Effects and Retention of Self-Re-Learning Using Video Recording of Advanced Life Support (ALS) On Nursing Students’ Knowledge, Self-Ecacy, and Skills Performance

Background In-hospital cardiac arrests account for 80% of hospital deaths, and the survival rate is not signicantly different from that of pre-hospitalized cardiac arrest patients. The nurses would presumably be the rst to see an in-hospital cardiac arrest patient. This study proposed to measure nursing students’ knowledge, self-ecacy, and skills performance of advanced life support (ALS) 6 months after the training by sending their videos taken during the nal skills test after the ALS training. Methods This is an experimental study with a randomized control group design. The participants’ knowledge, self-ecacy, and skill performance of ALS were evaluated immediately after the training and participants were videotaped during the nal skills test. Thereafter, we sent the video to the experimental group through the mobile phone messenger application once a month from the third month after training. Approximately six months after the training day, we conducted a follow-up test of the measured variables using the blinded method. Results Six months after the ALS training, knowledge scores decreased signicantly in both groups (p<0.001). Self-ecacy decreased by about 3 from 50.55 to 47.18 in the experimental group (p = 0.089), while it decreased by 10 in the control group from 50.67 to 39 (p<0.001). The skills performance decreased from 27.5 to 26.68 in the experimental group, while it decreased signicantly from 27.95 to 16.9 in the control group (p< 0.001). Conclusion Self-study with videos taken during an ALS skills test helps enhance the sustainable effects of training, such as knowledge, self-ecacy, and skills performance. ALS self-ecacy modied comprised 12 questions, including 2 questions on BLS, on ECG 2 on teamwork, 3 on ALS, and 2 on PCAC. Each a ve-point Likert scale with 5 for ‘very condent’ and 1 point for ‘very uncondent’. higher the score, higher the self-ecacy for professional resuscitation. The tool’s internal reliability at the time of alpha value of 0.91 and 0.87 a on and appropriate instructions, and 2 on post-cardiac care. Each question was scored two, one, and zero points for correct, insucient, and incorrect performance, respectively.


Introduction
The most common causes of death in Korea include malignant neoplasms, heart diseases, and cerebrovascular diseases. Among them, 62.4 per 100,000 people died due to heart diseases in 2016, which is steadily increasing to 43.8% compared to 43.4 in 2007 [1]. Particularly, due to ageing and increase in chronic diseases that cause cardiac arrest, the proportion of patients with acute cardiac arrest discharge were 23.7% and 6.4%, respectively. These ndings may be related to the healthcare provider's ability to perform CPR besides the patient's age and health status.
In-hospital cardiac arrest chains of survival in the 2015 Cardiopulmonary resuscitation guidelines recommended by the American Heart Association include the ve-stage chain of cardiac arrest: early detection -rapid reporting -rapid cardiopulmonary resuscitation -rapid de brillation -effective specialized cell and post-cardiac arrest integrated treatment [6]. Chances of brain injury begin about 4-5 minutes after cardiac arrest; therefore, the most important part of cardiac arrest is rapid CPR by the rst witness [6]. Nurses are more likely to detect cardiac arrest early in their care by observing the patient's condition and its changes over 24 hours [7]. Therefore, the nurses' early response is crucial [8]. When cardiac arrest occurs in a hospital, Advanced Life Support (ALS), including Basic Life Support (BLS), monitor use, emergency medication, and advanced airway maintenance is performed [9]. During treatment, the necessity of de brillation and the drugs to be administered vary depending on the electrocardiogram (ECG) rhythm [10,11]. Therefore, nurses require ALS training. Speci cally, nursing students who will become nurses in the future are more likely to face cardiac arrest patients for the rst time; therefore, training them in ALS is essential [12].
Previous studies on the effects of ALS training revealed that simulation-based ALS training can improve nursing student's knowledge, critical thinking, and problem-solving [13,14], and can also enhance their self-con dence and clinical performance [15]. However, three to six months after ALS training, the effect decreases [16,17], and retraining is required to maintain the continuous effect of ALS training [18,19].
Although prior studies have shown the effectiveness and persistence of ALS, most of these are crosssectional studies that con rm the effects before and after education. Intervening research is needed to con rm the continuity of educational effects and to maintain knowledge, self-e cacy, and performance. Therefore, this study compared the continuous effects of ALS training by measuring the nursing students' knowledge, self-e cacy, and skills performance of ALS immediately after the training and 6 months after the training.

