Development of a research mentorship guide and consensus statement for low- and middle-income countries: Results of a modified Delphi process

Background: Institutional research mentorship is a form of mentorship whereby institutions foster mentor-mentee relationships. Research mentorship improves research effectiveness and supports relationships. However, resources are needed in order to institutionalize research mentorship tailored to low- and middle- income countries (LMICs). The aim of this study was to develop a consensus document on institutionalizing research mentorship through a modified Delphi process as part of the practical guide development process. Methods: This study used a two-round modified Delphi process, which is an iterative, structured approach of consensus decision making. Each participant was asked about a series of items related to research mentorship using Likert scale questions. Agreement for each item was pre-defined as ≥80% of participants rating the item as “agree” or “strongly agree.” The items that reached agreement, were then discussed during round two at an in-person conference in Ethiopia. A separate group of individuals only participated virtually. For the final consensus survey, response rates and commenting rates (participants who wrote two or more comments) were compared among conference and non-conference participants. Results: The Delphi process led to the inception of three main themes in terms of developing research mentorship: leveraging existing resources, measuring and evaluating institutional mentorship, and encouraging a research mentorship life cycle. During the virtual first round, 59% (36/61) participants who were emailed completed the survey. In the second round, conference participants had a response rate of 79% (11/14) compared to non-conference participants with a response rate of 45% (21/47). Conference participants had a 100% (11/11) commenting rate whereas non-conference participants had a 38% (8/21) commenting rate. This study achieved consensus in both survey rounds for all 35 items on the consensus document. Conclusions: The data suggest that an in-person conference may increase participant engagement. The consensus developed through a modified Delphi method directly informed a practical guide on institutionalizing research mentorship in LMICs.