Key result
The SARIMA model demonstrated superior predictive accuracy for hand, foot, and mouth disease incidence compared to the Prophet model, with a lower Mean Absolute Error (8.30 vs 274.03).
Why the study?
It is crucial to understand the epidemiological characteristics of hand, foot, and mouth disease (HFMD) in Quzhou and establish a prediction model to lay the foundation for early warning.
Observational (n=66,601)
The SARIMA model provides better predictive accuracy than the Prophet model for forecasting hand, foot, and mouth disease incidence, which can aid in public health resource allocation and prevention strategies.
Background HFMD is a common infectious disease that is prevalent worldwide. In many provinces in China, there have been outbreaks and epidemics of whooping cough, posing a threat to public health. Purpose It is crucial to grasp the epidemiological characteristics of HFMD in Quzhou and establish a prediction model for HFMD to lay the foundation for early warning of HFMD. Method Descriptive epidemiological methods were used to analyze the epidemic characteristics of HFMD, the incidence map was drawn by ArcGIS software, the Seasonal Auto Regressive Integrated Moving Average (SARIMA) and Prophet model were established by R software. Then, root mean square error (RMSE) and mean absolute error (MAE) were used to evaluate the fitting and prediction performances of the model. Results From 2010 to 2023, Quzhou City reported a total of 66,601 cases of hand, foot, and mouth disease (HFMD), with the annual number of reported cases fluctuating between 2,265 and 7,964. The average annual incidence rate was 216.88/100,000, with the lowest in 2010 (97.08 /100,000) and the highest in 2016 (373.37/100,000) ( P trend < 0.001). The cases exhibited a seasonal bimodal distribution, with the first peak occurring from April to July and the second peak from October to December. The incidence rate of HFMD in males (246.71/100,000) was higher than that in females (185.81 /100,000). The performance of the SARIMA (1,0,1)(2,1,0) 12 model was better than that of the Prophet model in terms of prediction accuracy. Conclusion The incidence of hand, foot and mouth disease in Quzhou is on the rise in 2010–2016 and 2022–2023. In this study, the SARIMA prediction model was compared with the FB Prophet model. Data for more years can then be observed to better predict trends in the incidence of HFMD, providing a basis for prevention strategies and resource allocation. Further research can optimize the model to enhance predictive ability to improve the understanding and management of Chinese rival foot and mouth disease.
No takes yet. Share an insight, caveat, or question.
Xu et al. (2024) conducted an observational in Hand, foot and mouth disease (HFMD) (n=66,601). SARIMA model vs. Prophet model was evaluated on Mean Absolute Error (MAE) for incidence prediction. The SARIMA model demonstrated superior predictive accuracy for hand, foot, and mouth disease incidence compared to the Prophet model, with a lower Mean Absolute Error (8.30 vs 274.03).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: