Morbi-mortalidade nos municípios brasileiros: associação com indicadores sócio-econômicos e de assistência à saúde. Um estudo multinível

• Autor(a) principal: Cavalini, Luciana Tricai
• Data de Publicação: 2005
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações da UERJ
• Texto Completo: http://www.bdtd.uerj.br/handle/1/19420
• topic: Morbidade; Mortalidade; Assistência à saúde; Sistemas de informação em saúde; Modelos multinível; Morbidity; Mortality; Health care; Information systems in health; Multilevel models; CIENCIAS DA SAUDE::SAUDE COLETIVA

Socioeconomic inequity in Brazil has reflections on health sector. The sector's services and complexity distribution is analogous to the general social goods, flowing the country economic development vectos. The size of social disparities observedf makes possible the hypothesis statement of association among socioeconomic and health care indicators and community health outcomes. The analysis of this process requires observation of basic principles of Unique Health System related to territory, as municipality, hierarchy and region-centered policies. The objective of this study was to analyse the association among socioeconomic and health care indicators and some morbidity and mortality indicators. An ecological multilevel study was performed, taking as observation units all Brazilian municipalities for the year 2001. Infectious diseases, neoplasm and circulatori disease mortality rates, and infectious bowel diseases and diabetes mellitus hospitalization rates were obtained. The source of data for these indicators was the Mortality Information System (HIS). In order to obtain more precise rates, and verify the real size of diferences within municipalities, the MIS under-registration was corrected and the deaths and hospital admissions with undefined causes were redistributed, using the James-Stein empirical Bayes estimators for event counts in small areas. The MIS under-registration was corrected in 5,85%, more concentrated in Northern and Northeastern Regions, for neonatal and undefined causes of death, and for extremes of age. The undefined causes of death redistribution has altered the proportional mortality for Northern and Northeastern Regions, while the same procedure for HIS caused no significant changes in hospital proportional morbidity. After morbidity and mortality indicators were corrected by this method, Poisson multilevel regression models were specified for three levels: municipality, mesorregion and State. Results showed, in general terms: (a) variance was identified for second and third levels; (b) socioeconomic and health care variables were associated to all outcomes; (c) predominance of associations for the municipality level; (d) general and high complexity health care infrastructure were directly associated to indicators related to late stage of demographic and epidemiologic transition process; (e) suggestion of adjustment in Health Family Program implementation for its target population, with no impact over the respective epidemiologic profile; (f) suggestion of effects relatyed to politics definition at the municipality policy level over the morbidity and mortality. The implications for observation of these phenomena are discussed under the light of foundation remarks of Unique Health System and its policy-definition present needs.