Objective: To describe the use of learning materials and to analyze their association with academic performance among undergraduate and postgraduate students at Hanoi Medical University in 2025.
Methods: A cross-sectional study was conducted on 500 undergraduate and postgraduate students at Hanoi Medical University, with data collected from June to September 2025 using a self-administered questionnaire. Academic performance was dichotomized (good-excellent vs. average-weak). Missing data were handled using multiple imputation by chained equations (MICE). Associations were examined using univariable and multivariable logistic regression, reporting odds ratios (ORs) with 95% confidence intervals (95% CI).
Results: The proportion of frequent/heavy use of artificial intelligence (AI) tools was 73.4%, and frequent/heavy use of traditional textbooks was 76.6%. In the multivariable model, frequent AI use showed a trend toward higher academic performance that did not reach statistical significance (aOR = 1.54; 95% CI: 0.52-4.51; p = 0.435). Tuition fee per semester was significantly associated with the outcome (p = 0.012). Subgroup analysis showed a near-significant trend among postgraduate students (OR = 2.86; 95% CI: 0.87-9.43; p = 0.083) but not among undergraduates (OR = 1.42; 95% CI: 0.75-2.67; p = 0.282).
Conclusion: Traditional materials and AI tools are used concurrently at high levels. There is insufficient evidence to conclude that AI use is independently associated with academic performance across the whole sample; the trend toward benefit among postgraduate students warrants confirmation in larger studies.