Table 3.

Description of the studies included in the review. VAM: voice advisory mannequin.

Article	Population (Sample)	Study Groups and Teaching Methods	Assessment Methods	Immediate Results	Results after Refresher/Retention Period
An evaluation of objective feedback in basic life support (BLS) training [6]	Birmingham health centre students (n = 100)	Teaching in BLS and AED (8 h). Subsequently two groups with loss of two participants. (1) Control group: CPR practice without feedback. (2) Intervention group: CPR practice with a skill reporter	Sequence checklist evaluated by instructors and mannequin with a skill reporter. Use of AED not evaluated	Statistically significant differences since participants with practice with skill reporter obtained better averages in depth and % of correct compressions and better administered air volumes. There were no differences in terms of execution of the algorithm.	At 6 weeks, percentage of correct compressions was higher in participants who practised with a skill reporter. Volume administered increased in both groups. There were no differences in terms of execution of the algorithm.
A randomized controlled trial comparing traditional training in cardiopulmonary resuscitation (CPR) to self-directed CPR learning in first year medical students: The two-person CPR study [9]	First-year medical students (n = 180); neligible = 240. Those who had done BLS training in the past 5 years were excluded	(1) Self-directed learning with VAM for 2 h. (2) Normal mannequins and theory on a DVD for free viewing and practice for 2.5 h.Groups 1 and 2 could practise 10 days prior to the measurement. (3) Traditional group with instructor for 4–5 h on the day of assessment.	Simulations checklist evaluated by instructors and mannequins with a skill reporter	No statistically significant differences were established. Traditional group obtained better results for knowledge in simulations than the others. Main failure: misuse of AED. In terms of skills, there were significant differences only in the compressions: minute ratio, which was higher in the traditional group.	No
Effects of two retraining strategies on nursing students’ acquisition and retention of BLS/AED skills: A cluster randomised trial [11]	Nursing students based in Almería (Spain) and the United Kingdom (n = 177). Prior 3-h course, 3 months before	(1) Self-directed group: 4 h refresher in which required aspects were reviewed. (2) Instructor-led group: same time frame, but the instructor set goals and provided teaching in use of the material and final evaluation.	Pre- and post- intervention knowledge questionnaire, confidence questionnaires and mannequins with the skill reporter.	Both groups improved their skills, knowledge and confidence	At 3 months, there was a new assessment, in which skills and knowledge were improved, with the self-directed group obtaining better results.
Comparison of two instructional modalities for nursing student CPR skill acquisition [12]	Nursing students (n = 604) with prior BLS knowledge	(1) Mannequin-based group: theoretical class and training with mannequin (VAM). (2) Traditional instruction group: 4 h with instructor who taught knowledge and trained in skills with mannequin.	Mannequin with a skill reporter	The sample that undertook self-learning with mannequin obtained best results in all of the individual skills except frequency and compressions: ventilations ratio.	No
Retention of basic life support knowledge, self-efficacy and chest compression performance in Thai undergraduate nursing students [13]	Third-year university nursing students based in Thailand (n = 30), randomised with neligible = 180	One group, comprising women only. They had knowledge acquired 1 year previously. 1 h. BLS video followed by CPR practice with 1 and 2 resuscitators for 20 min.	Questionnaire on pre- and post- knowledge, questionnaire on confidence and mannequin with a skill reporter	No one passed the knowledge pre-test but 100% passed the post-test. In terms of confidence, an increase with the pre-test was noted. Motor skills were only recorded post-course, with 100% results for hand placement and decompression.	At 3 months, there was a new knowledge and skills test. Worse results were obtained for knowledge (30% passed), but there were better results in all skills compared to the other assessment (without refreshers). Values were maintained in terms of confidence.
Repetitive sessions of formative self-testing to refresh CPR skills: A randomised non-inferiority trial [16]	Third-year medical students based in Ghent (n = 218). People with knowledge excluded: nfinal = 196	After excluding from sample those with appropriate skills knowledge, a computer created two groups: (1) performed self-assessments in BLS training; (2) same training, and also practised CPR. Had 6 weeks to be proficient in CPR skills. Retention at 6 months	Mannequin with a skill reporter (2 min).	At the end of the first 6 weeks, there were no significant differences between the groups that were deemed proficient in CPR.	At 6 months, decrease in the number of people in both groups that performed quality CPR. Despite this, those who did not practice during the first 6 weeks obtained better results in depth and ventilations.
High-fidelity simulation effects on CPR knowledge, skills, acquisition, and retention in nursing students [17]	First-year nursing students (n = 90)	(1) Control Group: 4 h theory and traditional training with AED. (2) Intervention group: 4 h theory and training with high-fidelity simulation with AED.	Pre- and post- knowledge questionnaires and evaluation of skills by instructor during cardiac arrest activity.	Significant differences in the skills and knowledge among groups, with improvements in the intervention group.	New measurement at 3 months, in which knowledge and skills remained better in the intervention group.
Effects of monthly practice on nursing students’ CPR psychomotor skill performance [18]	Nursing students from different universities in U.S. neligible = 727; nfinal = 606	All students were trained in BLS at their universities. After this: (1) Control group: no practice; (2) Intervention group: CPR skills practice, 6 min a month, with VAM. At 3, 6, 9, and 12 months a random subgroup of each main group underwent measurement. Another subgroup, 12R, was also created, which was given a BLS refresher and subsequent measurement at 12 months.	Mannequin with a skill reporter	Throughout the study, there were no differences found in the compressions: minute ratio, hands placement and volume: minute. There were differences in depth and volume administered, with decreases in the control group.	In the 12R groups, which had a refresher, there were no differences, since both groups received a refresher in CPR knowledge and skills.
Using a serious game to complement CPR instruction in a nurse faculty [19]	Nursing students based in Norway (n = 109)	(1) Three control groups (A, B, C). Pre-test and practice with simulation with mannequins and with AED. (2) Five LIfe Support Simulation Activities (LISSA-2) groups (D, E, F, G, H). Tutorial focused on serious game simulation program with problems to be solved before the intervention.	Checklist questionnaires and mannequins with a skill reporter	Improved knowledge among those belonging to LISSA groups. In terms of skills, there were no differences between the groups. They conclude that this is a good method to support theory but that it does not improve skills.	No
Voice advisory manikin versus instructor facilitated training in cardiopulmonary resuscitation [22]	Medical students based in Copenhagen (n = 43). Students undertook course 1 year before.	(1) Instructor-led group: received teaching in skills for 32 min. (2) Mannequin-based group: use of a VAM for 5 min	Measurement of skills through a skill reporter, pre and post, 2 min	In the post-measurement, the instructor-led group obtained better results for skills.	At 3 months, new 2-min measurement, in which no differences between groups were found.
Pre-training evaluation and feedback improve medical students’ skills in basic life support [23]	Third-year medical students based in Sichuan (China) (n = 40)	(1) Control group: 45 min theory class followed by 45 min of traditional training with mannequin; (2) Intervention group: same theory class, followed by pre-assessment after simulation with instructor feedback for 15 min, followed by 30 min training with mannequin.	Questionnaires on prior knowledge and mannequin with a skill reporter	There were no differences upon analysis of theoretical knowledge. After evaluation with a skill reporter, better results with significant differences in the intervention group, except in hand positioning, which was the same.	No