Protocol: Systematic mapping of recent global research on health system financing using machine learning

Health systems worldwide are facing critical pressures that threaten access to and quality of healthcare, with profound long-term social and economic consequences. In high-income countries, population ageing and declining health sector budgets are already leading to deteriorating healthcare, while low- and middle-income countries must meet the needs of rapidly growing populations despite significant declines in development assistance. In this context, governments and private actors urgently need to identify more effective and efficient approaches to financing healthcare. This systematic evidence map aims to comprehensively document the global empirical literature on health systems financing, published since 2010, examining how resources are raised, pooled, allocated and expended, as well as on more upstream issues linked to capital investment, procurement, and supply chains. Building on Berrang-Ford et al. (2021), we will use machine-learning-assisted screening, prioritization, and data extraction. The resulting systematic evidence map will outline the contours of the current health financing evidence base, providing researchers, funders, and decision-makers with insight into which policy-relevant areas are well studied and which remain underexplored. The primary purpose is to characterize the recent current evidence base to inform research funding allocation decisions. The outputs will include a dynamic filterable evidence map, data visualizations, and an interactive AI-powered chatbot.