a b s t r a c t

Objectives: This study was designed to investigate whether an audiovisual feedback (AVF) device is ben- eficial for quality retention of chest compression after repetitive practices (RP).

Methods: After completion of a 45-min CC-only cardiopulmonary resuscitation (CPR) training, partici- pants performed 3 sessions of practices on days 1, 3, and 7 under the guidance of an instructor with (RP + AVF) or without (RP) the AVF device. CC quality was determined after each session and was retested at 3 and 12 months.

Results: In total, ninety-seven third year university students participated in this study. CC quality was improved after 3 sessions in both the RP and RP + AVF groups. Retests at 3 months showed that the pro- portions of appropriate CC rate and correct hand position were significantly decreased in the RP group as compared with the last practice (p < 0.05). However, no significant changes in CC quality were observed in the RP + AVF group. However, the proportions of appropriate CC rate, depth, and complete recoil were significantly decreased after 12 months in both RP and RP + AVF groups (p < 0.05). There were no signif- icant differences in these parameters between the RP and the RP + AVF groups at 12 months after RP. Conclusion: With RP, the use of an AVF device further improves initial CC Skill acquisition and short-term quality retention. However, long-term quality retention is not statistically different between rescuers who receive verbal human feedback only and those who receive additional AVF device feedback after RP.

(C) 2019

Introduction

Out-of-hospital cardiac arrest (OHCA) continues to be a major cause of death in many countries of the world [1]. Nearly 1 in 1000 people every year will succumb to an OHCA, with a mean sur- vival rate at hospital discharge of only 10.6% [2]. Timely and cor- rect CPR is essential for patient prognosis and neurological rehabilitation, which could double or triple the likelihood of sur- vival after OHCAs [3,4]. The combination of rescue ventilation and CC is known as conventional CPR, whereas untrained lay res- cuers can also provide CC-only CPR for OHCA patients [5,6]. How- ever, the prevalence of OHCA bystander CPR is variable–in some regions as low as 10% [7]. In addition, the problem of poor reten- tion of CPR skills has been widely demonstrated in both health care

* Corresponding authors.

E-mail addresses: [email protected] (X.-Q. Jin), [email protected] (Y. Zhao).

¹ These authors contribute equally.

providers and the general public [8-10]. It was found that CPR skill retention significantly declines over time, and CC and rescue breathing quality were inadequate at 6 or 12 months after training. Therefore, it is essential to identify some training techniques to improve the retention of CPR skills. Current evidence suggests that an AVF device is efficient for improving CPR training quality and skills retention [11]. However, Feedback devices are not available in many undeveloped regions, whereas repetitive sessions of addi- tional practice have been proven to be effective in maintaining long-term CPR quality [12]. Thus, repetitive CPR practice would be a possible option for the improvement of quality retention when high-cost devices are unavailable. However, whether the effects of RP and AVF are comparable in bystander CPR training and long- term skills retention is unclear. Therefore, this study was designed to compare the training quality and long-term retention of CC skills by bystanders between those receiving RP with verbal human feedback and those receiving RP with real-time AVF device.

https://doi.org/10.1016/j.ajem.2019.04.025

0735-6757/(C) 2019

Methods

Participants and setting

This randomized controlled study was conducted between July 18, 2017, and July 24, 2018, at Zhongnan Hospital of Wuhan University (Wuhan, Hubei, China). Third-year medical students were recruited to participate in this study. Each participant signed an informed consent form prior to participation. The characteris- tics of participants, including age, sex, height, weight, number of previous CPR trainings, and the experience of actual CPR, were recorded. Participants with underlying health problems were excluded. This study was approved by the Ethics Committee of Zhongnan Hospital of Wuhan University, and all procedures were performed according to the Declaration of Helsinki.

Study design

Training

An American Heart Association (AHA) basic life support-certi- fied instructor performed a 45-minute CC-only CPR training pro- gram that included the following: (1) instruction and practice of Cardiac arrest recognition and calling 1-2-0 (emergency call in China) for help, (2) instruction and practice of CC, and (3) use of an Automated external defibrillator . The instructor empha- sized the characteristics of high-quality CCs, such as adequate compression rate, depth, recoil, and hand position according to the 2015 AHA CPR guidelines [11]. Each instructor performed training program for 8 to 10 participants, and all instructors were blinded to group allocation and did not participate in the following phases. During training, instructors were asked to ensure each par- ticipant perform 5 cycles (2 min/cycle) of CC on manikins.

