Effectiveness of the Flipped Classroom on Medical Undergraduates’ Cardiopulmonary Resuscitation Training in Large Class Through Self-Efficacy: A Randomized Quasi-Experimental Study

Abstract

Purpose

The Flipped Classroom (FC) is a promising method of learning, particularly in medical education. Studies show that FC improves student engagement, achievement, and satisfaction. However, its effectiveness in larger classroom settings remains debated. This study explores the impact of FC on theoretical knowledge and practical skills in cardiopulmonary resuscitation (CPR) among medical undergraduates in larger classes. Findings could inform innovative and effective pedagogical approaches to enhance learning and performance in medical education and training.

Method

This quasi-experimental study used a control group and recruited students from the Second School of Clinical Medicine of Southern Medical University in 2019. Students were divided into TL or FC. Each group was further divided into 8 units based on grade point average (GPA) stratification. After 8 months, a CPR skill test was conducted to assess CPR retention.

Results

The FC and traditional learning (TL) groups had similar demographics. Time spent studying did not affect academic achievement. However, the FC group had significant improvements in pre-class study completion rates on theory learning platform (82.14% vs 22.73%, P = .000) and skill learning platform (75.89% vs 58.18%, P = .005), pre-class theoretical grade (39.98 ± 7.91 vs 27.89 ± 10.19, P = .00), final theoretical grade (18.04 ± 2.39 vs 16.36 ± 2.22, P = .00), and final skill grade (83.10 ± 8.59 vs 78.24 ± 11.04, P = .00). The FC group also had better CPR skill scores in situational settings after 8 months and increased self-efficacy.

Conclusion

The study concludes that FC can improve CPR theory and skill for medical undergraduates in large classes. Additionally, students in FC retain better CPR skills after 8 months, especially for situational CPR.

Keywords

flipped classroom self-efficacy large class CPR training medical undergraduates

Background

Cardiopulmonary arrest (CA) is one of the leading public health concerns worldwide, with high morbidity and mortality as well as a low discharge survival rate. In the US, data from 2005 to 2014 show 605,000 new-onset attacks and 200,000 recurrent attacks, with a sudden cardiac arrest mortality of 393,872 in 2018.¹ In China, the morbidity of CA is approximately 544,000 annually, which is gradually approaching the levels of developed countries. However, the overall level of rescue in China is far lower than that in developed countries and regions. The discharge survival rate of CA patients with good neurological function is approximately only 1%.² Cardiopulmonary resuscitation (CPR) is the key measure to improve the survival rate of CA and is an essential skill for medical students.

Flipped classroom (FC) is student centered, replacing the traditional teacher-centered teaching style. The core of FCs is the transfer of knowledge occurring outside the classroom and the internalization of knowledge occurring in the classroom.^3–5 FC is considered a powerful teaching method that can improve academic scores,^6,7 application performance, class satisfaction,⁸ active learning, and critical thinking.^6,7,9 Because of these benefits, FCs have become increasingly widespread and caught the attention of researchers and educators in medical fields.^10,11,6,7,9 For medical students, developing competency in critical thinking and application is crucial and could provide them with the ability to discern questions in an attempt to search for better ideas, a deeper understanding and better solutions for a given issue,¹² especially in initially identifying and performing procedures under critical emergence situations such as CA. Moreover, FC emphasizes evaluating satisfaction, which is an important factor in students’ learning. Learning satisfaction is highly related to CPR training outcomes and critical thinking to deal with a sudden emergency in the clinic.

However, the satisfaction of FC is still debatable. Some studies have shown that FC could significantly increase student's satisfaction because it provides a stress-free and risk-free simulation for student to practice clinic procedures in a realistic environment.¹³ By contrast, Beom reports there is no difference between the FC and TL regarding advanced cardiac life support skills.¹⁴ And some preferred the TL because the FC requires huge efforts from instructors to prepare materials and students of overloading of the curriculum outside of class. Apart from that, Adequate internet and technology must be provided to both instructors and students, as well as technological support when it's necessary.¹⁵

Compared to standard training, alternative training shows better performance in early recognition and prevention, activation of emergency response, high-quality CPR, defibrillation, postcardiac arrest care, and recovery.¹⁶ In a systematic review and meta-analysis by Lim et al, technology-based components benefited self-directed theoretical instruction, but instructor guidance, hands-on practice, and real-time feedback were necessary to achieve better learning outcomes in adolescent CPR training.¹⁷ However, Megan Boysen-Osborn found that an FC format for advanced cardiac life support marginally improved written test results.¹⁸ In addition, the effectiveness of FC across environments and learners with different levels (beginner to advanced) should be investigated.⁸

In addition, high-quality CPR requires not only medical students to immediately respond and perform in various situations but also confidence and willingness. Self-efficacy is a person's ability to engage in certain behaviors and achieve the desired results in a given situation and is, to a large extent, related to the individual's own sense of self-relevant abilities. Therefore, self-efficacy is a predictor of performance in intensive situations, such as practicing CPR. Self-efficacy has been extensively studied worldwide.¹⁹ However, there is no consensus on the role of self-efficacy in CPR.^20–22 A study from Italy regarding self-efficacy in the psychomotor skills of nursing students reported that increasing self-efficacy may increase engagement and thus improve learning outcomes as a predictor of academic success and personal development.²³ Another study reported no correlation between participants’ self-efficacy and neonatal CPR skills.²⁰ A systematic review reported no associations between knowledge, self-rated confidence, or self-rated psychomotor skill competence following CPR training in laypeople. Thus, investigating whether and how self-efficacy influences CPR training could assist in developing education strategies to improve skills.

Moreover, the existing studies have relatively small samples. Few studies have had sample sizes greater than 100 people per experimental group. Carrick reported that there were no statistically significant improvements in neuro-otology among 137 participants in both an FC group and a traditional group.²⁴ Beom reported no significant difference in CPR competency (n = 108) skills between FC (n = 55) and traditional learning (TL) (n = 53).²⁵ Scarce studies have yet to determine the effect of FC on skills in a large class. Considering the large class sizes in China, it is urgent and necessary to explore the effect of FC in large classes.

The primary goal of this study is to explore whether FC has a positive effect on CPR theory and skill acquisition in large undergraduate CPR classes, as well as the amount of time CPR skills are retained. All participants were third-year medical students from Class 1 (112 students) and Class 2 (110 students) at the Second School of Clinical Medicine of Southern Medical University in China. In addition, factors that could affect the results of CPR training FCs are also investigated to inform future education strategies, by considering self-efficacy and GPA stratification of students. In addition, students’ satisfaction with FC is also considered.

Method

The reporting of this study conforms to the The Do CTRINEGuidelines: Defined Criteria To Report INnovationsin Education.

Study design

This study used a quasi-experimental design with a control group. Written informed consent was obtained from all participants prior to inclusion in the study, which was previously approved by the Ethics Committee of Zhujiang Hospital of Southern Medical University. The ethical approval number of this study is 2022-KY-176-02.

Participants

This study was carried out in the spring semester of 2019 in the Second School of Clinical Medicine of Southern Medical University, China.

We recruited 222 third-year undergraduate Clinical Medicine majors from Zhujiang Hospital, Southern Medical University. The classes were divided into Class 1 (N = 112) and Class 2 (N = 110) according to school class management. The classes were allocated to either the intervention group or the control group through a purposive sampling method with one class for each group. The control group learned CPR in a traditional classroom, and the experimental group learned CPR in an FC. The students who participated studied in accordance with the teaching arrangement assigned to them.

The full sample inclusion is because of the special policy of Chinese medical education. The CPR training class in this study is the mandatory core curriculum for Chinese medical undergraduates. Besides, all medical students must have the competency to diagnose, provide first aid, and deal with emergencies, which is required for degree award. There was no unsolicited inclusion process to form a natural cohort study. Although theoretically all participated, to ensure the integrity of the data, we did a secondary control that a sign-in system (class on-site sign-in and random roll call) to ensure the participate and those who did not participate in the whole process of the class, such as due to absence (sick leave, personal leave and signed for approval) withdrawal or suspension, were excluded.

