Utilizing ARIMA Model Forecasts to Inform Allocation of Resources to the Maternal and Neonatal Health Program in Guatemala

Neonatal mortality continues to be an important public health issue in Guatemala especially in the rural areas where deliveries are sometimes conducted by lay midwives who lack skills of conducting safe deliveries. Failure to identify high risk pregnancies and late referrals significantly contributes to mortality among newborn babies. This study uses annual time series data on neonatal mortality rate (NMR) for Guatemala from 1964 to 2019 to predict future trends of NMR over the period 2020 to 2030. Unit root tests have shown that the series under consideration is an I (2) variable. The optimal model based on AIC is the ARIMA (4,2,2) model. The findings of this study indicate that neonatal mortality will gradually decline from approximately 12 in 2020 to around 9 deaths per 1000 live births by the end of 2030. Hence, authorities are encouraged to promote institutional deliveries particularly in the rural areas to keep neonatal mortality under control.