What influences the impact of health financing reforms? Using qualitative comparative analysis to identify patterns in health financing systems and their effects on financial protection.

Background: Policy interventions to improve financial protection do not always work as expected and their impact can vary across settings, suggesting that underlying health financing attributes (i.e. system arrangements for revenue raising, pooling and purchasing) matter. The objective of this study is to systematically identify patterns of health financing system attributes which influence the impact of policy interventions on financial protection. Methods: We conducted a systematic literature review to identify studies evaluating the impact of health financing interventions on financial protection. We searched across five databases from their earliest record to October 2023 and applied no geographic restrictions. Data were extracted on the measured outcome of catastrophic health expenditures and on underlying attributes of the health financing system. Patterns in the relationship between system attributes and financial protection were identified using qualitative comparative analysis (QCA). QCA compares and contrasts different combinations and assesses through formal statistical tests the degree to which combinations are present (or absent) when the intervention has been successful (or not) in obtaining the outcome of interest. Results from QCA tests were illustrated by selected studies. Findings: We identified 128 studies in our literature review from which a rich dataset was formed, representing diverse health financing systems and country reform experiences. The QCA approach to synthesize the literature provided unique insights regarding the pathway from policy intervention to impact on financial protection. We found empirical support for the significance of system attributes influencing intervention impact on financial protection. Findings indicated that combinations of more than one attribute were needed for positive impact on catastrophic health expenditures whereas only an individual attribute was needed to avoid negative impact. Attributes can have a direct or indirect effect on impact and can be synergistic, opposing or mitigating in relation to improving financial protection. Conclusion: This is the first application of QCA to health financing data. The pathway between impact on financial protection and a health financing intervention is complex and mediated by combinations of explanatory factors. Further research is needed for more comprehensive analyses across all health financing functions. Taking into account the underlying attributes of the health financing system and their interactions is critical to ensuring effective policies and progress on financial protection.