Índice de desenvolvimento humano e tuberculose: modelagem e mapeamento da incidência dos casos na Paraíba no período de 2001-2011

Abstract

The main objective of this study is to map the incidence of TB in the state of Paraiba and in their municipalities, as well as expressing the relation of social determinants with cases in Brazil, in the period between 2001 and 2011. The study is presented as observational ecological predominantly analytical and exploratory. The study was conducted in the state of Paraiba, and specifically in each of its municipalities. Data from SINAN, IBGE and Ministry of Health of Paraíba. For the analysis of the study, we employed two software tools TabWin and R, version 32 and version 2.9.1, respectively. For statistical analyzes, we used Cointegration and Unit Root, besides the application of generalized linear models and Poisson regression. We conducted mapping the incidence and described, year by year, ranging between 2001 and 2011, were generated maps colorful and noticed a predominance of notifications between the priority municipalities, however the identification of small municipalities with high values reported. Set up a generalized linear model for Poisson family with identity link, however, the model in the final version, showed no adjustment to the database, feature limitations presented by the database. According to previous studies on this topic, the characteristics of the impact of TB, such as, suffering and economic loss caused by it, configured as an affront to the conscience of society, and its permanence or persistence of occurrence is considered a failure scientific, medical and political. The concept of statistics, can be described as a part of applied mathematics, which provides methods for collecting, organizing, description, analysis and interpretation of data as well as to their use in supporting decision making. It is observed that the relevance or representativeness of the study, according to the statistical criteria, is not just the end result, but in the process, which led to such conclusion. Finally, it is observed that both models support the decision making process used in this study were effective, however, the Poisson regression model presented limitation of not being adjusted, due to the characteristics of the database.