A framework to assess welfare mix and service provision models in health care and social welfare: case studies of two prominent Italian regions

BACKGROUND The mechanisms through which the relationships among public institutions, private providers and families affect care and service provision systems are puzzling. How can we understand the mechanisms in these contexts? Which elements should we explore to capture the complexity of care provision? The aim of our study is to provide a framework that can help read and reframe these puzzling care provision mechanisms in a welfare mix context. METHODS First, we develop a theoretical framework for understanding how service provision occurs in care systems that are characterised by a variety of relationships between multiple actors, using an evidence-based approach that looks at both public and private expenditures and the number of users relative to the level of needs coverage and compared with declared values and political rhetoric. Second, we test this framework in two case studies built on data from two prominent Italian regions, Lombardy and Emilia-Romagna. We argue that service provision models depend on the interplay among six conceptual elements: policy values, governance rules, resources, nature of the providers, service standards and eligibility criteria. RESULTS Our empirical study shows that beneath the relevant differences in values and political rhetoric between the case studies of the two Italian regions, there is a surprising isomorphism in service standards and the levels of covering the population's needs. CONCLUSION The suggested framework appears to be effective and feasible; it fosters interdisciplinary approaches and supports policy-making discussions. This study may contribute to deepening knowledge about public care service provision and institutional arrangements, which can be used to promote more effective reforms and may advance future research. Although the framework was tested on the Italian welfare system, it can be used to assess many different systems.