Abstract
Background: Dyad learning is a two-person learning dynamic in which one student observes the other performing tasks, with their roles then being reversed such that both students experience the observer and the performer role. The efficacy of dyad learning has been tested in medical education contexts, such as in medical simulation. To our knowledge, this is the first systematic review that has evaluated the efficacy of dyad learning in a medical simulation context. Methods: PubMed, Google Scholar, Cochrane Library databases were searched in September 2021 and January 2022. Prospective studies of randomized design that compared dyad learning to a single medical student or physician learning in a medical simulation were included. Non-English language studies, secondary literature papers, non-human based studies, and papers that were published prior to 2000 were excluded. The methodological quality of these studies was assessed using the Medical Education Research Study Quality Instrument (MERSQI). The Kirkpatrick model was used to conceptualize study outcomes. Results: The identified papers included eight studies from four countries that totaled 475 participants. Students reported positively on their experiences as dyads, especially regarding the social aspects of it. Studies showed non-inferior learning outcomes for dyads. As most studies were one or two days long, there is limited evidence that this non-inferiority extends to longer term training modules. There is some evidence to suggest that dyad learning outcomes may be replicable in a clinical context following simulation training. Conclusions: Dyad learning in medical simulation is a pleasant experience for students and may be as effective as conventional learning. These findings set the foundation for future studies of longer duration, which is needed to determine the efficacy of dyad learning in lengthier curriculums and long-term knowledge retention. While cost-reduction is an implied benefit, studies that explicate cost reduction are needed to formalize this.