Forecasting incidence of infectious diarrhea using random forest in Jiangsu rovince, China

Abstract

Background: Infectious diarrhea can lead to a considerable global disease burden. Thus, the accurate prediction of an infectious diarrhea epidemic is crucial for public health authorities. This study was aimed at developing an optimal random forest (RF) model, considering meteorological factors used to predict an incidence of infectious diarrhea in Jiangsu Province, China. Methods: An RF model was developed and compared with classical autoregressive integrated moving average (ARIMA)/X models. Morbidity and meteorological data from 2012 to 2016 were used to construct the models and the data from 2017 were used for testing. Results: The RF model considered atmospheric pressure, precipitation, relative humidity, and their lagged terms, as well as 1–4 week lag morbidity and time variable as the predictors. Meanwhile, a univariate model ARIMA(1,0,1)(1,0,0) 52 (AIC=−575.92, BIC=−558.14) and a multivariable model ARIMAX(1,0,1)(1,0,0) 52 with 0-1 week lag precipitation (AIC=−578.58, BIC=−578.13) were developed as benchmarks . The RF model outperformed the ARIMA/X models with a mean absolute percentage error (MAPE) of approximately 20% . The performance of the ARIMAX model was comparable to that of the ARIMA model with a MAPE reaching approximately 30%. Conclusions: The RF model fitted the dynamic nature of an infectious diarrhea epidemic well and delivered an ideal prediction accuracy . It comprehensively combined the synchronous and lagged effects of meteorological factors; it also integrated the autocorrelation and seasonality of the morbidity. The RF model can be used to predict the epidemic level and has a high potential for practical implementation.

Time series analysis of foodborne diseases during 2012–2018 in Shenzhen, China

“…As described earlier, the ARIMA models for predicting foodborne diseases constructed by different studies were different (Fang et al 2020;Li et al 2010;Tao et al 2015;Wang et al 2016), which may be related to the different prevalence patterns and trends of foodborne diseases in different regions, and the latter was related to influencing factors. For example, economic level, new foodborne pathogens, the types of food that people eat, the sources of those foods, and the possible decline in public awareness of safe food preparation practices were the factors altering foodborne disease patterns (Altekruse et al 1996).…”

“…The patterns of the plot of the autocorrelation function (ACF) and the partial autocorrelation function (PACF) were used to determine the order of autoregressive (AR) and moving average (MA) included in the ARIMA model (Grahn 1995;Juang et al 2017). The fitting of the ARIMA model involves the following three essential steps (Fang et al 2020): first, a time series graph and an augmented Dickey-Fuller test is conducted to detect whether the original time series is stationary (statistical properties such as the mean and variance are all constant over time). If not, a logarithmic transformation or/and difference is adopted to achieve stability.…”

“…An ARIMA model is a statistical model for analyzing and forecasting time series data, and the model can explain a given time series based on its own past values, that is, its own lags and the lagged forecast errors, so that the equation can be used to forecast future values (Altekruse and Swerdlow 1996). The ARIMA model has become one of the most popular and convenient models in time series analysis and been widely used as classical method for various infectious disease prediction, especially in infectious diarrhea (Fang et al 2020;Li et al 2010). In China, so far, there have been few studies using the ARIMA model to predict foodborne disease, and there are some differences in the models used by different researchers.…”

“…In China, so far, there have been few studies using the ARIMA model to predict foodborne disease, and there are some differences in the models used by different researchers. Based on the infectious diarrhea (excluding cholera, dysentery, typhoid and paratyphoid) morbidity data and meteorological data from 2012 to 2016 in Jiangsu province, a univariate ARIMA model (1,0,1) (1,0,0) 52 (AIC = − 575.92, BIC = − 558.14) and a multivariable ARIMAX model (1,0,1) (1,0,0) 52 with 0-1 week lag precipitation (AIC = − 578.58, BIC = − 578.13) were developed to predict the incidence of infectious diarrhea (Fang et al 2020). It also suggested that the introduction of meteorological factors did not significantly optimize the prediction accuracy of the ARIMA model (Fang et al 2020).…”

“…Another study also concluded that the etiological and meteorological factors had age-specific effects on the prevalence of infectious diarrhea in Jiangsu (Fang et al 2019). However, even if the ARIMA models are constructed based on foodborne diseases, there were still some differences in model parameters in different regions (Fang et al 2020;Li et al 2010;Tao et al 2015;Wang et al 2016).…”