Estatística Espacial aplicada na caracterização de áreas de risco para hipertensão e diabetes no estado de Minas Gerais
Data
2013-12-05
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Federal de Viçosa
Resumo
O objetivo deste estudo foi avaliar a distribuição espacial da taxa de prevalência de hipertensão arterial e diabetes mellitus do estado de Minas Gerais, entre 2002 a 2012. Para o cálculo das taxas de prevalência, foram considerados os dados obtidos por meio dos registros do cadastro familiar do SIAB-MG do DATASUS, considerando homens e mulheres hipertensos ou diabéticos, acima de 15 anos, bem como a população total por município, maior de 15 anos, conforme censo 2010 do IBGE. Foram considerados os 853 municípios de Minas Geais como unidades de análise espacial e utilizou-se as técnicas de análise espacial de dados de área, destacando-se a média móvel local, os coeficientes de autocorrelação global de Moran e o índice local de Moran (LISA). Além disso, foram construídos mapas temáticos de distribuição espacial, de autocorrelação local (Box Map) e de identificação de cluster, utilizando a estatística Scan espaço- temporal. Verificou-se que há autocorrelação espacial para a taxa de prevalência de hipertensão e de diabetes, entre os municípios, em que apresentou coeficientes de Moran (global) positivos e significativos para todos os anos estudados, indicando similaridade entre os municípios. Foi possível identificar, por meio da análise local, agrupamento de municípios situados na parte sul do estado com maiores taxas de prevalência de hipertensão e também de diabetes em todos os anos estudados. Também constatou-se cluster que persistiram no espaço e no tempo para ambas as variáveis e que se localizaram no sul do estado. Conclui-se que a análise espacial permitiu mapear e compreender a distribuição das taxas de hipertensão e diabetes no estado de Minas Gerais. Estudos como este fornecem informações para que os serviços de saúde local possam selecionar os principais locais com taxas altas, na tentativa de propor ações de controle, bem como monitorar as regiões que apresentam taxas baixas.
The objective of this study was to evaluate the spatial distribution of the prevalence of hypertension and diabetes mellitus in the state of Minas Gerais, between 2002-2012. For the calculation of prevalence rates were considered the data obtained from the registry cords of the family record of the SIAB - MG DATASUS considering men and women with hypertension or diabetes, aged over 15, as well as the total population by municipality, most of 15 years, according to the IBGE 2010 census. We considered the 853 municipalities of Minas Gerais as units of spatial analysis and used the techniques of spatial data analysis area, highlighting the moving average site, the autocorrelation coefficients of global and local content Moran (LISA). Moreover, were built thematic maps of spatial distribution of local autocorrelation (Box Map) and identification of the cluster, using the space-time scan statistic. It was found that spatial autocorrelation for the prevalence of hypertension and diabetes, among the municipalities, which showed coefficients Moran (global) positive and significant for all years studied, indicating similarity between the municipalities. It was possible to identify, by analyzing local grouping of municipalities in the southern part of the state with the highest prevalence rates of hypertension and diabetes also in all years studied. Also it was found that persisted cluster in space and time for both variables and is located in the southern state. We conclude that the spatial analysis to map and understand the distribution of rates of hypertension and diabetes in the state of Minas Gerais. Studies like this provide information to the local health services can select the main places with high rates in an attempt to propose actions to control and monitor the regions with low rates.
The objective of this study was to evaluate the spatial distribution of the prevalence of hypertension and diabetes mellitus in the state of Minas Gerais, between 2002-2012. For the calculation of prevalence rates were considered the data obtained from the registry cords of the family record of the SIAB - MG DATASUS considering men and women with hypertension or diabetes, aged over 15, as well as the total population by municipality, most of 15 years, according to the IBGE 2010 census. We considered the 853 municipalities of Minas Gerais as units of spatial analysis and used the techniques of spatial data analysis area, highlighting the moving average site, the autocorrelation coefficients of global and local content Moran (LISA). Moreover, were built thematic maps of spatial distribution of local autocorrelation (Box Map) and identification of the cluster, using the space-time scan statistic. It was found that spatial autocorrelation for the prevalence of hypertension and diabetes, among the municipalities, which showed coefficients Moran (global) positive and significant for all years studied, indicating similarity between the municipalities. It was possible to identify, by analyzing local grouping of municipalities in the southern part of the state with the highest prevalence rates of hypertension and diabetes also in all years studied. Also it was found that persisted cluster in space and time for both variables and is located in the southern state. We conclude that the spatial analysis to map and understand the distribution of rates of hypertension and diabetes in the state of Minas Gerais. Studies like this provide information to the local health services can select the main places with high rates in an attempt to propose actions to control and monitor the regions with low rates.
Descrição
Palavras-chave
Estatistica espacial, Distribuição espacial, Dados de area, Autocorrelação, Scan espaço-temporal, Saude publica, Spatial statistics, Spatial distribution, Data area, Autocorrelation, Scan-space, Public health
Citação
PINTO, Eliangela Saraiva Oliveira. Spatial Statistics applied in the characterization of risk areas for hypertension and diabetes in the state of Minas Gerais. 2013. 74 f. Dissertação (Mestrado em Estatística Aplicada e Biometria) - Universidade Federal de Viçosa, Viçosa, 2013.