Impact of an educational supervision mechanism on blended teaching outcomes in general practitioner transfer training: a comparative study

Abstract

Objectives: This study aimed to evaluate the effectiveness of a blended teaching model under an educational supervision mechanism in general practitioner (GP) transfer training. Methods: A non-randomized, historical cohort study was conducted involving 147 trainees. The intervention group (2025 cohort, n = 70) was managed under an educational supervision mechanism, in which a supervision team monitored and evaluated teaching quality throughout the training process, while the historical control group (2024 cohort, n = 77) received conventional blended training without formal supervision. Theoretical assessment scores were collected for both groups, while questionnaire satisfaction responses were gathered only from the intervention group. Analysis of covariance (ANCOVA) adjusting for pre-training scores was used to compare post-training outcomes. Results: The intervention group achieved significantly higher overall theoretical assessment scores compared with the historical control group (median [IQR]: 92 [88- 95] vs. 90 [80.5-95]; p = 0.044; rank biserial correlation [r = 0.16]). After adjusting for baseline performance via ANCOVA, the adjusted marginal mean post-training score was 90.59 (95% confidence interval [CI]: 88.58-92.60) in the intervention group and 87.23 (95% CI: 85.32-89.15) in the historical control group, with a significant difference (F = 5.706, exact p = 0.018, partial η2 = 0.038). In addition, 84.29% of trainees in the intervention group reported very satisfied (score = 5) of the model. However, only 75.71% considered the training content fully met daily clinical demands. Conclusions: Given the limitations of the single-center historical cohort design and potential temporal confounding, these findings should be interpreted with caution, although they suggest potential for broader application in medical education.