Cardiopulmonary resuscitation (CPR) education with a feedback device is known to result in better CPR skills compared to one without the feedback device. However, its long-term benefits have not been established. The purpose of this study was to evaluate the long-term CPR skill retention after training using real-time visual manikins in comparison to that of non-feedback manikins.
We recruited 120 general university students who were randomly divided into the real-time feedback group (RTFG) and the non-feedback group. Of them, 95 (RTFG, 48; non-feedback group, 47) attended basic life support and automated external defibrillation training for 1 hour. For comparison of retention of CPR skills, the two groups were evaluated based on 2-minute chest compression performed immediately after training and at 3, 6, and 9 months. The CPR parameters between the two groups were also compared using a generalized linear model.
At immediately after training, the performance of RTFG was better in terms of average chest compression depth (51.9±1.1 vs. 45.5±1.1, p<0.001) and a higher percentage of adequate chest compression depth (51.0±4.1 vs. 26.9±4.2, p<0.001). This significant difference was maintained until 6 months after training, but there was no difference at 9 months after training. However, there was no significant difference in the chest compression rate and the correct hand position at any time point.
CPR training with a real-time visual feedback manikin improved skill acquisition in chest compression depth, but only until 6 months after the training. It could be a more effective educational method for basic life support training in laypersons.
Previous studies have reported some immediate effects of cardiopulmonary resuscitation (CPR) training with feedback devices. However, there is little evidence on the effects of these devices on CPR skill acquisition in laypersons and their long-term benefits.
CPR training with a real-time visual feedback manikin improved skill acquisition in chest compression depth until 6 months after the training. It could be a more effective educational method for basic life support training of laypersons than manikins without feedback.
High-quality chest compressions are essential for improving survival following a cardiac arrest. The chest compression rate, depth, hand position, and a full chest recoil are important factors affecting the overall quality of chest compressions [
In a systematic review, Yeung et al. [
However, it is more meaningful to measure the benefits of CPR education among laypersons to popularize the technique and save more lives. We hypothesized that there would be long-term benefits of CPR training using real-time visual feedback on CPR skill acquisition and retention. Accordingly, this study aimed to investigate the long-term benefits of chest compression-only CPR (COCPR) using real-time feedback manikins in comparison to that of non-feedback manikins in laypersons. The primary outcome measure was the mean chest compression depth, whereas the secondary outcome measures were the percentage of adequate chest compression depth and chest compression rate minute.
This was a prospective randomized simulation study of general university students. We advertised the recruitment of participants and received applications on a first-come, first served basis without sex restrictions through google survey. The inclusion criteria were age 19 years and university students. There were no exclusion criteria. After stratifying the participants based on sex and prior CPR education, they were randomly assigned to two groups based on the CPR training manikin used as the real-time feedback group (RTFG) and the non-feedback group (NFG). Randomization was achieved via simple block randomization using a random numbering. All participants attended basic life support (BLS) and AED training program and answered a questionnaire survey on age, height, weight, sex, body mass index (BMI), and prior CPR education.
This study was approved by the institutional review board (KNUH 2017-05-005) and was conducted according to the tenets of the Declaration of Helsinki and its seventh revision in 2013. All participants provided a signed consent form.
The sample size calculation was based on the primary end point (rate of correct compression depth >80%) for comparison of CPR quality between the two groups. The values for the reference were obtained from the report by Skorning et al. [
The participants attended training without knowing whether they were divided into groups in advance. However, blinding of instructors was not applied because it was hard to mask using feedback device. To reduce bias, we maximized similarity of education, and all CPR parameters were evaluated and corrected objectively from the CPR simulation manikin. To minimize the educational differences, (1) both groups used the same official CPR program; (2) the instructor could provide feedback, but to minimize intervention by the instructor, the lecture was delivered using a Practice-While-Watching method; (3) two certified BLS instructors from the AHA lectured alternately to the RTFG and NFG groups, with each session attended by about 20 students; and (4) both groups used the same CPR training manikin, but one group turned on the feedback system, while the other just turned it off.
The official BLS and AED training programs from the Korean Association of Cardiopulmonary Resuscitation were employed to train both groups for 1 hour each. The CPR course comprised theory and practice classes that included performing a 2-minute COCPR, four times. The CPR training proceeded in a kneeling position on the hard floor. The ratio of attendees:manikin was 2:1, while the ratio of instructor:attendees was 1:20. We then evaluated their CPR performance by testing their 2-minute COCPR skill immediately after the training in August 2017. To evaluate skill retention, the groups were re-evaluated 3 (November 2017), 6 (February 2018), and 9 months (May 2018) after the training. No further feedback was provided after initial training.
