Implementing a Near-Peer Mentorship Program to Enhance Surgical Research Among Medical Students: A Protocol for a Mixed-Methods Evaluation.

Background: Research is vital in medical education, fostering evidence-based practice and professional growth. However, medical students in low- and middle-income countries, such as Uganda, face significant barriers to research participation, including limited mentorship and resources. At Makerere University, these challenges impede surgical research output among students. Near-peer mentorship programs (NPMPs) offer a promising solution by leveraging senior students and residents to guide peers, enhancing research skills and productivity. This study implements and evaluates an NPMP within the Makerere University Surgery Society (MUSS) to boost student-led surgical research.

Methods: A mixed-methods implementation research design, guided by the RE-AIM Framework, will assess the NPMP over 12 months at Makerere University College of Health Sciences. The program involves 120 participants: 15 surgical residents mentoring 30 senior students (Years 3-5), who mentor 75 junior students (Years 1-5). Convenience and purposive sampling will recruit participants. Quantitative data (surveys, logs) will measure reach, effectiveness (research output, competency), adoption, implementation, and maintenance. Qualitative interviews with 25 participants will explore feasibility, acceptability, and sustainability. Data analysis includes descriptive/inferential statistics and thematic analysis. This protocol has been approved by the Mulago Hospital Research and Ethics Committee (Protocol Approval Number MHREC 2931).

Discussion: The NPMP aims to increase surgical research output, improve skills, and establish a sustainable mentorship culture within MUSS. By addressing mentorship gaps, it could enhance student engagement and offer a scalable model for resource-limited settings. Limitations include the lack of a control group and potential selection bias. This implementation research aims to establish a scalable model rather than prove direct causality in this initial phase.