Objective To evaluate the effect of residency training on trainees' competence and to provide a reference for improving the training policy. Methods Totally 15 830 teachers from 310 resident training institutions across the country were sampled and surveyed by the entrustable professional activity scale to evaluate the competence of the trainees before and after the training. The Wilcoxon rank sum test was used to compare the scores of each index with the expected score and the Marginal Homogeneity test was used to compare the scores of each index before and after the training. Results When entering training institutions, clinical decision making, critical illness identification and treatment, patient transfer and handover, essential skills of specialty, notification of bad news, clinical teaching, and public health incident response were all scored at 1 (1,2). The scores of the other 8 indicators were all 2 (1, 2). At the end of the training, the residen's scores of critical illness identification and treatment, and public health incident response were all 3 (3, 4) and the other indicators are 4 (3,4) or 4(4,4) or 4(4,5).The scores of all indicators were higher than those when entering the training (such as the scores of receiving patients were 4(4, 4) vs. 2(1, 2), χ²=107.81, P<0.001). Conclusions After training, residents' competence has been significantly improved. However, whether trained or not, the scores of the indicators still fail to reach the level of expectations. The collaboration in clinical performance and education should be strengthened and the integration of under-graduate education and post-graduate education should be further designed to ensure that the residents are able to achieve the training goals before and after residents' training.