Pre-class learning assessment on integrating artificial intelligence into evidence-based medicine curriculum

Objective To investigate the knowledge base, learning attitudes, and practical abilities of students enrolling in an evidence-based medicine (EBM) course integrated with artificial intelligence (AI).Methods The study targeted students enrolled in the 2024–2025 academic year course “Evidence Synthesis and Application: Network Meta-Analysis” at Peking University Health Science Center. Data were collected through a self-developed electronic questionnaire covering AI-related knowledge, attitudes, practices, and expectations or concerns about the course. Differences between master’s and doctoral students were analyzed using the Mann-Whitney U test.Results A total of 46 students participated in the survey (34 Masters and below, and 12 PhDs). Most students demonstrated basic knowledge of AI and understood the core technologies of large language models, though their understanding of technical details was limited. Approximately 89.13% believed AI could enhance learning efficiency, and 84.78% expressed interest in using AI for Meta-analysis, yet many exhibited low trust in AI-generated content and were concerned about its impact on independent thinking. There were no statistically significant differences in attitudes towards the course between master and below students and doctoral students (P>0.05). In practice, students primarily used AI for tasks such as translation, writing, and information retrieval, with limited application to complex EBM-related tasks. Students also expressed concerns about the accuracy of AI-generated content and privacy security.Conclusion While students possess a foundational knowledge base and exhibit a positive attitude toward AI-assisted EBM courses, their trust in AI and practical application skills require improvement. These findings provide insights for reforming course content and teaching methods.