A specific CPR training manikin (BT-SEEM; BT Inc., Goyang, Korea) was used for the study. This manikin had a graphic interface that provided real-time visual feedback of the compression speed and depth. However, it could not provide feedback on the hand position and chest recoil because it is only suitable for training purposes. The tested CPR training simulator was a BT-CPTA, which was created by the same company (BT Inc.).
Information regarding the general characteristics of the participants was collected from the survey. Further, the CPR performances were evaluated using a CPR training simulator. Data on CPR parameters such as chest compression rate, the average depth of chest compression, the percentage of adequate chest compression depth, correct hand position (%), and chest recoil (%) were all collected from the CPR training simulator. Adequate chest compression was based on the 2015 guidelines, which included 5–6 cm chest compression depth and 100–120 per minute chest compression rate [
The general characteristics of the study participants were described as a number (percentage) or median (interquartile ranges). The chi-square test or Fisher exact test were used for comparing proportions. Non-normally distributed continuous variables were analyzed using a non-parametric test (Mann-Whitney test). Between-group comparisons of the CPR parameters were performed using the Student t-test at each follow up and a generalized linear model for the overall period. Kolmogorov-Smirnov test method was used for normality testing. All statistical analyses were performed using the IBM SPSS Statistics ver. 19.0 (IBM Corp., Armonk, NY, USA). A P-value <0.05 was considered statistically significant.
We analyzed all participants who underwent training to compare the two groups base on which they were originally allocated (intention-to-treat analysis). There were no significant differences in the general characteristics of the participants, including age, height, weight, BMI, and prior CPR education between the two groups (
Evidence on the long-term effects of feedback devices for BLS, particularly in laypersons, is still limited. In the current study, we recruited general university students as laypersons, and they participated in a 1-hour BLS/AED training. Compared to the participants who trained with a non-feedback manikin, those who trained with a real-time visual feedback manikin showed higher skill acquisition measured as “adequate chest compression depth.” Previous CPR feedback studies in laypersons include one by Wik et al. [
In our study, the CPR parameters were most significantly different between the two groups immediately after the training. The reason for this initial difference might be due to the different quality of feedbacks. We reduced the number of instructors to maximize the dependence and effectiveness of feedback devices, create a situation similar to public education, and minimize instructor interventions during the training. We believe that this initial difference in skill acquisition contributed to the significant differences in the average compression depth and percentage of adequate chest compression depth until 6 months after the training.
The optimal percentage of adequate chest compression in laypersons is hard to define, but it ranged from 58.9% to 77.8% with a feedback device and from 14.6% to 66.6% without a feedback device, respectively in previous studies [
A recent European Resuscitation Council guideline reported that the CPR skills deteriorate within months of training, and therefore, annual retraining strategies are recommended. However, the optimal intervals for retraining have not been determined and are likely to differ based on the characteristics of the participants [
According to the recent AHA guideline, training methods can be classified based on the baseline skill level of the trainees. For lay rescuers, COCPR is a reasonable alternative to conventional CPR, which is used for adult cardiac arrest patients. On this basis, the Korean Association of Resuscitation also created and distributed CPR training videos of different levels for laypersons (basic/advanced) and for healthcare providers. In our study, we used the basic CPR training program for a layperson, which included lessons on COCPR and the use of AED to investigate the effect of CPR training using a real-time feedback manikin on essential chest compression parameters. A literature review showed that various forms of CPR education yield different results. The results depend on the skill levels of the target participants (layperson or healthcare provider), previous CPR experiences, methods of training (instructor led or self-training), durations of training (very brief to full time), change of training material, and methods of training evaluation (COCPR, conventional CPR, or extended time CPR) [
Our study has a few limitations. First, the study group included university students, which restricts the generalizability of the results to the general population and other age groups. Second, this study is a randomized controlled trial, and we designed each arm to include 60 participants, taking into account potential dropouts. However, 25% of the participants did not participate in the CPR training, and a few trainees skipped one or more tests. Third, the instructors were not blinded. It would be meaningful to design a study using blinded assessors to reduce bias. Fourth, the CPR training manikin does not give feedback on the correct hand position and complete chest recoil. We believe that the training could be more beneficial if a device that provides feedbacks on multiple CPR parameters is used. Despite these limitations, we believe that our study is valuable because it compares the longterm benefits of CPR training in laypersons between traditional and real-time visual feedback manikins, helping verify the benefits of a real-time feedback manikin before extending this training as a part of public education.