In addition, all instructors (n = 4) involved in this study were all from the Zhujiang Hospital Clinical Skills Center of Southern Medical University and held the following credentials: (1) higher education teaching certificate issued by the department of Education, Guangdong Province; (2) American Heart Association Basic Life Support instructor certification; and (3) full-time faculty certification in the Clinical Skills Center. Besides, to ensure standardization, the teaching team implemented standardized collective lesson planning before class, which includes: (1) synchronous analysis of teaching videos regarding AHA; (2) standardized calibration of demonstration action parameters using Laerdal SimPad^®.

Group Classification

First, according to the school classroom management method, this study is divided into 2 groups, Class 1 and Class 2. Class 1 and Class 2 were randomly grouped, with Class 1 being the FC group and Class 2 being the TL group.

Second, the students in both groups were stratified into levels according to the Southern Medical University student handbook grade levels, ie, excellent (A) was converted to a 4.5 GPA, good (B) to a 3.5 GPA, moderate (C) to a 2.5 GPA, passing (D) to a 1.5 GPA, and failing (E) to 0. Because there were no students in levels A or E, each group was then divided into 3 levels: levels B, C, and D.

Next, those at the 3 levels in each group were randomly and evenly sub-grouped into 8 units via Excel software. The students at each level in both groups were given a random number through the Excel random number table function, and then the students in the same order in each level were grouped to form 8 units by sorting the random numbers. Therefore, each group exhibited GPA stratification, ie, levels B, C, and D.

The classification divided the FC and TL groups into 8 units.

Study Protocol

CPR is a part of the Clinical Skills curriculum and is a mandatory class for third-year clinical medicine undergraduates at the Second School of Clinical Medicine of Southern Medical University. It is a 4-h class. According to the structure of the class, we divided it into 3 parts: pre-class, in-class, and post-class, which are introduced as follows in Figure 1.

Figure 1.

Structure of flipped class.

All the learning contents are identical for both FC and TL group.

Pre-class

For FC

One week before classroom teaching, learning materials and learning task lists were uploaded on the Southern Medical University Aike platform for theory learning and Massive Open Online Course (MOOC) for skill learning. For better understanding, we would use Theory Learning Platform as Aike, Skill Learning Platform for MOOC. All the students in FC group had access to the learning materials and learning list.

The learning materials included CPR proper operation video, CPR theoretical micro lessons, and relevant PPT and papers.

The learning contents included video learning of CPR, CPR operations, Adult Basic Life Support PPT, 2020 CPR and Emergency Cardiovascular Care (ECC) guidelines, “What is high-quality CPR?,” “How to achieve high-quality CPR,” and a theoretical test.

According to the learning task list, FC group must learn and complete all theoretical learning content before the in-class phase begins.

To confirm the FC's pre-class effectiveness, the instructor monitored the entire process with the background data of the platforms to track the progress of the students toward completion. According to the data, the instructor gave adequate reminders.

In addition, the instructors were available for both groups if the students had any questions.

The FC had to meet specific requirements.

The above list of learning contents had a deadline. The students were asked to complete the learning of the materials and a theoretical test before the in-class day. The theoretical test contains 12 single-choice questions, which cover all the basic theoretical knowledge of CPR.

The students working in units were asked to think about and ask at least 3 high-quality basic life support-related questions to be answered by unit members and include but are not limited to CPR.

The students were asked to recite the CPR flow chart and informed that the instructor would randomly select one person from each unit to recite the chart in class. In addition, the students received immediate feedback as part of the formative evaluation grade.

The students were informed that the instructor would show videos of incorrect CPR operation and that they would need to report the error points and how to correct them as a unit in class.

The students were informed that they would discuss the CPR operation in different specific situations as a unit in class. In addition, the students were also notified that CPR operation is part of the final skill test.

As a controller, TL group only complete pre-class theoretical test before in-class section.

In-class

The entire in-class phase is divided into a theory section and a skill section.

In the in-class theory section, identical contents and class processes were provided to both groups. The content in both groups included 6 parts: in-class introduction, pre-class problem discussion, reciting the CPR flow chart, “identifying incorrect procedures,” CPR in various situations, and questionnaire completion.

Concerning the experimental settings, different teaching methods were applied to the 2 groups. For both the FC and TL groups, in the introduction, the instructors introduced themselves and how the class would proceed. Afterward, the questions that had a low correct rate in the pre-class phase were answered. Meanwhile, the difference between the 2 CPR survival chains,²⁶ in and out of the hospital chain of survival, was shared by the instructor. In the last part of the class, the instructor concluded the class, and the students were asked to complete a questionnaire. Apart from the above, the subsequent parts were completely different between the 2 groups. Different teaching methods were applied to the 2 groups in the later 4 parts, as illustrated below.

For FC,

During the CPR flow chart recitation, 4 instructors were assigned to 8 units, 2 units per instructor. Each instructor supervised and provided feedback on the students’ recitations individually.

During the “identify the incorrect procedures” part, the students in all units discussed the incorrect procedures and gave improvement suggestions in response to the video of incorrect CPR operation. However, for units 1, 2, 3, and 4, 1 student was selected randomly as a representative to present the results of the unit discussion. The remaining members of the unit could also make contributions. There were 2 videos of incorrect CPR operations, which were divided between the units. Video one was assigned to units 1 and 2, and video 2 was assigned to units 3 and 4. The instructor gave feedback on the students’ answers immediately.

During the CPR video or various situations, the students discussed how to proceed with CPR in 4 situations as a unit. Then, for units 5, 6, 7, and 8, the units reported their solutions for proceeding in specific CPR situations. Unit 5 was assigned to a drowning situation for emphasizing droning-specific CPR sequence. Unit 6 was assigned to a garage situation for pre-hospital scene safety assessment. Unit 7 was assigned to a cracking glass situation for safe relocation of CPR patients. Unit 8 was assigned to a stairs situation for emphasizing emergency scene safety. Meanwhile, the instructor provided feedback to each unit.

For TL,

The students attended a didactic lecture. The instructor led the students to recite the CPR flow chart, explained the incorrect procedures in the operation videos, and interpreted how to conduct CPR in the 4 situations.

Before the in-class skill section, the students of both groups were asked to rewatch the CPR operation video. At the beginning of the in-class skill section, both groups were given a pre-class test. The test included 13 questions, the answers to which were addressed in the online class material. The first 12 questions were single best answer (SBA) questions that covered basic CPR theoretical knowledge. The students are asked to sequence the CPR steps in the correct order in the last question.

During the in-class skill section, both groups received identical practice content in the Zhujiang Hospital Clinical Skill Center. There were 8 sets of manikin variants compliant with AHA QCPR standards for both groups. For each group, each unit had one set of manikins, with 13 students sharing each manikin. Every 2 units had one instructor, who conducted the sessions and provided feedback. The total practice time was the same for both groups, 2 and a half hours.

The students practiced CPR skills freely in their own units. However, every member of the unit was required to practice at least once. After practice, 2 students were randomly selected in each unit. One was for the situational CPR skill test, and the other was for the CPR theoretical test. Both scores were counted as part of the final grade.

Furthermore, to maintain objectivity, different instructors provided guidance and administered the situational CPR skill tests, ensuring that the same instructional content was provided to all student groups throughout the study period.

Post-class

An identical assessment and evaluation method was used for both groups; it contained a series of tests, including a CPR theoretical test, a situational simulated CPR skill test, a class satisfaction survey, and a self-efficacy survey.

In the CPR theoretical test section, students were asked to complete the CPR theoretical test in 5 min. The theoretical test included 2 types of questions, SBA and case-based questions. The SBA included 12 questions that asked the students to choose the correct basic CPR knowledge. The case-based question included 4 questions based on the specific CPR situation.

