Objectives: To investigate whether a real-time visual feedback device could improve the quality of chest compression (CC), and, if so, whether the mechanism is associated with dynamic indexes such as velocity and acceleration.
Methods: A self-control trial of 2-minutes CC on a manikin by trained rescuers compared the quality of CC without or with a visual feedback device. Demographic characteristics were recorded and CC metrics for the two tests were computed. Multivariable linear regression analyses were performed to examine the impact of variables on rate of qualified chest compression (RQCC). Multivariable logistic regression was performed to determine independent risk factors for achieving qualified chest compression (QCC) in the second test. 
Results: A total of 159 participants (average age: 29.36±9.0 years, 80 (50.3%) men) were recruited. RQCC of the second test was significantly greater than that of the baseline test. Multivariable linear regression analysis showed that Vcompression and acompression were independent risk factors for RQCC for both tests. The mean Vcompression and acompression of the second test were significantly greater than those of the baseline test. However, Vcompression was the only independent risk factor predicting QCC achievement during the second test. ROC curve analysis showed the area under curve (AUC) was 0.84, and the optimal cut-off value of Vcompression was 39.48 cm/s.
Conclusions: Increasing the Vcompression and acompression might improve the quality of simulated CC and should be recommended to improve QCC. Only Vcompression was an independent risk factor for achieving QCC during CC with a visual feedback device.

7 - Increasing the compression velocity and acceleration improved the quality of simulated chest compression under the direction of real time visual feedback device