Learning needs assessment for multi-stakeholder implementation science training in LMIC settings: findings and recommendations

Background Despite significant progress in the field of implementation science (IS), current training programs are inadequate to meet the global need, especially in low-and middle-income countries (LMICs). Even when training opportunities exist, there is a “knowledge-practice gap,” where implementation research findings are not useful to practitioners in a field designed to bridge that gap. This is a critical challenge in LMICs where complex public health issues must be addressed. This paper describes results from a formal assessment of learning needs, priority topics, and delivery methods for LMIC stakeholders. Methods We first reviewed a sample of articles published recently in Implementation Science to identify IS stakeholders and assigned labels and definitions for groups with similar roles. We then employed a multi-step sampling approach and a random sampling strategy to recruit participants (n = 39) for a semi-structured interview that lasted 30–60 min. Stakeholders with inputs critical to developing training curricula were prioritized and selected for interviews. We created memos from audio-recorded interviews and used a deductively created codebook to conduct thematic analysis. We calculated kappa coefficients for each memo and used validation techniques to establish rigor including incorporating feedback from reviewers and member checking. Results Participants included program managers, researchers, and physicians working in over 20 countries, primarily LMICs. The majority had over 10 years of implementation experience but fewer than 5 years of IS experience. Three main themes emerged from the data, pertaining to past experience with IS, future IS training needs, and contextual issues. Most respondents (even with formal training) described their IS knowledge as basic or minimal. Preferences for future training were heterogeneous, but findings suggest that curricula must encompass a broader set of competencies than just IS, include mentorship/apprenticeship, and center the LMIC context. Conclusion While this work is the first systematic assessment of IS learning needs among LMIC stakeholders, findings reflect existing research in that current training opportunities may not meet the demand, trainings are too narrowly focused to meet the heterogeneous needs of stakeholders, and there is a need for a broader set of competencies that moves beyond only IS. Our research also demonstrates the timely and unique needs of developing appropriately scoped, accessible training and mentorship support within LMIC settings. Therefore, we propose the novel approach of intelligent swarming as a solution to help build IS capacity in LMICs through the lens of sustainability and equity.