The students were asked to deal with the specific given situation and perform the standardized CPR procedure. In the situational simulated CPR skill test section, 4 situations were prepared in advance; the situations were different from each other but had the same difficulty level. The test was conducted in a single test room equipped with one Laerdal ResusciAnne^® QCPR Intelligent Feedback Manikin, and one real-time data acquisition system. Only 1 student was allowed to be present with 1 test officer in the test room. The students were asked to select 1 situation randomly, and they have 5 min to complete the CPR procedure. The test officer scored the students according to a rating scale. The skills test process for this part was as follows. First, the students were introduced to the CPR situation. Second, the students assessed the safety of the environment and the consciousness of the patient. Then, the test officer signaled to the student that the person in the situation had no response. The students started the CPR procedure. After finishing 5 cycles of CPR, the students assessed the effect of the CPR. Next, the test officer signaled to the students that the patient's pulse and breathing were in recovery. Finally, the students helped the patient stay in recovery position to end the test.

The rating scale contains 31 items divided into 2 main parts: situation handling and CPR procedures. In the situation handling part, the students were score based on whether they were able to determine how to start the CPR operations based on the situation. The students who completed the situation part completely and correctly were awarded 5 points. Otherwise, they were awarded 0 points. The CPR procedure parts tested the specific details of the students’ CPR operations to determine whether they could perform CPR correctly. This part consisted of 30 items, and each item was scored. The total score was obtained by summing the scores on the situation handling part and the CPR procedure part.

Test process standardization: (1) Manikin reset to initial state (compression depth sensor zeroed), student read a standardized clinical scenario; (2) CPR test (5 cycle); (3) Automated recording: QCPR monitoring of compression depth, rebound, frequency, duration of interruption of compression, artificial respiration chest lift. Examiners graded how the student deal with the situation, evaluating the status of the patient and initiating the CPR.

Final score standardization: test officers are all 4 instructors in the class.

After finishing the theoretical and skill tests, both the class satisfaction and self-efficacy questionnaires were waiting for students to complete.

According to the teaching schedule of the university, the students in both groups were be tested again for CPR skills 8 months later, immediately before they started their hospital internships. Both the rating scale and the content of the CPR skills test remained the same as before, including the CPR situation handling part and the CPR procedure part.

Appropriate Assessment

Appropriate assessment is the key to class redesign. Therefore, the final score of the entire class in this study was based on the formative evaluation of the students, including the test part and the performance part. The test part included the pre-class, in-class, and final theoretical and skill tests for both groups. However, the performance evaluation covered different groups. For the FC group, it included the pre-class steps of video watching, material study, and theoretical test completion, and the in-class steps of CPR flow chart recitation, identifying incorrect CPR operation in videos, performing CPR in different situations and active engagement activities, including class participation, question answering, etc For the TL group, the performance evaluation included attendance and active engagement activities, such as class participation and question answering.

Questionnaires

Three questionnaires were used in this study, pre-class and post-class questionnaire for basic information, and self-efficacy questionnaire.

Pre-class questionnaire includes 3 parts: demographics (name, student ID, class, gender, age); study habits.

Post-class questionnaire includes 3 parts: demographic information, satisfaction about the class.

The self-efficacy questionnaire was used to collect data on the students’ self-efficacy ability pre- and post- class. The questionnaire was constructed based on the published studies, the original study by pintrich and de groot's and reversion one by Yusong Liang.^27,28 In Liang's study, the questionnaire was proven valid and reliable. The reliability of self-efficacy for learning ability, Cronbach α coefficient 0.820 and self-efficacy for learning behavior Cronbach α coefficient 0.752. The self-efficacy questionnaire consisted of 20 items in 3 categories: learning ability, learning behavior, and total self-efficacy. Each item was rated on a 5-point Likert scale with the following response alternatives: 5 = strongly agree; 4 = agree; 3 = neutral; 2 = disagree; and 1 = strongly disagree. All participants in the experimental study (N = 212) completed the questionnaire's pre- and post-measures.²⁷

Data Collection

The data collected in this study included the pre-class and final CPR theoretical test scores, in-class and final CPR skill test scores, classroom satisfaction questionnaire results, pre-class and final self-efficacy questionnaire results, and the backstage learning data records of the study participants on the online platforms.

Data were collected from April 6 to June 10, 2022.

Statistical Analysis

The collected data were analyzed by the Statistical Package for the Social Sciences (SPSS) 26 software (IBM SPSS statistics for Windows, Version 26.0. Armonk, NY: IBM Corp., 2019). The significance level was defined as <.05 for all tests. Descriptive statistics and chi-square tests were used to analyze the basic demographic information of the study participants, study variables, and pre-class study completion rate. The Mann‒Whitney test and independent t-test were used to analyze academic achievement between the 2 groups. A paired t-test was used to analyze the stratification of the 2 groups and self-efficacy. Exam scores were expressed as the mean and standard deviation. The results of the questionnaire of class satisfaction were reported as percentages.

Result

Demographics of participants

The demographics of the participants are presented in Table 1. A total of 212 students were enrolled in the study, and 1 student withdrew from the study due to suspension. There were no significant differences between the FC and TL classes in terms of performance points, age, and sex, indicating homogeneity between the 2 groups.

Table 1.

Demographics of participants.

	Flipped classroom (n = 112)		Traditional learning (n = 110)
Characteristics	Mean	SD	Mean	SD	P-value
Previous GPA	3.239	0.444	3.242	0.485	.956
Gender	N	%	N	%
Male	67	59.8	55	50	.141
Female	45	40.2	55	50

Pre-class study completion rate

According to the platform background data, on Aike, the completion rate was 82.14% for the FC group. On MOOC, the completion rate was 75.89% for the FC group (Table 2).

Table 2.

Pre-class FC study completion rate.

		Flipped classroom (n = 112)
		N	%
Theory learning platform	Complete	92	82.14
	Incomplete	20	17.86
Skill learning platform	Complete	85	75.89
	Incomplete	27	24.11

Effects of the flipped classroom

Pre-class theoretical scores: the FC group had higher scores than the TL group, and there were statistically significant differences (39.98 ± 7.91 vs 27.89 ± 10.19, P = .00), in which the B level and C level were higher than those in the TL group, and there were statistically significant differences (39.18 ± 7.96 vs 27.31 ± 10.77, P = .00; 38.97 ± 7.90 vs 27.42 ± 10.77, P = .00). The D level was higher than in the TL group, but there were no significant differences (38.25 ± 8.73 vs 32.45 ± 8.28, P = .16) (Table 3).

Table 3.

Pre-class theoretical scores.

	Flipped classroom (n = 112)		Traditional learning (n = 110)
	Mean	SD	Mean	SD	P-value
Pre-class theoretical score	39.98	7.91	27.89	10.19	.00
Tier B	39.18	7.96	27.31	10.77	.00
Tier C	38.97	7.90	27.42	10.11	.00
Tier D	38.25	8.73	32.45	8.28	.16

The final theoretical scores: compared with the TL group, the final theoretical scores of the FC group improved, with significant differences (18.04 ± 2.39 vs 16.36 ± 2.22, P = .00), among which the SBA question scores improved, with significant differences (10.97 ± 1.12 vs 9.68 ± 1.26, P = .00); the case-based question scores improved but did not show significant differences (7.06 ± 1.90 vs 6.68 ± 1.68, P = .12).

In terms of performance stratification, compared with the TL group, the performance of the B unit improved, with significant differences (19.26 ± 1.93 vs 17.11 ± 2.13, P = .00), with the scores on both the SBA questions and case-based questions improving, with significant difference (11.29 ± 0.84 vs 10.00 ± 1.00, P = .00; 7.97 ± 1.53 vs 7.11 ± 1.95, P = .047); the C unit improved, with significant difference (17.64 ± 2.40 vs 16.09 ± 2.26, P = .00). The scores on the SBA questions improving, with a significant difference (10.09 ± 1.19 vs 9.55 ± 1.31, P = .00), and the scores on the case-based questions improved but without significant difference (6.74 ± 6.95 vs 6.54 ± 1.54, P = .50). In the D unit, the performance improved but without significant difference (16.25 ± 1.98 vs 15.50 ± 1.58, P = .40), with the SBA questions and case-based questions both improving slightly but without significant difference (10.25 ± 1.165 vs 9.4 ± 1.58, P = .22; 6.00 ± 1.41 vs 6.10 ± 1.29, P = .88) (Table 4).

