Health as a common welfare: the DVSS program as a case study of shared outcomes evaluation in decentralized policy and planning for universal coverage

Research Objective: The recent WHO report of the Commission on Social Determinants of Health made clear that socioeconomic disparities cause catastrophic effects on larger strata of our society on a global scale, and the gap is wider where the public organization delegates its role to the private sector. More efficient population-based policy and planning is required to counteract these trends. The conceptual body of “clinical governance” has been introduced to improve outcomes through the direct involvement of service providers. In the highly decentralized Italian national health system, Regione Umbria undertook a three-year program to strengthen health policy and planning through a systematic evaluation and cooperation between research and policy.Study Design: The program (2005-2008) was structured around the creation of a key “Document for the evaluation of health determinants and system strategies” (DVSS). The main features of the program were: 1) coordination of the Epidemiologic Observatory of Regione Umbria (OERU); 2) evaluation of outcomes by organizational level and priority areas; 3) specific “research projects” to produce document “chapters” delivered by multidisciplinary teams of health professionals; 4) supervision of field experts and Steering Committee; 5) competitive research calls granting a substantial part of the OERU annual budget for the realization of the DVSS; 6) centrality of health outcomes in the evaluation process; 7) a System for the Evaluation of Outcomes (SVE) to automate the production of indicators from very large linked databases.Population Studied: Overall population of Regione Umbria, Italy (900,000 approx) and associated health services, over a timeframe of 5 years (2001-2005).Principal findings: A core set of N=163 outcome indicators was selected on the basis of scientific soundness, relevance and feasibility. N=138 were successfully calculated over five years (system level (N=2); hospitals (N=44); districts (N=46); screening (N=56); general practice, pediatrics (N=26); mental health (N=10). Risk adjusted indicators were produced implementing the AHRQ methodology. Results were directly interpreted by health professionals writing the Document, for which Umbria performs well in terms of average outcomes, but shows unexpectedly high rates of events in specific areas, and excess variability across the region. Conclusions: The DVSS program is a potentially useful scheme to understand the complex mechanisms of decentralized health systems. The experience can be replicated and further improved either in Umbria or elsewhere. Results obtained must be interpreted with caution, due to the current scarcity of comparable data and the potential biases induced by administrative data.Implications for Policy, Delivery or Practice: The DVSS may represent a relevant case study for countries involved in the introduction of universal coverage, particularly the US, for which a revision of performance evaluation at a population level and a decentralized organization may represent valid options. On these grounds, DVSS was invited to present its experience at the International Workshop on Performance Evaluation, WHO Ministerial Conference on Health Systems, Tallinn 2008 (http://www.whoconference2008.org/agenda?day=1)