摘要目的:探讨混班制教学对于提升不同生源的全科医生临床诊疗思维能力的应用。方法:2021年9月至12月，选择同济大学医学院52名全科医学方向硕士作为研究对象，根据学习背景及执业经历分为住培组、社区组、专科组。在《全科临床诊疗思维》课程中，对全体学员进行混班制教学，在课程前后通过病史采集、体格检查、医患沟通和病历书写4站客观结构化临床考试（the objective structured clinical examination，OSCE）测试来评价教学效果。采用SPSS 22.0软件对收集数据进行统计分析。计量资料以（均数±标准差）表示，组间比较采用方差分析或Kruskal-Wallis检验；组内前后比较根据差值是否符合正态分布，采用配对样本 t检验或Wilcoxon符号秩和检验。计数资料以频数和百分数表示。 结果:从课程前后得分来看，体格检查方面社区组、专科组、住培组学员课程前分别为（44.88±9.17）分、（45.85±8.18）分、（42.38±15.30）分，课程后分别为（78.06±12.11）分、（68.65±13.10）分、（76.44±16.46）分，差异有统计学意义（ t=-12.49、-6.43、-5.98； P<0.001）；医患沟通方面社区组和住培组课程前分别为（63.00±13.84）分、（60.00±15.50）分，课程后分别为（78.69±7.10）分、（70.88±6.44）分，差异有统计学意义（ t/ Z值分别为-4.33、-2.12； P<0.05）；病历书写方面3组学员课程前分别为（60.19±17.96）分、（47.77±12.59）分、（37.00±15.50）分，课程后分别为（79.69±14.25）分、（80.31±12.93）分、（75.88±15.88）分，差异有统计学意义（ t=-2.96、-6.26、-7.25； P<0.05）。各模块得分组间比较来看，培训前病历书写得分社区组为（60.19±17.96）分、专科组为（47.77±12.59）分、住培组为（37.00±15.50）分，差异有统计学意义（ χ2=13.04， P=0.001）；培训后医患沟通得分社区组为（78.69±7.10）分、专科组为（63.46±19.40）分、住培组为（70.88±6.44）分，差异有统计学意义（ χ2=10.13， P=0.006）。 结论:混班制教学在全科临床诊疗思维课程中不仅能够提升教学效率、节约教学资源，还能达到扬长避短、优势互补的作用。

abstractsObjective:To investigate the application of blended team-based learning (BTBL) in improving the clinical reasoning ability of general practitioners from different sources.Methods:From September to December in 2021, 52 postgraduates of general practice in Tongji University School of Medicine were selected as subjects, and according to their study background and practice experience, they were divided into residency training group, community group, and specialist group. BTBL was adopted for all students in the course of clinical reasoning in general practice, and a 4-station objective structured clinical examination (OSCE) test was conducted before and after the course to evaluate teaching effectiveness from the aspects of medical history collection, physical examination, doctor-patient communication, and medical record writing. SPSS 22.0 was used to perform a statistical analysis of data. Continuous data were expressed as mean±standard deviation, and an analysis of variance or the Kruskal-Wallis test was used for comparison between groups; the paired samples t-test or the Wilcoxon signed rank sum test was used for comparison of normally distributed or non-normally distributed data within each group. Categorical data were expressed as frequency and percentage. Results:There was a significant change in the score of physical examination after the course in the community group (44.88±9.17 vs. 78.06±12.11, t=-12.49, P<0.001), the specialist group (45.85±8.18 vs. 68.65±13.10, t=-6.43, P<0.001), and the residency training group (42.38±15.30 vs. 76.44±16.46, t=-5.98, P<0.001). There was a significant change in the score of doctor-patient communication after the course in the community group (63.00±13.84 vs. 78.69±7.10, t/Z=-4.33, P<0.05) and the residency training group (60.00±15.50 vs. 70.88±6.44, t/Z=-2.12, P<0.05). There was also a significant change in the score of medical record writing after the course in the community group (60.19±17.96 vs. 79.69±14.25, t=-2.96, P<0.05), the specialist group (47.77±12.59 vs. 80.31±12.93, t=-6.26, P<0.05), and the residency training group (37.00±15.50 vs. 75.88±15.88, t=-7.25, P<0.05). The score of medical record writing before the course was 60.19±17.96 in the community group, 47.77±12.59 in the specialist group, and 37.00±15.50 in the residency training group, with a significant difference between the three groups ( χ2=13.04, P=0.001); after the course, the score of medical record writing was 78.69±7.10 in the community group, 63.46±19.40 in the specialist group, and 70.88±6.44 in the residency training group, with a significant difference between the three groups ( χ2=10.13, P=0.006). Conclusions:In the course of clinical reasoning in general practice, BTBL can improve teaching efficiency, save teaching resources, and play the role of promoting strengths and avoiding weaknesses.