Table 4.

Final theoretical scores.

	Flipped classroom (n = 112)		Traditional learning (n = 110)
	Mean	SD	Mean	SD	P-value
Final theoretical score	18.04	2.39	16.36	2.22	.00
SBA	10.97	1.12	9.68	1.26	.00
Case-based question	7.06	1.90	6.68	1.68	.12
Tier B	19.26	1.93	17.11	2.13	.00
SBA	11.29	0.84	10.00	1.00	.00
Case-based question	7.97	1.53	7.11	1.95	.047
Tier C	17.64	2.40	16.09	2.26	.00
SBA	10.90	1.19	9.55	1.31	.00
Case-based question	6.74	6.95	6.54	1.54	.503
Tier D	16.25	1.98	15.50	1.58	.40
SBA	10.25	1.165	9.4	1.58	.22
Case-based question	6.00	1.41	6.10	1.29	.88

Final skill scores: Compared to the TL group, the FC group achieved better and significantly different results in terms of final skills (83.10 ± 8.59 vs 78.24 ± 11.04, P = .00), with significantly higher situational and skill scores (2.72 ± 2.50 vs 1.32 ± 2.21, P = .00; 80.38 ± 8.32 vs 76.92 ± 10.61, P = .01).

As far as stratification was concerned, in level B, the total final skill scores, situational scores and skill scores were higher in the FC group but without statistical significance (83.79 ± 9.96 vs 80.86 ± 9.28, P = .21; 2.35 ± 2.53 vs 1.57 ± 2.36, P = .19; 81.44 ± 9.90 vs 79.29 ± 9.21, P = .35). In level C, the total final skill scores, situational scores and skill scores were significantly higher in the FC group (83.69 ± 7.01 vs 77.49 ± 11.68, P = .00; 3.00 ± 2.47 vs 1.00 ± 2.02, P = .00; 80.69 ± 6.61 vs 76.49 ± 11.00, P = .01). In level D, the total final skill scores and the skill scores were higher in the FC group but without statistical significance (75.00 ± 11.66 vs 73.90 ± 11.33, P = .84; 73.13 ± 11.77 vs 71.40 ± 11.27, P = .76). However, the situational scores in the FC group were lower than those in the TL group but without statistical significance (1.88 ± 2.56 vs 2.50 ± 2.64, P = .57) (Table 5).

Table 5.

Final skill scores.

	Flipped classroom (n = 112)		Traditional learning (n = 110)
	Mean	SD	Mean	SD	P-value
Final skill scores	83.10	8.59	78.24	11.04	.00
Situational scores	2.72	2.50	1.32	2.21	.00
Skill scores	80.38	8.32	76.92	10.61	.01
Tier B	83.79	9.96	80.86	9.28	.21
Situational scoresSkill scores	2.35	2.53	1.57	2.36	.19
Situational scoresSkill scores	81.44	9.90	79.29	9.21	.35
Tier C	83.69	7.01	77.49	11.68	.00
Situational scoresSkill scores	3.00	2.47	1.00	2.02	.00
Situational scoresSkill scores	80.69	6.61	76.49	11.00	.01
Tier D	75.00	11.66	73.90	11.33	.84
Situational scoresSkill scores	1.88	2.56	2.50	2.64	.57
Situational scoresSkill scores	73.13	11.77	71.40	11.27	.76

Skill scores after 8 months: After 8 months, another CPR skill test was administered, and the data were collected. The FC group had higher total scores but without significant differences (86.24 ± 0.635 vs 85.11 ± 0.772, P = .258). The FC group had higher situational scores, and there was a statistically significant difference (3.04 ± 0.232 vs 1.91 ± 0.233, P = .001) (Table 6).

Table 6.

CPR skill after 8 months.

	Flipped classroom		Traditional learning
	Mean	SD	Mean	SD	P-value
Situational scores	3.04	0.232	1.91	0.233	.001
Skill scores	83.21	0.614	83.20	0.735	.996
Total score	86.24	0.635	85.11	0.772	.258

In terms of stratification, for the students in level B, the FC group had higher situational scores (3.24 ± 0.416 vs 2.29 ± 0.427, P = .116), skill scores (84.06 ± 1.135 vs 83.86 ± 1.043, P = .896) and total scores (87.29 ± 1.293 vs 86.14 ± 1.143, P = .506); however, there were no significant differences. For the students in level C, the FC group had higher situational scores (2.86 ± 0.298 vs 1.92 ± 0.304, P = .030), skill scores (83.17 ± 0.739 vs 82.75 ± 1.076, P = .587) and total scores (86.03 ± 0.739 vs 84.68 ± 1.114, P = .308). Only the situational scores showed a significant difference. For the students in level D, the FC group showed only significantly higher situational scores (3.75 ± 0.818 vs 0.50 ± 0.500, P = .003). Both the skill scores (79.88 ± 2.973 vs 83.80 ± 1.896, P = .752) and the total scores of the FC group were lower than those of the TL group without a significant difference (Table 7).

Table 7.

CPR skill after 8 months in terms of GPA.

	Flipped classroom (n = 112)		Traditional learning (n = 110)
	Mean	SD	Mean	SD	P-value
Final skill scores	86.24	0.635	85.11	0.772	.258
Situational scores	3.04	0.232	1.91	0.233	.001
Skill scores	83.21	0.614	83.20	0.735	.996
Tier B	87.29	1.293	86.14	1.143	.506
Situational scores	3.24	0.416	2.29	0.427	.116
Skill scores	84.06	1.135	83.86	1.043	.896
Tier C	86.03	0.739	84.68	1.114	.308
Situational scores	2.86	0.298	1.92	0.304	.030
Skill scores	83.17	0.739	82.75	1.076	.587
Tier D	83.63	2.705	84.40	2.022	.841
Situational scores	3.75	0.818	0.50	0.500	.003
Skill scores	79.88	2.973	83.80	1.896	.752

Self-efficacy

The self-efficacy scores in the pre-class phase were not significantly different between the 2 groups (75.49 ± 8.52 vs 75.59 ± 8.73, P = .93; 77.3 ± 7.82 vs 77.31 ± 10.64, P = 1.00). Neither the learning ability nor the learning behavior scores were statistically significant in the pre-class phase (39.34 ± 5.63 vs 39.32 ± 5.81, P = .98; 36.15 ± 4.40 vs 36.27 ± 4.06, P = .83).

of the post-class phase (40.96 ± 5.16 vs 40.57 ± 7.24, P = .64; 36.34 ± 4.36 vs 36.74 ± 4.90, P = .53).

In terms of self-efficacy before and after class, the FC group showed significant improvements in total and learning abilities (75.39 ± 8.52 vs 77.30 ± 7.82, P = .03; 39.34 ± 5.63 vs 40.96 ± 5.16, P = .00). However, the TL group showed no significant improvement in terms of total, learning ability, or learning behavior scores.

In terms of performance stratification, there was no statistical significance between the FC and TL groups at all 3 levels during the pre-class phase. However, the students in tier D in the FC group had significantly higher self-efficacy scores than the TL group in tier D (78.88 ± 6.896 vs 68.50 ± 10.855, P = .032), especially for learning behavior (36.63 ± 3.114 vs 33.60, P = .048) (Table 8).

Table 8.

Pre- and post-class self-efficacy.

