The effectiveness of e-learning in maternal and neonatal health education: a systematic review and meta-analysis

Background and objective E-learning is increasingly used in health professions education, including maternal and neonatal care. This systematic review evaluated the effects of e-learning interventions on knowledge and clinical skills among learners receiving maternal and neonatal health education. Methods We searched PubMed, Scopus, ProQuest, Web of Science, Cochrane Library, ERIC, MEDLINE, SID, and Magiran from inception to June 2025. Additional searches were conducted in relevant systematic-review databases, Google Scholar, and grey-literature sources. Two reviewers independently screened studies and assessed risk of bias using RoB 2.0 for randomized controlled trials and the JBI quasi-experimental checklist. Certainty of evidence was assessed using GRADE. Where sufficient quantitative data were available, meta-analyses were conducted using RevMan 5.4 and Comprehensive Meta-Analysis version 4. Results Sixteen interventional studies were included, comprising randomized controlled, quasi-experimental, and before-and-after interventions. Interventions included web-based modules, videos, simulations, serious games, mobile applications, and blended learning. In controlled studies, e-learning showed a favorable but non-significant effect on knowledge (SMD = 0.89, 95% CI − 0.22 to 1.99; p = 0.12; I² = 93%; very low-certainty evidence). For clinical skills, the pooled effect favored e-learning (SMD = 0.43, 95% CI 0.05 to 0.82; p = 0.03; I² = 38%; low-certainty evidence). In two uncontrolled before-and-after studies, knowledge improved after e-learning (SMD = 1.108, 95% CI 0.273 to 1.943; I² = 88%); very low-certainty evidence, but the absence of a control group and substantial heterogeneity limit causal interpretation. Conclusion E-learning may improve clinical skills in maternal and neonatal health education (low-certainty evidence), whereas evidence for an advantage in knowledge acquisition is less certain. Interpretation is constrained by substantial heterogeneity, incomplete outcome reporting, risk of bias, and the limited number of studies contributing to each meta-analysis.