Blended Learning in Health Care Education: An Overview and Overarching Meta-analysis of Systematic Reviews.

INTRODUCTION: Blended learning is an educational approach that integrates face-to-face with online instruction. This overview of existing systematic reviews aims to evaluate the impact of blended learning on educational outcomes in health care professional education, identify gaps in the current evidence, and direction for future reviews. METHODS: Five databases were searched (January 1, 2000 to January 14, 2021) for systematic reviews of blended versus nonblended learning approaches for health care education. "Flipped classroom", unpublished studies, abstracts, and study protocols were excluded. The extracted data included details of included reviews, description of populations, and outcomes according to Kirkpatrick levels. A narrative review is presented, along with an overarching meta-analysis, which evaluates a synthesized estimate of the effect of blended learning based on standardized mean differences. RESULTS: Eleven systematic reviews were included, comprising of 160 primary studies (including 117 randomized controlled trials) from 56 countries with over 18,000 participants. The overall level of evidence was very low quality. An overarching meta-analysis of 93 studies addressing knowledge acquisition favored blended over nonblended learning methods (standardized mean difference 0.768 [95% confidence interval 0.594-0.941]; P < .001). None of the reviews identified an adverse effect on other educational outcomes. DISCUSSION: Blended learning may be superior to traditional teaching approaches in improving knowledge acquisition. We recommend further research to describe the relative benefits of blended learning in each individual context and identify which elements of instructional design are beneficial for each outcome. Finally, we recommend the use of clear and consistent terminology in reported studies.