Design and Setting
This experimental study, with a randomized control-group pretest-posttest design, veri ed the sustained effects of the Korean Advanced Life Support (KALS) training on the nursing students' knowledge, selfe cacy, and skill performance of ALS.

Subjects
The study participants were 4 th -grade students of the Department of Nursing at K University, who met the following inclusion criteria: 1) understands the research purpose, participates voluntarily, and agrees to shoot videos, 2) has completed the BLS training of the American Heart Association as a nal grade student, 3) who participated voluntarily in the KALS training of the KACPR and agreed to shoot videos, and 4) who uses a mobile phone messenger application that can send videos. Participants who refused to shoot videos or had already completed the Advanced Cardiovascular Life Support Provider Course (ACLS) of the American Heart Association were excluded from this study because of differences in the parameters to be measured.

Sample
The number of study participants was calculated using the G-Power (ver. 3.1.9 for Mac) program. For ttest, a statistical method used for comparing the means of two groups, the minimum number of participants was 21 per group. Therefore, considering the dropout rate of 25 per county based on a similar prior study25 per group were selected for each [20,21]. Allocation concealment was applied to the experimental and control groups and we did not inform the participants which group they belonged to until posttest. After the posttest, the control group were also sent their video of the nal skills test through the mobile messenger application. This training is a one-day ( ve to six-hour) course wherein the knowledge and skills required for rst aid treatment of cardiac arrest patients in hospitals or ambulances is provided.

ALS knowledge
The ALS knowledge measuring tool was developed by the researcher based on the contents of the ACLS provider manual and the 3rd edition of the KALS textbook by the KACPR KALS committee. The validity of the knowledge was evaluated by one emergency physician, two ACLS and KALS instructors, two nurses with more than 10 years of emergency room experience, and two nursing professors. All items had a Content Validity Index (CVI) of 0.8 or higher. It comprised 4 questions on BLS, 5 on ECG recognition, 4 on Teamwork, 4 on ACLS, and 3 on post cardiac arrest care (PCAC).
ALS self-e cacy In this study, resuscitation self-e cacy [22] was measured by ALS self-e cacy using a modi ed and supplemented tool. The revised tool comprised 12 questions, including 2 questions on BLS, 3 on ECG recognition, 2 on teamwork, 3 on ALS, and 2 on PCAC. Each item had a ve-point Likert scale with 5 points for 'very con dent' and 1 point for 'very uncon dent'. The higher the score, the higher the selfe cacy for professional resuscitation. The tool's internal reliability at the time of development was Cronbach's alpha value of 0.91 [22] and 0.87 in this study.

ALS skills performance
The Training of In-hospital Cardiac Arrest (TROICA) checklist of KALS committee, developed for the KALS provider course, was used after obtaining the KACPR ALS committee's consent to measure ALS skills. The TROICA, a measurement tool for KALS skills, comprised 15 questions, including 2 questions on BLS skills, 3 on teamwork, 3 on ALS algorithms, 5 on cardiac arrest cognition and appropriate treatment instructions, and 2 on post-cardiac care. Each question was scored two, one, and zero points for correct, insu cient, and incorrect performance, respectively.

Procedure
The study process followed the CONSORT 2010 Guidelines [23]. Immediately after ALS training, all the participants were surveyed for ALS knowledge and self-e cacy and evaluated for ALS skills performance. The process of their ALS skills test was recorded. Subsequently, participants were randomly assigned to the experimental and control groups using a blocked randomization method. Block randomization is a method designed to prevent imbalances in the number of experimental and control groups that can occur in simple randomization and is generally used when the sample size is small [24].
Referring to the previous study [17] that found ALS knowledge, ALS self-e ciency, and ALS skills performance to be greatly reduced after three to six months of ALS training, and another study [25] that found a meaningful decrease in ALS knowledge and ALS skills performance three months after ALS training, the participant's recorded video of their skills test process was sent to the experimental group through a mobile messenger application once a month from the third month after training while no arbitration was conducted on the control group. About six months after the date of the initial training, two evaluators participated in the evaluation in the same way as and conducted a post-test of measurement variables without knowing the experimental and control groups. Figure 1 present the ow chart of the research process.