Grouping

In total, 97 participants participated in this study. The use of real-time AVF devices for participants during CC-only CPR was assigned randomly using a card in a sealed envelope. If ”A” was on the card in the envelope, participants performed repetitive ses- sions of CC-only CPR under the guidance of the instructor (RP group, n = 49). If ”B” appeared on the card, participants performed repetitive sessions of CC-only CPR under the guidance of the instructor and the real-time AVF devices (RP + AVF group, n = 48).

CPR practice

CPR sessions were practiced on an adult Laerdal Resusci Anne QCPR torso manikin (Laerdal China Ltd., Hangzhou, China). The SimPad PLUS was used as an AVF device in the present study. This device is a tablet which can directly reflect the CC rate, CC depth, chest recoil, correct hand position, time of CC interruption and ven- tilation volume. The Laerdal SimPad PLUS was connected to the QCPR manikin and automatically recorded the following variables: CC rate, depth, percentage of compression with complete chest recoil, percentage of compression with correct hand location posi- tion, and the proportions of appropriate CC rate and depth. Since we performed CC-only CPR in this study, only those parameters regarding compressions have been evaluated for quality assess- ment. The repetitive sessions of CPR were conducted on day 1 (first session, day of CPR training), day 3 (second session), and day 7 (third session). In each session, participants first practiced CC under the supervision of instructors for a total of 30 min. One hour later, each participant performed a 2-min CC, and the quality of compressions was determined. The study protocol is summarized in Fig. 1.

Outcome measures

Primary outcome measures: the assessments of short- and long-time CC quality retention were performed at 3 months and 12 months after training. The variables of CC quality including appropriate CC rate, depth, correct hand position, and complete chest recoil were recorded by the SimPad PLUS. Comparison of resuscitation skills between different time points was used for the assessment of CC quality retention. Secondary outcome mea- sures: the real-time CC quality assessment were performed at 1 h after a 30 min-practice in each session. The appropriate CC rate, depth, correct hand position, and complete chest recoil were recorded by the SimPad PLUS.

Statistical analysis

The data were expressed as numbers, percentages, and mean +- standard deviation (SD). Comparisons across groups were performed using the chi-square test for categorical variables and Student’s t-test for continuous variables. CPR skills at RPs, 3 months, and 12 months after training were compared using Two-way Repeated Measures ANOVA. On the basis of our prelimi- nary trial related to the CC quality retention by RP, we estimated that the addition of AVF device would result in 10% absolute increase in appropriate CC proportion (from 65% to 75%). We calcu- lated that it would require at least 37 participants in each group to have a statistical power of 80% to detect the estimated difference under a two-sided alpha level of 0.05. We chose to include at least 20% more participants, anticipating some attrition. All statistical analysis was performed using commercial software (SPSS version 20, SPSS Inc., Chicago, IL). A two-tailed p value <0.05 was consid- ered statistically significant.

Results

Characteristics of participants

Among 97 participants, 48 were randomized to the RP group and 49 were randomized to the RP + AVF group. All participants were followed up for 12 months after RP. The demographic data of participants in the two groups are summarized in Table 1. The proportion of males was 42.9% in the RP group and 44.9% in the RP + AVF group (p = 0.875). The body mass index in the RP group was insignificantly higher than that in the RP + AVF group (21.1 +- 2.9 vs. 20.4 +- 2.2 kg/m², p > 0.05). In addition, there were no significant differences in age, height, or weight between groups (p > 0.05, respectively). We also documented the number of previ- ous CPR trainings and actual CPR for each participant, and we found only 1 participant in the RP + AVF group experienced actual CPR. There were no significant differences in previous CPR training between the RP and RP + AVF group (35.4 vs. 40.8%, p = 0.677).

CPR skills after each session of practices

The CC performance of three sessions are summarized in Table 2. In either the RP or the RP + AVF group, significant differ- ences were observed in compression depth but not in compression rate between sessions. The proportions of appropriate compression rate and depth were significantly increased after RP in both the RP and RP + AVF groups (p < 0.05, respectively). The proportion of cor- rect hand position remained unchanged after RP, and there were no significant differences in correct hand position between the RP and the RP + AVF group at all sessions (p > 0.05, respectively). Participants in the RP + AVF group showed a high level of appropri- ate chest recoil, which was significantly increased as compared

Fig. 1. Participant flow. RP = repetitive practice; AVF = audiovisual feedback; CPR = cardiopulmonary resuscitation; CC = chest compressions.

