The effect of narrative medicine-based education on academic performance and humanistic communication in residency training: a systematic review and meta-analysis

Our study aims to assess the effect of narrative medicine education on academic performance and humanistic communication. We performed a PRISMA guideline-based systematic review and meta-analysis. The literature was searched in PubMed, Embase, China National Knowledge Infrastructure, VIP database and Wanfang database up to January 26, 2026. Two independent reviewers performed the study selection, data extraction and quality assessment. We included 20 studies with 1,364 residents. Meta-analyses showed that compared with lecture-based learning (LBL), the application of narrative medicine was associated with significantly improved procedural skill (SMD = 0.98, 95% CI 0.42 to 1.53), humanistic care ability (SMD = 2.62, 95% CI 0.26 to 4.99), empathy ability (SMD = 1.51, 95% CI 0.76 to 2.26), and doctor-patient communication (SMD = 2.19, 95% CI 0.21 to 4.17). Our study also showed that residents were more satisfied with the narrative medicine education than LBL (OR = 6.67, 95% CI 1.74 to 25.51). In conclusion, this is the first meta-analysis to show that narrative medicine may potentially improve the procedural skills and humanistic communication compared with LBL. However, given the substantial heterogeneity and low-certainty evidence, our findings are insufficient for strong practice recommendations. Further high-quality studies are needed to validate these findings. Systematic review registration: https://www.crd.york.ac.uk/PROSPERO/view/CRD420261323721.