Ethical consideration
This study was approved by the Institutional Review Board Committee of the hospital to which the rst author belongs (IRB approval number: 20180518 / 20-2017-33 / 062). The participants were explained the purpose and procedures of the study and informed consent were obtained written consent from voluntary participants. Considering the ethical aspect, the recorded video of their ALS skills test process was also sent to the control group after the study.

Statistical analysis
Collected data were analysed using SPSS 24.0 (for Windows), and the selected statistical signi cance level for hypothesis testing was p<0.05. The general characteristics of the experimental and control groups were analysed using descriptive statistics of frequency, percentage, mean, and standard deviation. To test the normality of the measured variables, the participants were analysed using the Shapiro-Wilk test, which is mostly used for 3 to 50 participants [26]. The t-test and Mann-Whitney U test were used to verify the general characteristics of the experimental and control groups and the homogeneity of the dependent variables before the experiment. To con rm the pre-post change of the experimental group and the control group, the normal distribution was analysed by paired t-test and the non-normal distribution was analysed by using the Wilcoxon signed-rank test. The reliability of the measurement tool was analysed using Cronbach's alpha.

Results
General characteristics of participants and homogeneity test of experimental and control groups Three of the 25 participants in the experimental group failed to follow up on a personal schedule, and four of the 25 participants in the control group dropped out of contact. Therefore, the analysis included 22 and 21 participants in the experimental and control groups, respectively (Fig. 1). There was no signi cant difference in the experimental group and control group homogeneity test for general characteristics, including gender, age, average grades, nursing degree satisfaction, and university life satisfaction (Table 1).  Comparison of dependent variables immediately and 6 months after training of experimental and control groups Table 3 shows the differences between the experimental and control groups' knowledge, self-e cacy, and skills performance immediately after and six months after training. Regarding the total knowledge score, the experimental group's pretest score (16.68) showed a statistically signi cant decrease than the posttest score (15.32) (p = .001), and the control group's pretest score (17.19) also showed a signi cant decrease compared to the posttest score (14.67) (p < .000). Self-e cacy scores and skills performance scores were reversed. Regarding the total self-e cacy score, the test group's pretest score (50.55) was not signi cantly decreased compared with the posttest score (47.18), but the control group's pretest score (50.67) was signi cantly decreased compared with the posttest score (39). Regarding the skills Comparison of dependent variables 6 months after training between experimental and control groups The knowledge, self-e cacy, and skills performance of the experimental and control groups were examined six months after the training. In the total score, the experimental group scored higher than the control group; however, there was no signi cant difference in the sub-items. In the self-e cacy score, the experimental group was higher than the control group in all the sub-items. ECG reading of the sub-items was higher in the experimental group (7.55) than in the control group ( .00, the experimental group was signi cantly higher than the control group (p = .002, p = .002, p = .02, p = .043, and p = .007). In the skills performance score, the experimental group was higher than the control group in all the sub-items (p < .001) ( Table 4; Fig. 2).
performance score, the pretest score of the experimental group (27.5) was not signi cantly decreased compared to the posttest score (26.68), but the pretest score of the control group (27.95) was signi cantly lower than the posttest score (16.9).

