ObjectiveTo explore the new hospital management method about diagnosis-related groups (DRGs), and put forward some strategic suggestions.MethodsIn March 2019, using literature research method, relevant documents were consulted to understand the research policy and background. In April 2019, the DRGs data and first pages of medical records of a tertiary grade A hospital in 2018 were obtained through field survey. The DRG with the largest quantity of patients was selected, and then the top two treatment centers ranked by the quantity of patients were selected for analysis.ResultsA total of 11 936 patients’ face sheets for medical records were investigated, covering 18 major disease categories (MDCs) and 93 DRGs. Treatment center A and B were the top two treatment centers ranked by the quantity of patients, covering 8 MDCs and 34 DRGs. There were 1 116 patients in treatment center A and 470 patients in treatment center B, with the same case-mix index (0.820). There was no statistically significant difference in the average length of hospital stay between the two treatment centers (t=−1.926, P=0.054). The average hospitalization expenses [(45 902.64±30 028.22) vs. (40 763.34±25 141.12) yuan, t=−3.260, P=0.001], drug expenses [21 481.43 (10 663.16, 34 251.64) vs. 11 740.36 (5 818.37, 21 572.09) yuan, Z=−9.812, P<0.001], and other expenses [138.00 (84.00, 178.00) vs. 120.00 (72.00,155.28) yuan, Z=−3.573, P<0.001] in treatment center B were higher than those in treatment center A. But the medical technology expenses [(7 319.11±3 781.52) vs. (10 995.61±4 784.55) yuan, t=12.324, P<0.001] and nursing expenses [(578.42±226.82) vs. (882.99±781.63) yuan, t=8.187, P<0.001] in treatment center B were lower than those in treatment center A.ConclusionsThe disease diagnosis and treatment specifications need to be strengthened and the process needs to be optimized. In the next hospital management, we should pay attention to key indicators to improve performance appraisal, standardize the diagnosis and treatment process to promote clinical path, and mine deep data to make performance management detailed.