Análise da evolução das taxas de mortalidade de crianças com até um ano de idade residentes no estado do Rio Grande do Sul

Abstract

Infant mortality rates represent indicators of quality of life and measure human development, social inequality, and the living and health conditions of a given population. This study aims to highlight their main characteristics and compare the evolution of infant mortality rates in the state of Rio Grande do Sul (RS) through the ARIMA and 𝛽ARMA methodologies. The infant mortality rates were obtained from the Unified Health System (SUS) Department of Informatics (DATASUS), through the deaths registered in the Mortality Information System (SIM), stratifying them by state. A descriptive analysis was carried out and the vulnerability characteristics of babies, mothers, and childbirth were demonstrated. In addition, the time series was modeled using the Box-Jenkins or ARIMA and 𝛽ARMA methodologies, considering monthly observations in the period of 2000 to 2017. Cacique Doble was the city of residence with the highest infant mortality rate for the state of RS registered in this period. Using the residual analysis and the AIC and BIC penalizing criteria to evaluate the quality of the models, a better performance was observed in the adjustment of the 𝛽ARMA model, while the best model for the series was 𝛽𝐴𝑅𝑀𝐴(4,6) in the 6-month time horizon. However, the accuracy measures MAE, MAPE and RMSE were lower in the SARIMA model. The proposed methodologies can guide the planning of preventive and educational policies aimed at the risk of a born alive dying during its first year of life.