Dados do Trabalho

Título

BAYESIAN MODEL FOR MAPPING IN THE PRESENCE OF UNDERREPORTED DATA: THE CASE OF NEONATAL DEATHS IN MINAS GERAIS

Resumo

In poor and more socially deprived areas, economic and social data are typically underreported. As a consequence, quantities of interest for, e.g., political, social and scientific purposes, such as income, rates of death and spread of diseases, tend to be underestimated. The great challenge, in those cases, is to build models able to provide reliable estimates for such quantities, despite the poor quality of data. In this paper we introduce a Bayesian model for mapping data that are subjected to underreporting. The usual practice to overcome the problem is to assume that data are censored. The proposed model considers count data in different areas, modeled using Poisson distributions, whereas prior information is used to build an appropriate distribution for the probability of underreport in each area. To illustrate the use of the proposed model, we map the early neonatal hospital mortality in the Minas Gerais state, a Brazilian state which presents heterogeneous characteristics and a relevant socio-economical inequality. Levels of functional illiteracy are then considered as covariates affecting underreporting. Finally, microregions are clustered according to their mortality rates and findings are compared with other studies.