	Pre-class		Post-class
	Mean	SD	Mean	SD	P
Flipped classroom (n = 112)
Total	75.49	8.52	77.3	7.82	.03
Learning ability	39.34	5.63	40.96	5.16	.00
Learning behavior	36.15	4.40	36.34	4.36	.70
Traditional learning (n = 110)
Total	82.52	6.69	83.38	8.38	.42
Learning ability	41.32	4.03	42.37	5.43	.11
Learning behavior	41.20	3.43	41.01	4.17	.72

Learning Satisfaction

According to the learning satisfaction section described in the post-class questionnaire, a majority of students in FC considered the pre-class online learning was effective. And video quality and internet played the biggest role in it. Besides, most students in FC would use normal or 1.25× playback speed. The most important reason for the multiple playback speed was because they can fast-forward to the key points, the teacher's speaking speed varies, and uneven difficulty of content. The most favorite video format was slide + narration + Teacher's video + subtitles. And preferred video length was 10 to 15 min. The most common frequency to master the learning content was 1 to 2 times.

For FC-specific part, majority students prefer FC and their favorite FC activities are CPR skill practice, special scenario CPR training, and Q&A session. Most of the students thought FC met or even beyond their expectations.

For CPR specific part, most student were confident or very confident about practicing CPR on Manikin or a person. And they were willing or very willing to provide CPR when needed. However, they still had concerns for that. The most concerns were lack of experience and panic/stress (Table 9).

Table 9.

Learning satisfaction from post-class questionnaire.

		FC count	FC %
Pre-class online learning experience	Highly effective	36	32.10%
	Somewhat effective	42	37.50%
	Neutral	23	20.50%
	Somewhat ineffective	10	8.90%
	Ineffective	1	1.00%
Factors affecting learning motivation	Video quality	65	58.00%
(Select all that apply)	Internet speed	59	52.70%
	Platform friendliness	51	45.50%
	Video length	48	42.90%
	Test difficulty	43	38.40%
	Learning environment	36	32.10%
Most common playback speed	1× (normal)	48	42.90%
	1.25×	40	35.70%
	1.5×	10	8.90%
	1.75×	8	7.10%
	2×	6	5.40%
Reasons for using playback speed	Fast-forward to key points	51	45.50%
	Teacher's speaking speed varies	47	42.00%
	Uneven difficulty of content	43	38.40%
	To finish quickly for attendance	28	25.00%
	Video length varies	24	21.40%
Preferred video format	Slide + narration + teacher's Video + subtitles	47	42.00%
	Slide + narration	40	35.70%
	Animated videos	32	28.60%
	Slide + narration + 2 Teachers discussing	28	25.00%
Views needed to master content	1-2 times	54	48.20%
	3 times	36	32.10%
	4+ times	22	19.60%
Preferred video length	10-15 min	59	52.70%
	15-20 min	28	25.00%
	5-10 min	14	12.50%
	20-30 min	11	9.80%
Preference for flipped classroom	Very like	32	28.60%
	Like	48	42.90%
	Neutral	24	21.40%
	Dislike	8	7.10%
Favorite classroom activities	CPR skill practice	81	72.30%
(select all that apply)	Special scenario CPR Training	65	58.00%
	Q&A session	51	45.50%
	Group discussion	40	35.70%
	Video error correction	36	32.10%
Met expectations?	Exceeded expectations	28	25.00%
	Met expectations	62	55.40%
	Neutral	14	12.50%
	Below expectations	8	7.10%
Confidence in CPR	On Mannequin (Confident/very confident)	88	78.60%
	On real person (Confident/very confident)	59	52.70%
Willingness to perform CPR on real person	Very willing	40	35.70%
	Willing	36	32.10%
	Neutral	28	25.00%
	Unwilling	8	7.10%
Concerns about performing CPR	Lack of experience/skill	70	62.50%
(select all that apply)	Fear of legal responsibility	43	38.40%
	Panic/stress	51	45.50%
	Physical fatigue	36	32.10%
	Fear of infection	28	25.00%

Discussion

This study mainly aimed to explore the effectiveness of FCs in large classroom teaching of CPR skills, as well as the possible influential factors. This discussion is presented in the following order: effectiveness of FC, possible influential factors (self-efficacy and GPA stratification), retention of CPR skill, and class satisfaction.

Effectiveness of the flipped classroom

Compared to the TL group, the FC group scored better on the pre-class theoretical test, final theoretical test, and final skill test. Meanwhile, all of the difference in these scores in the FC group had higher statistical significance than those in the TL group. The results of this study were consistent with previous findings on the use of FCs in ophthalmology classes for nursing, dental, and medical students.⁷ Positive results were also found in a systematic review and meta-analysis of 22 studies on nursing skills in China²⁹ and on surgical education in the US regarding preconference quizzes.³⁰ A previous study reported that there was a significant improvement in CPR skills after a simulation-based FC.³¹

After the intervention in the FC, the self-rated learning ability and skill test scores were significantly higher than those of the comparison group. It is meaningful that the FC approach benefits students by improving learning goals, self-efficacy and social learning problem solving,⁷ cooperative spirit, practical ability, enjoyment of the class and participation interest,^32,33 expression and communication, ability to think and analyze problems,²⁹ deep understanding of learning, and effectiveness in accelerating learning.³¹

To achieve these benefits, one important method is designing FCs based on the Miller pyramid model.

The Miller pyramid model has been widely used in medicine assessment since it was first introduced by George Miller in 1990. The primary pyramid consists of 4 levels: Knows (knowledge), Knows How (competence), Shows How (performance), and Does (action).³⁴ The FC class was designed based on this pyramid. In the pre-class phase, the students could study the material on the specific platforms 1 week before classroom teaching. This segment involves the first pyramid level, student knowledge. At the same time, the process of flipping the classroom to transfer knowledge to students in advance was completed. To ensure the effectiveness of knowledge transfer, learning checklists and facilitated tests were also made available on the platform. The effect of these methods has been validated by a thesis that reported that using clear and comprehensive instructions is a key factor of FCs.³⁵ Through the back-end data of the platforms, the instructor could track the students’ learning at any time and promptly supervise the FC group to complete the online resources before the classroom learning began. The instructor's supervision guaranteed the FC group's possession of CPR knowledge, ie, the first level is effectively established and lays the foundation for the second level.

Since the students had already acquired basic theoretical knowledge before class, class time was used to carry out activities that internalize the knowledge and enable students to acquire intellectual competencies. The activities are considered successful factors in FCs, including discussion and explanation of high error rate questions on theoretical tests, discussion and explanation of pre-class questions, discussion and correction of incorrect actions as units, discussion of CPR operations involving scenarios, and reciting CPR flow charts.^8,35 All of the activities were attributed to the second level “Knows how,” which ensured that the students acquired the competence of knowing how to conduct CPR. Student-teacher discussions and group discussions and feedback enhance students’ understanding and internalization of knowledge.^8,35 This is different from the traditional teacher-centered to student-centered teaching philosophy.

Based on the competence level, the third level “Shows how” was built accomplished through in-class skill teaching. The students worked in units to perform CPR skills under the supervision and guidance of the teacher. The unit members were free to discuss and instruct each other on CPR operations, and each unit had a specific teacher to supervise and provide feedback to the students at any time.³⁵ Unlike in TL, the units where the teacher instructs were based on the needs of the students, continuous teacher supervision and feedback, which contributed to the accuracy of CPR skills operation in the FC groups.

The post-class scenario corresponds to the fourth level of Miller's pyramid, “Does” (action), which examined and predicted how the students would perform CPR in real-world scenarios.

The FC was effective in improving the students’ mastery and application of CPR skills, and the FC group had significantly higher pre-class theoretical scores than the TL group, indicating that the FC group had a higher foundation than the TL group at the first level of Miller's pyramid. Moreover, this result was reflected in the final theoretical scores, indicating that the FC group was significantly better than the TL group in terms of their ability to apply knowledge. In addition, in the final skills section of the test, the FC group scored higher than the TL group in both the operation of CPR and the handling of scenarios, indicating that the FC group had a stronger grasp of CPR operation skills and the use of CPR.