Table 1

Characteristics of participants.

Characteristic	RP group	RP + AVF group	p value
	(n = 48)	(n = 49)
Male	21 (42.9)	22 (44.9)	0.873
Age (year)	21 +- 1	21 +- 1	0.412
Height (cm)	166.9 +- 7.6	167.0 +- 7.2	0.934
Weight (kg)	59.0 +- 10.6	57.2 +- 8.6	0.382
BMI (kg/m2)	21.1 +- 2.9	20.4 +- 2.2	0.228
Previous CPR training	17 (35.4)	20 (40.8)	0.677
Experience of actual CPR	0 (0)	1 (0.02)	1.000

RP = repetitive practice; AVF = audiovisual feedback; BMI = body mass index. CPR = cardiopulmonary resuscitation. Data are expressed as number (percentage) or mean +- SD.

with that in the RP group at each session (p < 0.001, respectively). Although the proportion of appropriate chest recoil in the RP group was slightly increased after RP, the differences were insignificant. Together, these results suggested that repetitive sessions of prac- tice can improve CC quality. In addition, the AVF devices are effec- tive for enhancing CC skill acquisition, which may shorten the training period of bystanders.

Quality retention of CC skills at 3 and 12 months

After 3 months, 49 participants in the RP group and 48 partici- pants in the RP + AVF group were retested (Table 3). In the RP group, the proportions of correct hand position (p = 0.042) and appropriate CC rate (p < 0.001) were significantly decreased after

3 months as compared with the last session of practice. No signif- icant differences in the proportion of appropriate compression depth (p = 0.275) and complete chest recoil (p = 0.054) were observed between the last practice and the retest at 3 months in the RP group. In addition, the proportions of appropriate CC rate, depth, correct hand position, and complete chest recoil were slightly decreased after 3 months as compared with the last practice in the RP + AVF group. However, the differences were insignificant (p > 0.05, respectively). These results suggested that the RP + AVF group had better quality retention of CC as compared with the RP group.

CC skills were retested for all participants at 12 months after the last session of practice. As seen in Table 4, the quality of CCs was apparently decreased after 12 months in both the RP and the RP + AVF groups. The proportions of correct hand position (p = 0.001) and appropriate CC rate (p < 0.001) in the RP group were further decreased as compared with the last practice. In addi- tion, the differences in the proportions of appropriate CC depth (p = 0.039) and complete chest recoil (p = 0.001) were also signifi- cant between the last practice and the retest results at 12 months. Moreover, the CC quality in the RP + AVF group was also signifi- cantly decreased after 12 months as compared with the last prac- tice. In detail, the proportions of appropriate CC rate (p = 0.001), depth (p = 0.002), correct hand position (p = 0.019), and complete chest recoil (p = 0.012) were significantly decreased at 12 months as compared with practices. These results demonstrated significant decreases in CC quality for both the RP and RP + AVF groups.

We also performed a comparison of CC quality between the RP

and the RP + AVF groups. As seen in Table 5, our results demon- strated significant differences in Compression quality between

Table 2

chest compression quality after each session (mean +- SD).

RP (n = 49) RP + AVF (n = 48)

Compression rate (cpm)

1st session (d1)	112 +- 19	113 +- 17	n.s.
2nd session (d3)	107 +- 12	112 +- 16
3rd session (d7)	111 +- 16	112 +- 12

Compression depth (mm)

p value¹

the short-term (3 months) quality retention but failed to maintain high-quality CCs after 12 months.

The 2015 Guidelines for Cardiopulmonary Resuscitation recom- mended that rescuers, whether laypersons and health care provi- ders, compress the adult chest with adequate rate, depth, recoil, and hand position and attempt to minimize the frequency and duration of interruptions [13]. It is difficult to judge whether these criteria are met, even by an experienced instructor [14]. Our results

1st session (d1) 50.6 +- 6.4 51.7 +- 4.4 n.s.

