目的：对某中医院医师端与编码员复核后的主要诊断编码在DRG分组模拟器中入组情况进行对比，分析DRG入组有偏差的原因，并针对性提出改进建议。方法：随机抽取某中医院2022年7月1日-2022年12月31日病案DRG数据，分析版块中体现编码员复核主要诊断编码后（复核后）与医师端填写的主要诊断编码（复核前）病例DRG入组不一致的病案2 805份。利用配对t检验对复核后与复核前数据进行分析。结果：各科室DRG病组总点数在复核后整体都有提升，其中主要诊断编码复核后与复核前对比，DRG病组点数上升1 819份，占比（64.85%）；复核前后对比入组DRG病组点数无差异262份，占比（9.34%）；复核后与复核前对比DRG病组点数降低724份，占比（25.81%），DRG病组点数复核前后变化有统计学意义（t=-10.82，P < 0.001）。住院病案首页主要诊断编码选择不准确的原因归纳为编码误选为肿瘤形态学（1.78%）、未明确患者来院治疗目的（87.02%）及主要诊断选择太笼统（11.20%）。结论：通过加强临床医师疾病编码培训、与病案室的沟通以及病案信息化管理，有效降低主要诊断编码复核前后DRG分组的偏差，提高病案主要诊断编码准确率。

Objective：The purpose of this study is to compare the main diagnostic codes reviewed by TCM doctors and coders in the HIS system of a certain traditional Chinese medicine hospital in the DRG grouping simulator，analyze the reasons for the deviation in DRG grouping，and propose targeted improvement suggestions. Method：Randomly select 2 805 medical records in the DRG data analysis section of a certain traditional Chinese medicine hospital from July 1，2022 to December 31，2022，which showed that the DRG enrollment of the case was inconsistent between the main diagnostic code filled in by the coder（after review）and the main diagnostic code filled in by the physician（before review）. Analyze the data after and before the review using paired t-tests. Results：showed that the total number of DRG disease groups in each department showed an overall increase after review，with 1 819 cases（64.85%）showing an increase in DRG disease group points compared to before review for the main diagnostic code；262 samples（9.34%）showed no difference in the number of DRG disease points before and after the review；The number of DRG disease group points decreased after review compared to before review was 724，accounting for 25.81%. The change in DRG disease group points before and after review was statistically significant（t=-10.82，P < 0.001）. The reasons for the inaccurate selection of the main diagnostic code on the homepage of the inpatient medical record can be summarized as incorrect selection of the code for tumor morphology（1.78%），unclear patient treatment purpose（87.02%），and too general selection of the main diagnostic code（11.20%）. Conclusion：By strengthening the training of clinical physicians on disease coding，strengthening communication between clinical physicians and medical record rooms，strengthening information management of medical records，reducing the deviation of DRG grouping before and after main diagnostic review，and improving the accuracy of main diagnostic codes in medical records.