation of ECG and advice on medical techniques that did not perform well in conventional voice calls, increased in VAMD. Also, VAMD showed a higher effectiveness through survey than that of con- ventional direct medical direction using only voice calls. Therefore, VAMD may play an important role in the prehospital emergency care.

Funding

This work was supported by the National Emergency Management Agency under Grant 201809013F3-00.

Declaration of Competing Interest

The authors have no conflicts of interest to report.

Acknowledgment

This work was supported by the Soonchunhyang University Re- search Fund.

References

1. Kim CS, Kim HH, Kim GW, Lee SK, Kim ST, Choi S, et al. Analysis of the 119 emer- gency situation control center usage including dispatcher-assisted cardiopulmonary resuscitation instructions. J Kor Soc Emerg Med 2018;29(4):350-7.
2. Pittet V, Burnand B, Yersin B, Carron P-N. Trends of pre-hospital emergency medical services activity over 10 years: a population-based registry analysis. BMC Health Serv Res 2014;14(1):380.
3. Veser A, Sieber F, Gross S, Pruckner S. The demographic impact on the demand for emergency medical services in the urban and rural regions of Bavaria, 2012-2032. J Public Health 2015;23(4):181-8.
4. Studnek JR, Fernandez AR, Margolis GS, O’Connor RE. Physician medical oversight in emergency medical services: where are we? Prehosp Emerg Care 2009;13(1):53-8.
5. Coats T, Davies G. Prehospital care for road traffic casualties. BMJ 2002;324(7346): 1135.
6. Alonso-Serra H, Blanton D, O’Connor RE. Physician medical direction in EMS. Prehosp Emerg Care 1998;2(2):153-7.
7. Polsky S, Krohmer J, Maningas P, McDowell R, Benson N, Pons P. Guidelines for med- ical direction of prehospital EMS. Ann Emerg Med 1993;22(4):742-4.
8. Jung SH, Jeong J, Chung JY, Yun YH, Lee JH. Communication for medical advices be- tween Prehospital providers and physician medical directors. J Kor Soc Emerg Med 2015;26(5):430-6.
9. Lee G. The effects of the 4G service on the usage of main communication services. Kor Telecom Policy Rev 2016;23(2):1-16.
10. Park HA, Ahn KO, Shin SD, Cha WC, Ro YS. The effect of emergency medical service use and inter-hospital transfer on prehospital delay among ischemic stroke patients: a Multicenter observational study. J Korean Med Sci 2016;31(1):139-46.
11. Park KJ, Yun YH, Jeong J, Lee JH, Chung JY, Han SK, et al. Efficacy and necessity of the certification program for emergency medical system directors on the actual on-line medical direction. J Kor Soc Emerg Med 2012;23(4):449-54.
12. Peberdy MA, Ornato JP, Larkin GL, Braithwaite RS, Kashner TM, Carey SM, et al. Sur- vival from in-hospital cardiac arrest during nights and weekends. Jama 2008;299 (7):785-92.
13. Delbridge TR, Bailey B, Chew JL, Conn AK, Krakeel JJ, Manz D, et al. EMS agenda for the future: where we are… where we want to be. Prehosp Emerg Care 1998;2(1): 1-12.
14. Millin MG, Brown LH, Craven CK, Hawkins SC, Tan DK, Piazza GM, et al. Evolution of the literature identifying physicians’ roles in leadership, clinical development, and

    practice of the subspecialty of emergency medical services. Prehosp Disaster Med 2011;26(1):49-64.

    Munk M-D, White SD, Perry ML, Platt TE, Hardan MS, Stoy WA. Physician medical di- rection and Clinical performance at an established emergency medical services sys- tem. Prehosp Emerg Care 2009;13(2):185-92.

15. Wuerz RC, Swope GE, Holliman CJ, Vazquez-de Miguel G. On-line medical direction: a prospective study. Prehosp Disaster Med 1995;10(3):174-7.
16. Lee JT, Park SH, Lee CH, Kim YK, Lee CJ, Je DW, et al. Effectiveness of indirect medical oversight on the quality of prehospital emergency care by emergency medical tech- nicians. J Kor Soc Emerg Med 2013;24(6):684-93.
17. Ripley DL, Politzer T. Vision disturbance after TBI. NeuroRehabilitation 2010;27(3): 215.
18. Erder MH, Davidson SJ, Cheney RA. On-line medical command in theory and prac- tice. Ann Emerg Med 1989;18(3):261-8.
19. Pointer JE, Osur MA. Effect of standing orders on field times. Ann Emerg Med 1989; 18(10):1119-21.
20. Raaber N, Botker MT, Riddervold IS, Christensen EF, Emmertsen N-C, Grofte T, et al. Telemedicine-based physician consultation results in more patients treated and re- leased by ambulance personnel. Eur J Emerg Med 2018;25(2):120-7.
21. Belson D, Thompson J, Sun J, Moller R, Sintorn M, Huston G. akamai’s [state of the in- ternet] Q1 2017 report. Akamai; 2017https://www.akamai.com/fr/fr/multimedia/ documents/state-of-the-internet/q1-2017-state-of-the-internet-connectivity- report.pdf, Accessed date: 3 May 2019.