Second, the FC had a facilitating effect on improving student learning at higher levels. This facilitation was reflected in the students’ handling of different CPR scenarios, ie, scenario simulation. Scenario-based simulations can also promote higher-level learning for students. Nancy Sullivan found that compared to traditional clinical training, simulation accounts for a greater proportion of the highest level of Miller's pyramid “Does.”³⁶

In the final skills test, the FC group scored significantly higher than the TL group in the scenario simulation. In the TC classroom, the students better gained the ability to apply CPR skills by discussing and analyzing scenarios, and in the scenario portion of the final skills test, the students’ abilities were reflected in their scores for processing the various scenarios.

However, FCs may increase students’ time consumption and workload. Fen Tang found that students in an FC group felt more “burden and pressure.”³⁷ Before class, students must spend more time studying overwhelming material to prepare for class. To relieve the burden on students, a study list was used to guide the students to complete their studies, and the length of videos was kept below 10 min. This was based on Jinlei Zhang's research on the key factors of FCs, that is, videos of approximately 10 min are more appropriate (Zhang 2013). In the meantime, many pictures and videos were covered in the material to improve the concretion and interest in their studies. In addition, in the questionnaire, students expressed adoration for the videos. Some students commented that “the teacher in the lecture video is too interesting” and “I like the video so much.” However, RC Chick found that performance on preconference quizzes was increased without increasing preparation time after FC learning.³⁰ In addition, total study time and academic achievements were not correlated in this study. Therefore, whether FC increases the burden on students should be further investigated.

GPA stratification on performance

FC benefits all tiers of students in terms of various aspects. This is consistent with Han's finding, which suggested that the characteristics of participants should be analyzed before starting class.⁸

In this study, after FC learning, the B- and C-tier students showed significant improvements, especially in the situation simulation part, while the D-tier students mainly reflected mastery of basic knowledge. According to Miller's pyramid, the B- and C-tier students mainly obtained the higher levels of Shows how and Does, while the D-tier students obtained the lower levels of Knows and Knows how.^3–5 This could be explained as follows. Higher-tier students with strong learning ability can acquire basic knowledge more quickly before class and perform better in FCs for higher-level learning such as knowledge internalization, which is related to the fact that the discussion and scenario simulation in FCs can facilitate higher-level learning activities. Lower-tier students with weaker learning ability must devote more time to remembering and understanding the basic knowledge and then proceed to higher-level learning.

RC Chick found that junior residents benefit the most from the FC mode, as they are less experienced and are easily overwhelmed by multiple sources without clear guidance.³⁰ In a study of student performance in a pharmacotherapy oncology module before and after flipping the classroom, there were no significant differences between upper fiftieth percentile and the lower fiftieth percentile.³²

A study list designed to guide students was distributed to the students³⁵; however, the effect was mild in this study. According to the pre-class questionnaire about the study list, only 44 (39.3%) members of the FC group reported that the study list was very clear, 31 reported that they were most impressed with the study list, 2 (1.8%) members of the TL group reported that the study list was unclear, only 39 (35.5%) reported that it was very clear, and 26 were impressed with the study list. The results showed that the study list that was designed was insufficient to facilitate pre-class study. Therefore, designing a clear study list should be a key focus in the future.

Effect of FC in the long term

FC could benefit students at higher study levels in the long term. After 8 months, the FC group showed significantly higher situational scores than the TL group. The results of the long-term influence of FC are consistent with a thesis in which FC not only improves short-term physiology scores but also facilitates learning in follow-up medical classes in the long term. The long-term effect of FC can be explained as follows. FC students move up the Miller Pyramid from the pre-class phase, preparing the knowledge for deep learning, to the in-class phase, deeper understanding of and applying the knowledge through various activities, which could help enhance students’ long-term memory.³⁸ In addition, the entire learning process helps students develop a deeper understanding of the CPR procedure, as well as situational handling. Therefore, FC students could be better at refreshing their knowledge on their own. Furthermore, FCs train students in logic, analysis and knowledge application, which contribute to the significantly higher scores in the situational handling part.³⁹ In terms of stratification, the tier D FC students had lower skill scores but significantly higher situational scores. It could be that skill knowledge deteriorates more rapidly than situational knowledge, as reported by Riggs in a systematic review.⁴⁰

Self-efficacy

FC contributed to academic scores and CPR performance by improving the students’ self-efficacy. Self-efficacy may be weakly associated with improved skill.⁴⁰ On the one hand, when self-control was taught in this study, there was a significant increase in self-efficacy in the FC group, where learning ability was the main contributor to the increase in self-efficacy; there was no significant increase in the TL group. This suggests that the FC contributed to the improvement of learning ability by increasing the students’ self-efficacy. The results of this study are consistent with the findings of a study that used an FC approach to develop surgical competencies in veterinary students. This study showed a significant increase in student self-efficacy after FC learning, especially in the area of skill acquisition, but self-efficacy was associated with pre-FC grades rather than post-FC grades.⁴¹ Another study of nursing students in blended learning of CPR also found that the intervention group showed significantly higher self-efficacy after education and higher self-efficacy scores than the TL group.⁴²

Albert Bandura proposed self-efficacy in social-cognitive theory⁴³ and defined it as follows: “People's judgments of their capabilities to organize and execute classes of action required to attain designated types of performances.”⁴⁴ Anthony R. presented several ways to apply self-efficacy to practice as follows: (1) help students set clear and specific goals; (2) encourage the use of challenging and proximal goals; (3) provide honest, explicit feedback to increase students’ efficacy beliefs; (4) facilitate accurate calibration of self-efficacy; and (5) use peer modeling to build self-efficacy.⁴³ During the pre-class phase, the students were asked to complete a pre-class questionnaire, 1 question of which was about the goals of the class: “What are your expectations for this semester's Basic Life Support class?” “Able to learn some CPR skills and achieve proficiency” and “Perform CPR more fluently in emergency situations” were the most selected options, followed by “Completion of established academic tasks,” “Able to get high marks in this class” and finally “Challenge yourself to improve your overall capacity.” This shows that the students set their goals for this lesson primarily based on how to apply CPR skills. It also provides students with a clear class structure to help students succeed in FCs.³⁵ Second, during the in-class phase, the students could discuss and communicate with their peers in the units to receive help; encouragement; and frequent, focused, task-specific feedback, which is consistent with Howard Margolis's suggesting that using peer models, encouragement and feedback can improve self-efficacy.⁴⁵ Additionally, such interactive activities create a safe environment for students to freely ask questions and correct mistakes to succeed in mastering their CPR skills.³⁵ In addition, group activities and scenarios provide the basis for the construction of students’ social-cognitive skills.⁴⁶

On the other hand, based on the data analysis results, there was no difference in self-efficacy between the TL and FC groups before and after the class, and there was no correlation between self-efficacy and students’ prior GPA. The results of this study are consistent with the results of a study that correlated self-efficacy with neonatal CPR skills. That study suggested that the lack of correlation between participants’ self-efficacy and neonatal CPR skills was related to the participants overestimating or underestimating their proficiency in the skills. Additionally, work experience was found to be the only neonatal CPR-related factor.²⁰

In addition, in terms of performance stratification, FC is more beneficial for improving the self-efficacy of high-tier students than that of low-tier students. After class, all the tier B and C students of both groups improved. However, all the tier D students in both groups had lower self-efficacy. However, the number of D-tier students in the FC group decreased slightly more than the number in the TL group.

