Abstract
BACKGROUND: Serious games have emerged as an innovative educational approach, holding potential to improve the quality and efficacy of cardiopulmonary resuscitation (CPR) training. The objective of this meta-analysis was to systematically evaluate the impact of serious games on CPR training and educational outcomes. METHODS: We systematically searched eight databases to identify randomized controlled trials (RCTs) that compared the effectiveness of serious games with that of traditional training in CPR education. The literature search was conducted up to September 20, 2025. All statistical analyses were performed using RevMan 5.4 software. RESULTS: A total of 11 RCTs involving 409 participants in the serious game groups and 388 participants in the control groups were included. We found no statistically significant differences between groups in CPR compression depth (SMD = 0.63, 95% CI: -0.09 to 1.35, p = 0.08), compression rate (SMD = -0.04, 95% CI: -0.53 to 0.44, p = 0.86), theoretical knowledge scores (SMD = -0.22, 95% CI: -0.96 to 0.51, p = 0.55), or practical skill scores (SMD = -0.18, 95% CI: -1.02 to 0.66, p = 0.67). Egger regression analysis indicated significant publication bias in theoretical scores (p = 0.015) and skill scores (p = 0.022) in this meta-analysis. The high heterogeneity observed across studies necessitates cautious interpretation of these pooled results. CONCLUSION: Current evidence does not show a statistically significant advantage for serious games over traditional CPR training methods. Importantly, the evidence is also insufficient to confirm their true equivalence. A cautious stance regarding their widespread adoption is therefore warranted. Future research should prioritize high-quality, adequately powered studies with rigorous designs, standardized intervention protocols (e.g., game types, training duration), and unified internationally recognized assessment standards.