In conclusion, CPR training with a real-time visual feedback manikin improved skill acquisition in chest compression depth metrics, but the advantage disappeared 6 months after the training. It could be a better educational method for BLS training in laypersons comparing with traditional training.
No potential conflict of interest relevant to this article was reported.
This study was supported by the Daegu Metropolitan City (number 20161221-0057). The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. We appreciate the Public Health Division of Daegu Metropolitan City Hall for their continued support.
Flow chart of the study design and participant recruitment. RTFG, real-time feedback group; NFG, non-feedback group.
Comparison of average chest compression depth at each time point of evaluations. Statistical significance was tested using the two sample t-test as a parametric approach. RTFG, real-time feedback group; NFG: non-feedback group.
Comparison of the percentage of adequate chest compression depth at each time point of evaluation. Statistical significance was tested using the two sample t-test as a parametric approach. RTFG, real-time feedback group; NFG: non-feedback group.
General characteristics of the participants (n=95)
RTFG (n = 48) | NFG (n = 47) | P-value | |
---|---|---|---|
Age (yr) | 21.0 (20.0–22.0) | 21.0 (20.0–22.0) | 0.268 |
Height (cm) | 163 (158–170) | 162 (156–168) | 0.156 |
Weight (kg) | 52.0 (50.3–64.5) | 52.0 (48.0–60.0) | 0.142 |
Sex, male | 10 (20.8) | 9 (19.1) | 0.521 |
Body mass index (kg/m2) | 20.4 (19.3–22.7) | 20.3 (19.0–21.7) | 0.320 |
Prior CPR education | 16 (33.3) | 15 (31.9) | 0.529 |
Values are presented as median (interquartile ranges) or number (%).
RTFG, real-time feedback group; NFG, non-feedback group; CPR, cardiopulmonary resuscitation.
Comparison of the cardiopulmonary resuscitation parameters by time and group
Immediately after training | 3 Months | 6 Months | 9 Months | Group | Time | Group & time | ||
---|---|---|---|---|---|---|---|---|
Feedback parameters | ||||||||
Compression rate (per minute) | ||||||||
NFG | 122.4 ± 2.31 | 114.7 ± 2.36 | 117.8 ± 2.36 | 116.2 ± 2.39 | 0.072 | < 0.001 | 0.919 | |
RTFG | 120.3 ± 2.29 | 110.1 ± 2.29 | 114.3 ± 2.39 | 114.6 ± 2.39 | ||||
Average chest compression depth (mm) | ||||||||
NFG | 45.5 ± 1.06 | 44.6 ± 1.08 | 44.8 ± 1.08 | 45.5 ± 1.09 | < 0.001 | 0.093 | 0.095 | |
RTFG | 51.9 ± 1.05 | 50.2 ± 1.05 | 48.4 ± 1.09 | 46.9 ± 1.09 | ||||
% of adequate chest compression depth | ||||||||
NFG | 26.9 ± 4.15 | 22.9 ± 4.24 | 22.5 ± 4.24 | 25.0 ± 4.29 | < 0.001 | 0.015 | 0.088 | |
RTFG | 51.0 ± 4.11 | 42.3 ± 4.11 | 31.8 ± 4.29 | 30.1 ± 4.29 | ||||
Non-feedback parameters | ||||||||
Correct hand position rate | ||||||||
NFG | 43.1 ± 6.36 | 35.8 ± 6.50 | 38.2 ± 6.50 | 43.7 ± 6.57 | 0.759 | 0.531 | 0.943 | |
RTFG | 41.3 ± 6.29 | 35.5 ± 6.29 | 43.7 ± 6.57 | 46.0 ± 6.57 | ||||
Full chest recoil | ||||||||
NFG | 99.2 ± 1.41 | 99.9 ± 1.44 | 99.7 ± 1.44 | 99.9 ± 1.46 | 0.024 | 0.994 | 0.994 | |
RTFG | 97.9 ± 1.39 | 97.6 ± 1.39 | 97.2 ± 1.46 | 96.8 ± 1.46 |
Values are presented as mean±standard deviations. Statistical significance was tested using a generalized linear model.
NFG, non-feedback group; RTFG, real-time visual feedback group.