This may be due to the following reasons. First, self-efficacy is variable. Therefore, it is problematic to assess self-efficacy only twice before and after the class. Three or more waves of data are needed to establish a longitudinal pattern of the change.¹⁹ Furthermore, skills deteriorate within 3 months and then plateau from 3 to 6 months, which is attributed to uncertainty. In addition, self-efficacy is task specific, which means that people judge their capability based on the particular domain of function.⁴³ In this study, not only CPR skill self-efficacy but also information and technology self-efficacy should be considered. The latter could be defined as “judgment of one's capability to use a computer, which also plays a vital role in transferring knowledge to students.”⁴⁶ However, to ensure the validity and reliability of the questionnaire, this study directly applied the modified Chinese version of the general self-efficacy scale and did not refine the questionnaire according to CPR skills. This may also have led to inaccurate self-efficacy survey results. Future studies should focus on the specific feature of self-efficacy.

Learning satisfaction

FC helps improve class satisfaction. The result is consistent with current published papers. Hassan reports that simulation-based FC has significantly increasing satisfaction.¹³ Joseph found that FC improves nursing student's learning satisfaction in anatomy and physiology.³⁸ Khanittanuphong concluded the proportion of FC satisfaction was greatly increased.⁴⁷ A systemic review reports that a large portion of faculty and students are satisfied with FC.¹⁵ This could be the core of the FC that students become the center of the class. They study actively instead of the Didactic passive learning. They can study on their own pace. In pre-class, they can multiply the playback speed to get the key points, watch the videos multiple times. During the class, they practice CPR skill in a unit where they can discuss and learn simulation-based CPR from both the instructors and their peers. In the post-class questionnaires, the most favorite activities of FC are CPR skill practice, special scenario CPR training.

On the other hand, FC does increase student workload. We did see some students said they multiplied the speed of the videos to finish fast for the attendance. Beom concludes there is no significant difference between the FC and TL.¹⁴ A systemic review also reports some faculty and students does not prefer FC. It’ demanding for faculty to prepare materials and overloading for students to study by themselves. Besides, the internet and technology are the key to succeed in FC, which requires support when needed.¹⁵ This is consistent with our findings that video quality, internet speed, and platform friendness are playing an important role for students to finish the pre-class learning. Besides, we can see students need to watch the video multiple times to master the learning content, which could increase their workload. Therefore, the video length should be short, like 10 to15 min.

Limitations

This study was limited by its nonrandomized design; however, there was no statistically significant difference between the FC group and the TL group in terms of demographic characteristics and baseline self-efficacy. In addition, the measurement tool is also limited. Although the self-efficacy scale was analyzed for reliability and validity, its specificity for specific case-based simulation still needs to be strengthened. Moreover, in the future, fNIRS neuroimaging technology can be combined to quantify the effect of FC on students’ activation patterns in decision-making brain regions. Psychological mechanism can also be explored. Assessment of objectivity limitations can be analyzed by integrating multimodal data (eg, compression waveform consistency index, thoracic rebound kinetic parameters) from the Laerdal QCPR third-generation feedback system.

Conclusion

In a large classroom with over 100 students, FC significantly improved students’ theoretical CPR scores, particularly among high-achieving students (those with higher GPAs). Among these students, case-based questions received higher scores. On the skill-learning level, FC significantly enhanced students’ skill acquisition, especially in simulation scores. After 8 months, although there was no significant difference in CPR skills between the FC and TL groups, simulation scores remained significantly higher in the FC group. Additionally, FC significantly increased students’ self-efficacy and learning satisfaction.

Footnotes

Abbreviations

Acknowledgments

Thanks for Clinical Skill Center of Zhujiang Hospital, Southern Medical University, for their support.

ORCID iDs

Xin Wang

Min Mao

Dandan Qian

Ethical Approval

This study has been approved by the Ethics Committee of Zhujiang Hospital of Southern Medical University. The ethical approval number of this study is 2022-KY-176-02.

Consent to Participate

This study obtained written consent informed consent from all the participants and was approved by the Ethics Committee of Zhujiang Hospital of Southern Medical University. The ethical approval number of this study is 2022-KY-176-02.

Author Contributions

Xin Wang was responsible for research design, data analysis, data presentation, manuscript writing, and revision. Min Mao contributed to research design, conducted the research, and collected data. Dandan Qian was responsible for research design, conducting research data collection, data presentation, and drafting and revising the manuscript.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and publication of this article.

Data Availability Statement

The data of this study have been shared as attachments under the name of Research Data for review.

References

Virani

Alonso

Aparicio

, et al. Heart disease and stroke statistics-2021 update: a report from the American Heart Association. Circulation. 2021;143(8):e254–e743. doi:https://doi.org/10.1161/cir.0000000000000950. published Online First: 2021/01/28.

Liu

Wang

. Cardiopulmonary Resuscitation Specialized Committee Of Chinese Research Hospital Association. Zhonghua Wei Zhong Bing Ji Jiu Yi Xue2019;31(4):385-389. https://doi.org/10.3760/cma.j.issn.2095-4352.2019.04.001

Mok

. Teaching tip: the flipped classroom. J Inf Syst Educ. 2014;25(1):7.

Lee

Lim

Kim

. Development of an instructional design model for flipped learning in higher education. Educ Technol Res Dev. 2017;65:427–453.

Jakobsen

Knetemann

. Putting structure to flipped classrooms using team-based learning. Int J Teach Learn Higher Educ. 2017;29(1):177–185.

DeRuisseau

. The flipped classroom allows for more class time devoted to critical thinking. Adv Physiol Educ. 2016;40(4):522–528. doi:https://doi.org/10.1152/advan.00033.2016. published Online First: 2017/02/02.

Zhu

Lian

Engström

Use of a flipped classroom in ophthalmology courses for nursing, dental and medical students: a quasi-experimental study using a mixed-methods approach. Nurse Educ Today. 2020;85:104262 doi:https://doi.org/10.1016/j.nedt.2019.104262 [published Online First: 2019/11/24].

Han

. Flipped classroom: challenges and benefits of using social media in English language teaching and learning. Front Psychol. 2022;13:996294. doi:https://doi.org/10.3389/fpsyg.2022.996294. published Online First: 2022/10/11.

Day

. A gross anatomy flipped classroom effects performance, retention, and higher-level thinking in lower performing students. Anat Sci Educ. 2018;11(6):565–574. doi:https://doi.org/10.1002/ase.1772. published Online First: 2018/01/23.

10.

Vanka

Wali

. Flipped classroom in dental education: a scoping review. Eur J Dent Educ:. 2020;24(2):213–226. doi:https://doi.org/10.1111/eje.12487. published Online First: 2019/12/07.

11.

Hew

. Flipped classroom improves student learning in health professions education: a meta-analysis. BMC Med Educ. 2018;18(1):38. published Online First: 2018/03/17.

12.

Chan

ZCY

. A systematic review on critical thinking in medical education. Int J Adolesc Med Health. 2016;30(1):/j/ijamh.2018.30.issue-1/ijamh-2015-0117/ijamh-2015-0117.xml. Published 2016 Apr 18. doi:10.1515/ijamh-2015-0117.

13.

Hassan

Elsaman

SEA

. The effect of simulation-based flipped classroom on acquisition of cardiopulmonary resuscitation skills: a simulation-based randomized trial. Nurs Crit Care. 2023;28(3):344–352. doi:https://doi.org/10.1111/nicc.12816. published Online First: 20220708.

14.

Beom

Kim

Chung

Kim

Cho

. Flipped-classroom training in advanced cardiopulmonary life support. PLoS One. 2018;13(9):e0203114. doi:https://doi.org/10.1371/journal.pone.0203114. published Online First: 20180905.

15.

Barranquero-Herbosa

Abajas-Bustillo

Ortego-Maté

. Effectiveness of flipped classroom in nursing education: a systematic review of systematic and integrative reviews. Int J Nurs Stud. 2022;135:104327. doi:https://doi.org/10.1016/j.ijnurstu.2022.104327

16.

Panchal

Bartos

Cabañas

, et al. Part 3: adult basic and advanced life support: 2020 American Heart Association guidelines for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation 2020;142(16_suppl_2):S366–S468 doi:https://doi.org/10.1161/cir.0000000000000916 [published Online First: 2020/10/22].

17.

