Using outcome measures in sub-national level performance management: when and under what circumstances?

There is growing interest in using outcome-based measures in sub-national level health system performance management, particularly in high-income countries. Increasingly, population health indicators used for making international comparisons are being applied at a sub-national level. This study aims to understand whether and under what circumstances population health outcome-based measures can be used for performance measurement and management at the sub-national level health systems. We have integrated empirical population-based data with key health system expert perspectives to evaluate the appropriateness of two population health indicators - amenable mortality and ambulatory-sensitive hospitalization of young children. Our assessment focused on two key aspects: (i) the technical validity of these indicators, ensuring they accurately measure these outcomes, and (ii) the functionality and legitimacy of performance information, determining whether it meets stakeholders' program or policy needs and supports strategic decision-making. Overall, we found that the 'intermediate' outcome measure, childhood ambulatory sensitive hospitalization, was more useful for identifying district-level health system performance variation than the 'end' outcome measure, amenable mortality. Performance information based on childhood ambulatory-sensitive hospitalization is more appropriate for improving decision-making, and it is more likely to be accepted by a wide range of stakeholders involved in health system performance improvement.