BackgroundHigh-quality chest compression is essential for successful cardiac arrest resuscitation. High-quality cardiopulmonary resuscitation (CPR) can effectively improve the survival rate of patients with cardiopulmonary arrest. However, bystanders untrained in cardiopulmonary resuscitation may provide inadequate chest compressions. Previous studies have shown that the use of feedback devices in training alone or in simulated cardiopulmonary arrest alone can improve cardiopulmonary resuscitation. This study aims to determine whether using an audiovisual feedback (AVF) device during CPR training or a simulated cardiopulmonary arrest (CA) scenario would be more effective in improving the quality of chest compressions (CC).MethodsWe use a prospective, randomized, 2 × 2 factorial design trial. A total of 160 participants from Wuhan University and senior clinical medicine undergraduates who had not participated in any CPR training before and had no actual CPR experience are recruited. Each participant is randomized to 1 of 4 permutations, including AVF device vs. no AVF device during CPR training and AVF device vs. no AVF device during simulated CA. Main outcomes and measures are the depth, the percentage of CCs with correct depth (5–6 cm), the rate of CCs, and the percentage of CCs with the correct rate (100–120 cpm).ResultsThe use of the AVF device during simulated CA resulted in improved CC quality. In CA without AVF device, the average compression depth and the percentage of adequate depth with AVF device are 5.1 cm, 5.0 cm and 55.5%, 56.3%, respectively, which are higher than those without AVF device (4.5 cm, 4.7 cm and 32.8%, 33.6%). (p = 0.011, p = 0.000, both < 0.05).Compared with CA without AVF device, the average compression rate and the percentage of adequate rate with AVF device are 112.3 cpm, 111.2 cpm and 79.4%, 83.1%, respectively. The average compression rate and the percentage of adequate rate without using the AVF device are 112.4 cpm, 110.3 cpm and 71.5%, 68.5%, respectively. (p = 0.567 > 0.05, p = 0.017 < 0.05)Although the average compression rate in group D is slightly lower than that in group C, the percentage of suitable frequency with the feedback device is still higher than that without AVF device.ConclusionUsing a feedback device during simulated cardiopulmonary arrest is more effective in improving cardiopulmonary resuscitation than during training.