Lim

XMA

Liao

Wang

Seah

. The effectiveness of technology-based cardiopulmonary resuscitation training on the skills and knowledge of adolescents: systematic review and meta-analysis. J Med Internet Res. 2022;24(12):e36423. doi:https://doi.org/10.2196/36423. published Online First: 2022/12/16.

18.

Boysen-Osborn

Anderson

Navarro

, et al. Flipping the advanced cardiac life support classroom with team-based learning: comparison of cognitive testing performance for medical students at the University of California, Irvine, United States. J Educ Eval Health Prof. 2016;13:11. doi:https://doi.org/10.3352/jeehp.2016.13.11. published Online First: 2016/02/20.

19.

Klassen

JRL

. Self-efficacy beliefs of medical students: a critical review. Persp Med Educ. 2018;7(2):76–82. published Online First: 2018/02/28.

20.

Donohue

Underwood

Hoffman

. Relationship between self-efficacy and performance of simulated neonatal chest compressions and ventilation. Simul Healthcare: J Soc Simul Healthcare. 2020;15(6):377–381. doi:https://doi.org/10.1097/sih.0000000000000446. published Online First: 2020/05/07.

21.

Shin

Song

, et al. Public awareness and self-efficacy of cardiopulmonary resuscitation in communities and outcomes of out-of-hospital cardiac arrest: a multi-level analysis. Resuscitation. 2016;102:17–24. doi:https://doi.org/10.1016/j.resuscitation.2016.02.004. published Online First: 2016/02/24.

22.

Dick-Smith

Power

Martinez-Maldonado

Elliott

. Basic life support training for undergraduate nursing students: an integrative review. Nurse Educ Pract. 2021;50:102957. doi:https://doi.org/10.1016/j.nepr.2020.102957. published Online First: 2021/01/10.

23.

Bulfone

Fida

Ghezzi

, et al. Nursing student self-efficacy in psychomotor skills: findings from a validation, longitudinal, and correlational study. Nurse Educ. 2016;41(6):E1–e6. doi:https://doi.org/10.1097/nne.0000000000000285. published Online First: 2016/10/22.

24.

Carrick

Abdulrahman

Hankir

, et al. Randomized controlled study of a remote flipped classroom neuro-otology curriculum. Front Neurol. 2017;8:349. doi:https://doi.org/10.3389/fneur.2017.00349. published Online First: 2017/08/10.

25.

Beom

Kim

Chung

Kim

Cho

Flipped-classroom training in advanced cardiopulmonary life support. 2018.

26.

Merchant

Topjian

Panchal

, et al. Part 1: executive summary: 2020 American Heart Association guidelines for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation. 2020;142(16_suppl_2):S337–SS57. doi:doi:https://doi.org/10.1161/CIR.0000000000000918

27.

Liang

. Study On Achievement Goals, Attribution Styles and Academic Self-efficacy of Collage Students. Central China Normal University; 2000.

28.

Pintrich

De Groot

. Motivational and self-regulated learning components of classroom academic performance. J Educ Psychol. 1990;82(1):33–40. doi:https://doi.org/10.1037/0022-0663.82.1.33

29.

Chen

Nie

, et al. The effectiveness of a flipped classroom on the development of Chinese nursing students’ skill competence: a systematic review and meta-analysis. Nurse Educ Today. 2019;80:67–77. doi:https://doi.org/10.1016/j.nedt.2019.06.005. published Online First: 2019/07/07.

30.

Chick

Adams

Peace

, et al. Using the flipped classroom model in surgical education: efficacy and trainee perception. J Surg Educ. 2021;78(6):1803–1807. doi:https://doi.org/10.1016/j.jsurg.2021.05.008. published Online First: 2021/07/03.

31.

Hassan

Elsaman

SEA

. The effect of simulation-based flipped classroom on acquisition of cardiopulmonary resuscitation skills: A simulation-based randomized trial. Nurs Crit Care. 2023;28(3):344-352. doi:https://doi.org/10.1111/nicc.1281. published Online First: 2022/07/09.

32.

Bossaer

Panus

Stewart

Hagemeier

George

. Student performance in a pharmacotherapy oncology module before and after flipping the classroom. Am J Pharm Educ. 2016;80(2):31. published Online First: 2016/04/14.

33.

Marshall

Nykamp

Momary

. Impact of abbreviated lecture with interactive mini-cases vs traditional lecture on student performance in the large classroom. Am J Pharm Educ. 2014;78(10):189. published Online First: 2015/02/07.

34.

Cruess

Steinert

. Amending Miller’s pyramid to include professional identity formation. Acad Med. 2016;91(2):180–185. doi:https://doi.org/10.1097/acm.0000000000000913

35.

Sointu

Hyypiä

Lambert

Hirsto

Saarelainen

Valtonen

. Preliminary evidence of key factors in successful flipping: predicting positive student experiences in flipped classrooms. High Educ. 2023;85(3):503–520. doi:https://doi.org/10.1007/s10734-022-00848-2

36.

Sullivan

Swoboda

Breymier

, et al. Emerging evidence toward a 2: 1 clinical to simulation ratio: a study comparing the traditional clinical and simulation settings. Clin Simul Nurs. 2019;30:34–41.

37.

Tang

Chen

Zhu

, et al. Comparison between flipped classroom and lecture-based classroom in ophthalmology clerkship. Med Educ Online. 2017;22(1):1395679. doi:https://doi.org/10.1080/10872981.2017.1395679. published Online First: 2017/11/04.

38.

Joseph

Roach

Natarajan

Karkada

Cayaban

ARR

. Flipped classroom improves Omani nursing students performance and satisfaction in anatomy and physiology. BMC Nurs. 2021;20(1):1. published Online First: 20210102.

39.

Luo

Feng

Xiang

. Short- and long-term influences of flipped classroom teaching in physiology course on medical Students’ learning effectiveness. Front Public Health. 2022;10:835810. doi:https://doi.org/10.3389/fpubh.2022.835810. published Online First: 2022/04/15.

40.

Riggs

Franklin

Saylany

. Associations between cardiopulmonary resuscitation (CPR) knowledge, self-efficacy, training history and willingness to perform CPR and CPR psychomotor skills: a systematic review. Resuscitation. 2019;138:259–272. doi:https://doi.org/10.1016/j.resuscitation.2019.03.019. published Online First: 2019/04/01.

41.

Decloedt

Franco

Martlé

Baert

Verwulgen

Valcke

. Development of surgical competence in veterinary students using a flipped classroom approach. J Vet Med Educ. 2021;48(3):281–288. published Online First: 2020/02/29.

42.

Moon

Hyun

. Nursing students’ knowledge, attitude, self-efficacy in blended learning of cardiopulmonary resuscitation: a randomized controlled trial. BMC Med Educ. 2019;19(1):414. published Online First: 2019/11/11.

43.

Artino

Jr . Academic self-efficacy: from educational theory to instructional practice. Persp Med Educ 2012;1(2):76–85 doi:https://doi.org/10.1007/s40037-012-0012-5 [published Online First: 2013/01/15].

44.

Bandura

Social Foundations of Thought and Action. NJ, 1986, 23–28.

45.

Margolis

Mccabe

. Improving self-efficacy and motivation:what to do, what to say. Interv Sch Clin. 2006;41(4):218–227. doi:https://doi.org/10.1177/10534512060410040401

46.

Cai

Yang

Gong

MacLeod

Zhu

. Understanding the continued use of flipped classroom instruction: a personal beliefs model in Chinese higher education. J Comput Higher Educ. 2019;31(1):137–155. doi:https://doi.org/10.1007/s12528-018-9196-y

47.

Khanittanuphong

Iamthanaporn

Bvonpanttarananon

. The impact of the transition from flipped classroom to online lectures on learning outcomes and student satisfaction in a rehabilitation medicine clerkship during the COVID-19 pandemic. BMC Med Educ. 2022;22(1):885. published Online First: 20221220.