2nd session (d3) 53.9 +- 5.3^*** 54.0 +- 4.2^*

3rd session (d7) 55.2 +- 4.4^*** 54.3 +- 3.2^* Correct hand position (%)

1st session (d1)	95.7 +- 16.8	98.7 +- 3.9	n.s.
2nd session (d3)	96.8 +- 12.5	99.3 +- 2.6
3rd session (d7)	97.4 +- 14.5	99.9 +- 0.3
Appropriate compression rate (%)
1st session (d1)	68.8 +- 29.6	77.7 +- 27.2	0.012
2nd session (d3)	76.0 +- 31.1*	82.0 +- 25.4*	0.034
3rd session (d7)	86.7 +- 20.5***,-	87.5 +- 20.9**,-	n.s.
Appropriate compression depth (%)

suggested that RP improved CC quality through increasing the pro- portion of appropriate rate and depth. These results were in line with a previous study that already demonstrated that repetitive sessions of formative self-testing are effective for refreshing CPR skills [15]. Importantly, the increases in the proportion of CC rate and depth were more apparent in the RP group as compared with the RP + AVF group. These results indicated that RP is an acceptable method for the improvement of CC quality when high-cost AVF devices are unavailable. However, in the absence of AVF devices, 3 repetitive sessions of practice under the guidance of an instructor

3rd session (d7) 69.2 +- 27.8^**,- 76.9 +- 20.7^**,– 0.003

1st session (d1)	55.6 +- 31.9	67.7 +- 29.6	0.002	were insufficient to maintain high-quality CCs. The proportions of
2nd session (d3)	64.6 +- 33.4*	70.6 +- 27.0	0.017	correct CC rate and hand position were markedly decreased after

Complete chest recoil (%)

1st session (d1) 82.2 +- 25.8 93.7 +- 10.6 <0.001

2nd session (d3) 82.9 +- 24.6 94.7 +- 10.0 <0.001

3rd session (d7) 84.0 +- 18.5 95.9 +- 8.0 <0.001

Appropriate chest compression rate, depth, correct hand position, and complete chest recoil were defined according to the 2015 American Heart Association (AHA) guideline. CC = chest compression; RP = repetitive practice; AVF = audiovisual feedback.

¹ RP vs. RP + AVF group, n.s. = not significant.

^* p < 0.05.

^** p < 0.01.

^*** p < 0.001 vs. the 1st session.

^– p < 0.05.

^— p < 0.01 vs. the 2nd session.

these two groups at 3 months after repetitive sessions of practice. In the RP + AVF group, the proportions of appropriate CC rate (p = 0.028), depth (p = 0.005), correct hand position (p = 0.028), and complete chest recoil (p = 0.011) were significantly higher than that in the RP group. However, there were no significant dif- ferences in these parameters between the RP and RP + AVF groups at 12 months (p > 0.05, respectively).

Discussion

This randomized controlled trial is designed to compare the long-term retention of CC skills by bystanders between those receiving verbal human feedback only and those receiving verbal human feedback + real-time AVF device training. The major find- ings of this study can be summarized as follows: (1) repetitive ses- sions of practice were able to improve CC quality, (2) participants performed higher quality of CC under the guidance of a real-time AVF device, and (3) the use of a real-time AVF device increased

3 months, and all quality parameters, including the proportions of CC rate, depth, correct hand position, and complete chest recoil, were sharply decreased after 12 months. Together, our results indi- cated that RP can improve CC skills but fails to maintain long-term quality.

During the past decade, the real-time automated Feedback devices such as Zoll PocketCPR^(R), Laerdal CPRmeter^(R), iPhone app Zoll Pocket CPR^(R) have been introduced to enhance both training and performance of CPR [16,17]. They generally provide feedback based on direct assessment of current parameters of CPR and their real-time visualization through a software interface. Trainees may therefore read the current CPR parameters and eventually adopt

