Forecasting incidence of tuberculosis cases in Brazil based on various univariate time-series models

Abstract

Tuberculosis (TB) remains the world's deadliest infectious disease and is a serious public health problem. Control for this disease still presents several difficulties, requiring strategies for the execution of immediate combat and intervention actions. Given that changes through the decision-making process are guided by current information and future prognoses, it is critical that a country's public health managers rely on accurate predictions that can detect the evolving incidence phenomena. of TB. Thus, this study aims to analyze the accuracy of predictions of three univariate models based on time series of diagnosed TB cases in Brazil, from January 2001 to June 2018, in order to establish which model presents better performance. For the second half of 2018. From this, data were collected from the Department of Informatics of the Unified Health System (DATASUS), which were submitted to the methods of Simple Exponential Smoothing (SES), Holt-Winters Exponential Smoothing (HWES) and the Integrated Autoregressive Moving Average (ARIMA) model. In the performance analysis and model selection, six criteria based on precision errors were established: Mean Square Error (MSE), Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percent Error (MAPE) and Theil's U statistic (U1 and U2). According to the results obtained, the HWES (0.2, 0.1, 0.1) presented a high performance in relation to the error metrics, consisting of the best model compared to the other two methodologies compared here.

Predicting Tuberculosis Morbidity Rate in Indonesia Using Weighted Markov Chain Model

“…The Tuberculosis morbidity rate in Indonesia is shown in Tables 4 and 5, which are divided into four states and six states, respectively. (3) Hence by Table 4 and Table 5, the frequency matrix for 𝑚 = 4 and 𝑚 = 6 is given by Equation (10) and Equation (11), consecutively.…”

“…By making predictions or performing forecasting, which is the process of determining what will happen in the future using historical morbidity rate data, it is possible to determine the morbidity rate in the future. Several researchers have conducted research on forecasting morbidity rates, as in [6], [7], [8], [9], [10], [11], and [12]. Some of them used machine learning and the other used statistical approaches.…”

Time series forecasting for tuberculosis incidence employing neural network models

“…The novelty of this work is the use of ANN techniques such as recurrent neural networks and nonlinear autoregressive models to analyze the TB incidence time series. There are similar studies in the region, where context similarities can be applied; for example, models based on statistical models such as ARMA, ARIMA, simple exponential smoothing, Holt-Winters, and its modified exponential smoothing were studied in Brazil, reaching the best results with the implementation of the ARIMA (4,1,5) model ( Ribeiro et al., 2019 ). In addition, a complementary study compared the effectiveness of the GeneXpert technology with an ARIMA (5,0,0) model ( Berra et al., 2021 ).…”

“…In Latin America, Brazil has been mentioned as the most relevant country in terms of TB. There, studies based on TSF have been reported, in which different univariate models have been employed ( Ribeiro et al., 2019 ). In addition, ARIMA and Holt-Winters models have been used to analyze the incidence reported by the Brazilian Unified Health System ( Achcar et al., 2021 ), as well as the effectiveness of the use of GeneXpert within TSF ( Berra et al., 2021 ).…”

Towards Tuberculosis Incidence Time Series Forecasting in Colombian Regions: The Antioquia Case

“…Authors of [9] conducted a study in Brazil with a database of confirmed cases of Pulmonary TB in the period from 2011 to 2018. The authors considered three statistical models for prediction: the Simple Exponential Smoothing Model, Autoregressive Integrated Moving Average Model, and Holt-Winters Exponential Smoothing Model.…”