2.1 Study location
The research was conducted from January 2018 to December 2020 at the Medical University of Lublin, Eastern Poland, during First Aid classes on basic life support (BLS) in the following faculties: dietetics, cosmetology, physiotherapy, pharmacy, public health, and medical emergency.
2.2 Study population
All seven hundred and nine students receiving training participated in the study (100%). After setting aside incomplete questionnaires, 694 students (98%) qualified for further evaluation, of which 416 were from the Faculty of Health Sciences (dietetics, physiotherapy, medical emergency, public health), and 278 came from the Faculty of Pharmacy (pharmacy, cosmetology).
2.3 Study preparation
CPR training began with a theoretical introduction to first aid and CPR. The algorithm for OHCA management in an adult was presented (compliant with the 2015 European Resuscitation Council guidelines) [13]. High-quality CPR was emphasized, and students were familiarized with all the parameters that should be achieved for optimum CPR performance. A demonstration was conducted on an adult Resusci Anne® Skill ReporterTM mannequin from Laerdal Medical with a basic control panel. Afterwards, the students practiced CPR on the same mannequin, using the same control panel. The students could observe their CPR's quality on two types of feedback, presented in figure 1 and 2.
The basic feedback device contained a panel with LEDs that lit up, informing the students about the following: compression depth level (correct or incorrect), the amount of air during rescue breaths (correct or incorrect), incorrect hand placement for chest compressing - signaled on an image of the chest on the panel, incorrect airway patency - signaled by the diode placed on the neck shown on the figure indicator.
The feedback received by the basic control panel indicators (figures 1 and 2) consisted of lighting up the indicator lamps, where green indicated a well-done activity, while orange and red indicated elements that needed improvement. After connecting the mannequin to the laptop, SkillReporterTM displayed the exact results of ongoing activities on the monitor screen. Consequently, during chest compression, the student could observe an amplitude graph showing the exact depth of each individual compression, and the pace of chest compressions (compressions per minute). During ventilation, the student could observe an amplitude graph showing at what rate and amount of air is given. Students could constantly observe the effects of their actions and incorrect measures at a given moment and try to correct areas of improvement on a regular basis. This detailed feedback gave the opportunity to adjust their hand movements to achieve better quality compressions. (figures 1 and 2)
During student’s CPR performance, the instructor observed the following: careful hand placement for resuscitation, the quality of chest compressions performed, and the quality of rescue breaths.
All methods were carried out in accordance with relevant guidelines and regulations.
2.4 Next phase
In the next class, the same students continued to learn CPR with the same mannequin. However, this time with an altered form of feedback. The mannequin was connected to a laptop, from which the indicators of resuscitation were displayed on the screen to the class using a projector. The outcome of the resuscitation was displayed synchronously in milliseconds response. Figures 3, 4 and 5 presents a picture of the feedback screen, as seen by the students. (figures 3-5)
A detailed record presented the following:
- Current and average compression rate,
- The exact number of compressions implemented,
- The ratio of the number of chest compressions to the number of rescue breaths,
- Compression depth given in millimeters,
- Relaxation error,
- Hand position error indicating the direction of the wrong hand position,
- The airway patency,
- The exact amount of air supplied,
- The speed with which the air is introduced during rescue breaths)
- Resuscitation time.
Students were able to observe their performance and correct the quality of CPR activities on an ongoing basis. Moreover, the instructor conducting the classes observed the students and analyzed their behavior during exercises with the device and gave detailed feedback.
2.5 Questionnaire
A questionnaire was developed using multiple strategies. First, the authors conducted a literature review to identify the critical dimensions for developing a questionnaire to be used after practical sessions and for the aim of this study. For the review purpose, the following keywords: first aid, education, basic life support, cardiopulmonary resuscitation, feedback device, simulation, quality, skill retention, motivation, chest recoil, mannequins alone or in combination were used. The acquired data from PUBMED, SCOPUS, and WEB OF SCIENCE were organized, categorized, and mapped to create the questions in the questionnaire. Secondly, a total nine questions were chosen through Nominal Group Technique [14] with five students from a local university to evaluate the questionnaire (Supplementary File 1). These participants were later excluded from the study, and their responses were not used in the final analysis. Validity was reviewed based on a combination of logic, relevance, comprehension, legibility, clarity, and usability before the final administration.
2.6 Statistical analysis
Statistical analysis was performed using SPSS Statistics version 25. A level of α <0.05 was considered statistically significant. Due to the ordinal and nominal nature of the measurement of variables considered essential for the study, non-parametric tests were used. The median, mean, and standard deviation values, percentage distribution, and mean rank were used to describe the obtained results' distribution. The mannequin dependent score measurements were compared using the Wilcoxon test. The Glass' delta expressed the effect size for the calculated difference. The Mann-Whitney U test was used to compare the two groups in terms of the ordinal variable, and the H Kruskal-Wallis test was used to compare more groups. The Mann-Whitney U test was also used as a post hoc test. Its significance level was adjusted using the Bonferroni correction. The effect size for the calculated differences was expressed using Glass's r coefficient and eta - square. Comparing the groups in terms of the qualitative variable was made using the chi-square test, with a Z test with a Bonferroni correction used to compare the columns' proportions. The effect size for the calculated differences was expressed using Cramer's V coefficient.