ObjectiveTo analyze the long-term effect on cardiopulmonary resuscitation skill between video-led and scene simulation training and traditional instructor-led courses in medical student with eight-year program.MethodsNinety-nine medical students with eight-year program who studied in Peking Union Medical College were trained in cardiopulmonary resuscitation skill from January to February 2018. They were randomly divided into two groups, 53 students participated in basic life support course training, which belonged to video-led and scene simulation training as the trial group, and 46 students were trained by traditional instructor-led courses as the control group. In January 2019, the above 99 students were re-evaluated for cardiopulmonary resuscitation, and the outcome of cardiopulmonary resuscitation skill test in total scores and sub-items scores between two groups were compared. The data were analyzed using t test and Wilcoxon rank sum test.ResultsThe total average scores of the trial group (8.02±1.11) was higher than that of the control group (6.85±1.50) (P<0.05). The sub-items scores of the trial group in the three aspects of on-site assessment, chest compressions and simple respirators (1.64±0.37, 3.38±0.46, 1.52±0.58) were higher than those of the control group (1.33±0.45, 2.80±0.76, 1.19±0.58) (P<0.05). In terms of opening airway, there was no significant difference in scores between the two groups (1.02±0.47 vs. 1.10±0.45, P>0.05). The excellent rate of the trial group (60.3%) was significantly higher than that of the control group (30.4%) (P<0.05), and the unqualified rate (5.6%) was significantly lower than that of the control group (21.7%) (P<0.05).ConclusionsThe video-led and scene simulation training has a better effect on cardiopulmonary resuscitation skills acquisition and long-term maintenance than traditional instructor-led courses for medical student with eight-year program.