corrective measure to promote a better agreement with guidelines recommendations. Those real-time AVF devices generally work with high-cost manikins. In fact, a number of simple, low-cost feedback devices have also been implemented for CPR training [18-20]. In this study, we used a high-cost Laerdal QCPR manikin and SimPad PLUS tablet as AVF device, which are rarely used for public CPR training. Anyhow, almost all kinds of feedback devices were applied in the clinical research and teaching, and it has been demonstrated that CPR training with a real-time AVF device is bet- ter than traditional teaching methods [18-21]. In addition, when applied in learning strategies involving different health care pro- fessionals and students, these devices improved CPR acquisition, retention, and adherence to parameters recommended by guideli- nes [22]. However, with deeper research, some problems were also found. There are only a few nonrandomized studies on the use of such devices during laypersons’ training, and their results are con- troversial. Noemi Pavo randomized 370 medical students in a two- rescuer scenario comparing the Quality of CPR with two different feedback methods. Receipt of verbal human feedback (acquired in a short training session) was not inferior to a mechanical audio- visual feedback prompt device [23]. Another study demonstrated no significant differences in any of the parameters between the

Table 3

Chest compression quality retention after 3 months.

RP (n = 49) Difference (95% CI) p value RP + AVF (n = 48) Difference (95% CI) p value

	Practicea	3 months			Practicea	3 months
Correct hand position (%)	97.4 +- 14.5	93.4 +- 19.9	–8.9 (–16.5, –1.4)	0.042	99.9 +- 0.3	98.6 +- 3.6	–1.5 (–3.0, 0.4)	0.271
Appropriate CC rate (%)	86.7 +- 20.5	65.4 +- 38.2	–22.1(–30.3, –13.9)	<0.001	87.5 +- 20.9	77.1 +- 25.9	–9.0 (–20.6, 2.6)	0.112
Appropriate CC depth (%)	69.2 +- 27.8	64.8 +- 29.1	–6.8 (–19.3, 5.6)	0.275	76.9 +- 20.7	76.9 +- 21.9	–1.2 (–2.1, 1.7)	0.109
Complete chest recoil (%)	84.0 +- 18.5	81.9 +- 16.5	–5.48 (–12.1, 0.2)	0.054	95.9 +- 8.0	92.1 +- 6.2	–4.3 (–12.2, 6.9)	0.242

^a The last session of practice. Appropriate chest compression rate, depth, correct hand position, and complete chest recoil were defined according to the 2015 American Heart Association (AHA) guideline. CC = chest compression; RP = repetitive practice; AVF = audiovisual feedback.

Table 4

Chest compression quality retention after 12 months.

RP (n = 49) Difference (95% CI) p value RP + AVF (n = 48) Difference (95% CI) p value

	Practicea	12 months			Practicea	12 months
Correct hand position (%)	97.4 +- 14.5	90.1 +- 25.4	–12.5 (–19.9, –5.2)	0.001	99.9 +- 0.3	96.8 +- 9.2	–3.1 (–6.2, –0.3)	0.019
Appropriate CC rate (%)	86.7 +- 20.5	62.9 +- 30.8	–23.9 (–35.6, –11.8)	<0.001	87.5 +- 20.9	68.4 +- 29.3	–15.4 (–24.5, –6.5)	0.001
Appropriate CC depth (%)	69.2 +- 27.8	61.9 +- 28.6	–11.2 (–23.3, –0.6)	0.039	76.9 +- 20.7	63.2 +- 26.8	–14.2 (–19.6. -7.9)	0.002
Complete chest recoil (%)	84.0 +- 18.5	80.5 +- 19.4	–7.8 (–11.9, –4.6)	0.001	95.9 +- 8.0	83.5 +- 14.4	–13.1 (–22.4, –5.4)	0.012

^a The last session of practice. Appropriate chest compression rate, depth, correct hand position, and complete chest recoil were defined according to the 2015 American Heart Association (AHA) guideline. CC = chest compression; RP = repetitive practice; AVF = audiovisual feedback.

Table 5

Chest compression comparison between RP and RP + AVF groups (mean +- SD).