Discussion
This study veri ed the effectiveness and retention of nursing students after ALS training. Consequently, the experimental group that received the video showed higher persistence in knowledge, self-e cacy, and skills performance than the control group. Based on these results, the ways to increase and maintain the effectiveness of ALS training are discussed.
In a study [27] evaluating BLS and ACLS knowledge of healthcare providers, the average score of participants did not exceed 50%. However, healthcare providers with BLS or ACLS training had higher knowledge scores than those without training. Therefore, ALS training can improve knowledge scores. For nursing and medical school students, a study [17] measuring knowledge scores immediately after, after 3-6 months, and after 6-9 months, using traditional ACLS training and high delity mannequins showed that knowledge scores measured after 3-6 months and 6-9 months were reduced, respectively, compared to knowledge scores immediately after training. Additionally, a study [28] comparing the existing and high-delity simulations on the persistence effect of ACLS knowledge among medical students also reported that ACLS knowledge scores decreased signi cantly after one year compared to immediately after training. In the results of this study, the knowledge score measured after 6 months decreased in both the control group and the experimental group, but the experimental group showed a smaller decrease than the control group. These results are similar to a previous study [14] on nursing students in which the control and experimental groups underwent traditional lecture-based ACLS training and simulation-based ACLS training, respectively. The simulation training was more effective in acquiring and maintaining ACLS knowledge than lecture-based education. In a study [29] measuring knowledge 2 weeks and 8 weeks after ACLS training for nurses in the emergency department, knowledge retention decreased eventually. The authors emphasized that knowledge retention could be increased through simulation-based re-learning after the ACLS training. Reduction of knowledge after ACLS training is considered to be a natural result over time, and a renewal education program is necessary to maintain knowledge retention. Additionally, it may be possible to consider ways to maintain the knowledge level through newsletters and email noti cations.
Self-e cacy is an individual's belief in her/his ability to perform a speci c task or activity [30]. Simulation-based ACLS training can increase self-e cacy for ACLS skills performance. In a study [31] measuring the self-e cacy before and after simulation-based ACLS training undertaken by medical students, the self-e cacy increased signi cantly after the training. Furthermore, the experimental [14] group showed a signi cantly higher self-e cacy than the control group in a study comparing the experimental group with the simulation-based ACLS training and the control group with the traditional lecture-based resuscitation training. However, the persistence of self-e cacy was found to decrease signi cantly over time [17]. In a study [32] comparing healthcare e cacy immediately after and 6 months after PALS (Pediatric Advanced Life Support) training of healthcare providers, self-e cacy measured after 6 months was signi cantly reduced compared to immediately after training. Additionally, in a study [33] comparing nursing students immediately after and 3 months after BLS training, self-e cacy decreased signi cantly after 3 months.
The self-e cacy of nursing students can be determined by the ongoing interaction between cognitive, behavioral, and environmental factors [34]. The self-e cacy measured immediately after the training in this study was not different between the control group and the experimental group. Although the selfe cacy measured 6 months after the training was not signi cantly decreased in the experimental group, it was signi cantly decreased in the control group. These results suggest that they could have re-learned their knowledge and skills of ACLS by watching their skills test videos, and this could improve their retention of self-e cacy.
Simulation-based ALS training is di cult to improve skills performance based on previous studies.
However, this performance ability decreases rapidly over time [25]. Only 30% of the nurses passed the skills test measured 3 months after ALS training [35], and another studies [36,37] reported a decrease in ALS skills performance 6 months after ALS training. Furthermore, a systematic literature review of retention of skills performance after healthcare providers' ALS training showed a decrease in skills performance between 6 months and 1 year after training [38]. On comparing the skills performance of the control group, that had no intervention in this study, immediately after and 6 months after training, the skills performance measured 6 months after training decreased signi cantly. As such, skills performance begins to decline between three and six months and appears to decrease signi cantly after one year. As a method of retaining these skills performance, iterative simulation-based ALS training can improve the retention of skills [38,39]. After six months of clinical experience, training results showed a longer-lasting effect on skills performance than those without clinical experience [38,40]. Moreover, in ACLS training, practicing for a 2-minute cycle similar to the actual time resulted in higher skills performance measured 3 months later compared to short training [41].
In this study, the experimental group was sent a video of their process of nal skills test three months later, a relatively simple and cost-effective method, to retrain themselves. Six months later, the experimental group that received the video had better retention of skills performance than the control group. This video delivery method can increase the retention of skills performance and retraining time.
Timely reminders to participants who have received ALS training will be required for self-retraining.

Limitations
This study is a randomized control study; however, there are some limitations. First, this study's results are di cult to generalize because the sample size was small and the experiment was conducted in one institution. Second, since the same standardized tool was used at each data collection point, participants could have possibly remembered previous answers. Lastly, this local cross-sectional study of nursing students cannot be generalized in Korea. Still, further trials are needed to improve retention of ACLS knowledge, self-e cacy, and skills performance.

Conclusion
Our study showed that sending videos to the nursing students of their process of nal skills test between 3-6 months after training, a relatively simple and cost-effective method after ALS training, to induce selflearning can be effective in retaining knowledge, self-e ciency, and skills performance for ALS. Further studies should con rm the proper timing of sending the video. Replication studies are needed to recon rm these ndings.

Declarations
Availability of data and materials The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request. Study process