PIN79 VACCINATION COVERAGE RATE: WHAT CAN WE EXPECT FROM BRAZIL?

Abstract

The maintenance of high vaccine coverage rates is an important strategy for infectious disease control in the context of migration and tourism. The annual rate of vaccination coverage (ARVC) has decreased in Brazil in recent years. Our aim is to analyze historical ARVC data and forecast Brazilian performance in the next five years. We applied the Autoregressive Integrated Moving Average (ARIMA) model to the Brazilian ARVC time series from 1994 to 2018. Data were collected from the National Immunization Program and evaluated in four steps: (1) Data analysis: data were analyzed to identify stationarity, seasonal variation, level, cycle and trend; (2) Model Identification: this involved the estimation of the parameters p, d and q using the Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) values; (3) Model checking: this step determined whether the specified model was valid based on multi-step-ahead forecasting; and (4) Forecasting: the model was tested on real data to forecast the next five years of ARVC. The data showed a growing in ARVC between 1994 and1999, random variations from 2000-2010 and irregular variations between 2011 and 2018. The largest reduction in ARVC occurred in 2016 (50.4%). The AIC automatically identified a best model with ARIMA (1, 1, 2) in software Gretl. The ARIMA (1, 1, 2) forecast showed that ARVC will increase on average by 1.53% per year and a total of 7.06% over the next 5 years to reach an ARVC of 77.07 by 2023, if no government action is taken during this period. The ARIMA model can be used for forecasting future ARVC and will help decision-makers to establish strategies for the Brazilian vaccination program. Health authorities should make every effort to reestablish ARVC greater than 90% to avoid the re-emergence of infectious diseases that were previously controlled or eliminated in Brazil.