	3 months				12 months
	RP	RP + AVF	p value		RP	RP + AVF	p value
Correct hand position (%)	93.4 +- 19.9	98.6 +- 3.6	0.028		90.1 +- 25.4	96.8 +- 9.2	0.051
Appropriate CC rate (%)	65.4 +- 38.2	77.1 +- 25.9	0.028		62.9 +- 30.8	68.4 +- 29.3	0.248
Appropriate CC depth (%)	64.8 +- 29.1	76.9 +- 21.9	0.005		61.9 +- 28.6	63.2 +- 26.8	0.122
Complete chest recoil (%)	81.9 +- 16.5	92.1 +- 6.2	0.011		80.5 +- 19.4	83.5 +- 14.4	0.270

Appropriate chest compression rate, depth, correct hand position, and complete chest recoil were defined according to the 2015 American Heart Association (AHA) guideline. CC = chest compression; RP = repetitive practice; AVF = audiovisual feedback.

group receiving short feedback (1-minute training with real-time visual feedback) and the group receiving long feedback (10- minute training with real-time visual feedback) [21]. In the present randomized trial, participants performed repetitive CPR practices under the guidance of an instructor with or without AVF devices, and bystanders who participated in a CPR training with an instruc- tor and an AVF device showed higher technical skill acquisition compared with those who received training with standard instructor-based feedback only. In addition, our results demon- strated that participants performed better CCs with AVF devices at 3 months but not at 12 months after RP. The repeated assess- ments were performed after each session of CC practice, and at 3 and 12 months after training. Participants performed a 2-min CC for each assessment, which probably has additional effects on CC quality retention. However, the duration of CC for testing (2 min) is short as compared to CC practice (30 min). In addition, the repeated assessments were performed without any feedback (human verbal feedback or AVF device). Together, in addition to verbal human feedback, AVF devices are possibly beneficial for CC training and for short-term, but not long-term, quality retention.

Limitations

There are three major limitations in the present study. Firstly, a previous study indicated that compression-only CPR training appears to help bystanders retain CPR skills better than conven- tional CPR training [24]. Therefore, in the present study, a short- ened CC-only CPR training was performed instead of conventional training. Although our results effectively demon- strated the effects of RP and AVF devices in CC skill acquisition and quality retention, it is unclear whether other CPR skills, includ- ing the recognition of cardiac arrest, ventilation, and the use of an automated external defibrillator, can also be affected by RP and feedback devices. Secondly, in order to determine whether AVF is beneficial for CC quality retention after RP, a RP group and a RP + AVF group were included in this study. A separate ”control group” without repeated practices and the use of AVF device has not been considered. Previous study has already reported that RP was able to improve CPR quality retention [12]. However, the RP implement in the present study differs from that of previous study,

which possibly resulted in different outcomes. Thirdly, all repeti- tive practices were completed in 1 week in this study. Previous study reported that ”rolling refresher” training with a 1-month interval improved retention of CC skill quality at 6 months [25]. The limited effects of RP on CC quality retention in the present study were possibly due to short intervals between repetitive prac- tices. Therefore, regulator CC skill refreshing with an appropriate interval is possibly important for quality retention.

Conclusion

The use of a real-time AVF device can further improve rescuers’ initial acquisition of CC skills and short-term quality retention after RP. However, long-term quality retention was not different between rescuers performing RP under the guidance of AVF devices and those practicing with verbal human feedback from instructors.

Competing interests

None.

Acknowledgments

The authors would like to thank the students of Wuhan Univer- sity (Wuhan, Hubei, China), as their efforts made this study possi- ble. We would like to thank all of the staff of the American Heart Association Emergency Cardiovascular Care Training Center (Zhongnan Hospital of Wuhan University). This study was finan- cially supported by the Emergency Diagnostic & Therapeutic Cen- ter of Central China. This study was also supported by Beijing Union Medical Fund – Ruiyi Emergency Medical Research Founda- tion (No. 201615; No. 201725).

