JMIR Publications

ABSTRACT

Background:

Hemorrhagic fever with renal syndrome (HFRS) is a severe natural focal infectious disease that remains a major public health concern in Shandong Province, China. Understanding the epidemiological characteristics of HFRS and forecasting its incidence are crucial to provide guidance for policy making and implement timely interventions against this disease.

Objective:

This study aims to systematically analyze the epidemiological characteristics of HFRS in Shandong Province, China, over the past 15 years and to construct an optimal time-series forecasting model, providing a theoretical basis for the scientific prevention, control, and early warning of the disease.

Methods:

HFRS case surveillance data in Shandong Province from 2010 to 2024 were collected through the China Information System for Disease Control and Prevention. Descriptive epidemiological methods were employed to analyze the basic characteristics of HFRS. Four time-series models, namely SARIMA, Prophet, Holt-Winters, and XGBoost, were constructed using R software for fitting and predictive analysis. Their predictive performances were evaluated and compared using evaluation metrics such as root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE).

Results:

From 2010 to 2024, a total of 15,184 HFRS cases were reported in Shandong Province, with 188 deaths, representing a case fatality rate of 1.24%. The annual incidence rate ranged from 0.45/100,000 to 1.84/100,000, showing an overall fluctuating downward trend. The peak incidence occurred during the autumn and winter months from October to December each year; spatially, the eastern and southeastern regions of Shandong Province were the primary high-incidence areas; demographically, the high-risk populations mainly consisted of middle-aged and elderly individuals, males, and farmers. The Holt-Winters model outperformed C the SARIMA, Prophet and XGBoost models in predicting the trend in HFRS incidence, with RMSE, MAE, and MAPE values of 7.56, 6.43, and 34.56%, respectively.

Conclusions:

The Holt-Winters model demonstrates the optimal performance in forecasting HFRS incidence trends and can serve as the preferred tool for routine early warning. It is recommended to apply this model to enhance precise monitoring and to implement targeted prevention and control measures focusing on key areas, peak periods, and high-risk populations.