以临床推理为导向的数字化赋能病理学教改评价

Evaluation of Pathology Education Reform Empowered by Digitalization with a Clinical Reasoning Orientation

摘要: 目的：评价新医科背景下面向临床推理能力培养的病理学数字化赋能教学改革效果。方法：采用基于届别的历史对照准实验设计。以2022级临床医学本科生为对照组(传统讲授，N = 352)，以2023级为改革组(数字化赋能教学，N = 459)。两组同一课程与考核体系下参加结业考试(满分100分)，授课教师、学时与考核流程一致，命题教师为同一人；比较卷面成绩、基础题与临床推理题得分率及试卷认知层次分布。对改革组开展Likert 5级满意度问卷(发放459份，回收率约79%，n ≈ 363)。结果：改革组平均分62.48分，高于对照组60.22分(均值差2.26，95% CI 0.28~4.24，p ≈ 0.026)。基础题与临床推理题得分率分别由47.65%升至66.80%、由54.35%升至70.50%。认知层次中“解释”占比由29.79%升至47.83%，回忆略降(40.43%→39.13%)，问题解决下降(29.79%→13.04%)。非常满意比例：AI智能体84.8%、AI数字人78.7%、数字病理标注92.0%。结论：数字化赋能的推理导向病理学教改与成绩提升及推理相关题型表现改善相关，学生认可度高；但仍需稳定命题蓝图、进行难度等值或采用独立的临床推理测评工具进一步验证。

Abstract: Objective: To evaluate the reform effects of digital empowerment teaching in pathology aimed at cultivating clinical reasoning skills under the background of new medical sciences. Methods: A quasi-experimental design with a matched control was adopted. The control group consisted of 2022 clinical medicine undergraduates (traditional lecture, N = 352), while the reform group comprised 2023 students (digital empowerment teaching, N = 459). Both groups participated in the final examination (maximum score 100) under the same course and assessment system, with identical instructors, teaching hours, and assessment procedures, and the same question setter. The written test scores, scoring rates for basic questions and clinical reasoning questions, and cognitive level distribution in the test papers were compared. A Likert 5-point satisfaction questionnaire was administered to the reform group (459 copies distributed, recovery rate approximately 79%, n ≈ 363). Results: The average score of the reform group was 62.48, higher than that of the control group (60.22) (mean difference 2.26, 95% CI 0.28~4.24, p ≈ 0.026). The scoring rates for basic questions and clinical reasoning questions increased from 47.65% to 66.80% and 54.35% to 70.50%, respectively. The proportion of “interpretation” in cognitive levels rose from 29.79% to 47.83%, while recall slightly decreased (40.43%→39.13%) and problem-solving declined (29.79%→13.04%). The proportions of very satisfied respondents were: AI agent 84.8%, AI digital human 78.7%, and digital pathology annotation 92.0%. Conclusion: The digitally empowered reasoning-oriented pathology teaching reform is associated with improved academic performance and enhanced performance in reasoning-related question types, with high student recognition. However, further validation is required by stabilizing the test blueprint, achieving equivalent difficulty levels, or adopting independent clinical reasoning assessment tools.