References

1. Lavonas EJ, Drennan IR, Gabrielli A, et al. Part 10: special circumstances of resuscitation: 2015 American Heart Association guidelines update for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation 2015;132:S501-18.
2. Benjamin EJ, Blaha MJ, Chiuve SE, et al. Heart disease and stroke statistics-2017 update: a report from the American Heart Association. Circulation 2017;135: e146-603.
3. Cummins RO, Eisenberg MS. prehospital cardiopulmonary resuscitation. Is it effective? JAMA 1985;253:2408-12.
4. Dowie R, Campbell H, Donohoe R, et al. ‘Event tree’ analysis of out-of-hospital cardiac arrest data: confirming the importance of bystander CPR. Resuscitation 2003;56:173-81.
5. Bobrow BJ, Spaite DW, Berg RA, et al. Chest compression-only CPR by lay rescuers and survival from out-of-hospital cardiac arrest. JAMA 2010;304:1447-54.
6. Kitamura T, Kiyohara K, Nishiyama C, et al. Chest compression-only versus conventional cardiopulmonary resuscitation for bystander-witnessed out-of- hospital cardiac arrest of medical origin: a propensity score-matched cohort from 143,500 patients. Resuscitation 2018;126:29-35 [Epub 2018 Feb 21].
7. Sasson C, Magid DJ, Chan P, et al. Association of neighborhood characteristics with bystander-initiated CPR. N Engl J Med 2012;367:1607-15.
8. Gass DA, Curry L. Physicians’ and nurses’ retention of knowledge and skill after training in cardiopulmonary resuscitation. Can Med Assoc J 1983;128:550-1.
9. Kaye W, Mancini ME. Retention of cardiopulmonary resuscitation skills by physicians, registered nurses, and the general public. Crit Care Med 1986;14:620-2.
10. Riegel B, Nafziger SD, McBurnie MA, et al. How well are cardiopulmonary resuscitation and automated external defibrillator skills retained over time? Results from the Public access defibrillation Trial. Acad Emerg Med 2006;13:254-63.
11. Allan KS, Wong N, Aves T, et al. The benefits of a simplified method for CPR training of medical professionals: a randomized controlled study. Resuscitation 2013;84:1119-24.
12. Mpotos N, De Wever B, Cleymans N, et al. Repetitive sessions of formative self- testing to refresh CPR skills: a randomised non-inferiority trial. Resuscitation 2014;85:1282-6.
13. Kleinman ME, Brennan EE, Goldberger ZD, et al. Part 5: adult basic life support and cardiopulmonary Resuscitation quality: 2015 American Heart Association guidelines update for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation 2015;132:S414-35.
14. Brennan EE, McGraw RC, Brooks SC. Accuracy of instructor assessment of chest compression quality during simulated resuscitation. CJEM 2016;18:276-82.
15. Singleton MN, Allen KF, Li Z, et al. Rolling-refresher simulation improves performance and retention of paediatric intensive care unit nurse code cart management. BMJ Simul Technol Enhanc Learn 2018;4:77-82.
16. Cortegiani A, Russotto V, Baldi E, et al. Is it time to consider visual feedback systems the gold standard for chest compression skill acquisition? Crit Care 2017;21:166.
17. Zapletal B, Greif R, Stumpf D, Nierscher FJ, Frantal S, Haugk M, et al. Comparing three CPR feedback devices and standard BLS in a single rescuer scenario: a randomised simulation study. Resuscitation 2014;85:560-6.
18. Kim SC, Hwang SO, Cha KC, Lee KH, Kim H, Kim YK, et al. A simple audio-visual prompt device can improve CPR performance. J Emerg Med 2013;44:128-34.
19. Gonzalez-Calvete L, Barcala-Furelos R, Moure-Gonzalez JD, Abelairas-Gomez C, Rodriguez-Nunez A. Utility of a simple lighting device to improve chest compressions learning. Rev Esp Anestesiol Reanim 2017;64:506-12.
20. Sarma S, Bucuti H, Chitnis A, Klacman A, Dantu R. Real-time mobile device- assisted chest compression during cardiopulmonary resuscitation. Am J Cardiol 2017;120:196-200.
21. Baldi E, Cornara S, Contri E, et al. Real-time visual feedback during training improves laypersons’ CPR quality: a randomized controlled manikin study. CJEM 2017;19:480-7.
22. Yeung J, Meeks R, Edelson D, et al. The use of CPR feedback/prompt devices during training and CPR performance: a systematic review. Resuscitation 2009;80:743-51.
23. Pavo N, Goliasch G, Nierscher FJ, et al. Short structured feedback training is equivalent to a mechanical feedback device in two-rescuer BLS: a randomised simulation study. Scand J Trauma Resusc Emerg Med 2016;24:70.
24. Nishiyama C, Iwami T, Kitamura T, et al. Long-term retention of cardiopulmonary resuscitation skills after shortened chest compression-only training and conventional training: a randomized controlled trial. Acad Emerg Med 2014;21:47-54.
25. Niles DE, Nishisaki A, Sutton RM, et al. Improved retention of chest compression psychomotor skills with brief ”rolling refresher” training. Simul Healthc 2017;12:213-9.