Análise de dados de vigilância epidemiológica com auxílio de gráficos de controle estatístico

Abstract

The statistical control charts play an important role in processes of health surveillance. However, implementation of these charts has not yet become a routine amid the healthcare practitioners. This aim proposes the statistical control charts play an important role in processes of health surveillance. However, implementation of these charts has not yet become a routine amid the healthcare practitioners. This aim proposes two statistical methods for early detection of outbreaks of notifiable diseases: Graphs x and R (for individual measurements and moving range) and exponentially weighted moving average (EWMA). For this, we performed a crosssectional study using secondary data from the SINAN (Information System for Notifiable Diseases) consisting of cases of hepatitis C, reported in Rio Grande do Sul, in the period January 2007 to December 2011. The graphics performance were analyzed and discussed comparatively. For this purpose, we used the methodology of Box and Jenkins to obtain a model to captures the behavior of the series and to meet the presuppositions of the use of control charts. The model found in the work was a SARIMA (0,1,1)×(1,0,0)12. It can concluded that the EWMA chart is substantially faster than X and R graphics to detect process disturbances allowing to complement epidemiological investigations.