Unleashing the potential of real-world data for learning health systems. Context and applications of international studies in diabetes care

Real-World Data (RWD) are data collected as part of daily life operations that form the building blocks of modern information systems to measure quality of care continuously and consistently. Learning Health Systems (LHS) have been described as systems "at the crossroads of people and information systems, which enable virtuous learning cycles informed by evidence and actionable real-world data for meeting systems-wide, clinically oriented, and patient-relevant delivery targets". The thesis explores pillars, processes and outcomes of LHS, using RWD in a set of European comparative studies in diabetes, to identify best practices that can improve health systems performance under rapidly evolving circumstances. In Part I, the policy context and health data governance are targeted as pillars of LHS, in the broader scenario of international studies carried out at the OECD and EU level. In Part II, the thesis focuses on outcomes and processes in specific applications in diabetes. The results of this research highlight the need of identifying actionable indicators to fully unleash the potential of RWD in LHS. Accurate measures must be supported by a coherent and sustainable information infrastructure, whose compliance with legal and ethical principles shall be constantly monitored. The reliability and clinical value of RWD for LHS can be substantially improved by promoting the implementation of population-based disease registries to aid clinical interpretation and enhance predictive modeling. Information systems implementing these principles can be used routinely to inform policies and generate performance improvement spinoffs through standardized international comparisons.