Enhancing public health surveillance: SARIMAX model incorporating Baidu search index for HCV prediction in China

• DOI: 10.1186/s12874-025-02562-w
• Publication Date: 2025-04-23T09:58:45Z
• Authors: Huan Ning; Yingji Lan; Qianqian Huang; Tingyan Luo; Jing Yang; Jie Zhou; Zhuoxin Li; Ping Cui; Hao Liang; Jiegang Huang
• Keywords: Public health surveillance

Hepatitis C virus (HCV) has become a serious global public health issue. Existing study confirmed the potential role of internet search data in prediction infectious diseases. This aims to explore predictive models suitable for HCV, providing references and recommendations actively responding World Health Organization's strategy eliminate HCV by 2030. We collected publicly available daily reported case numbers volume HCV-related keywords Baidu Search Index (BSI) from January 2011 September 2023. Identify highly correlated with using Spearman rank correlation time series cross-correlation analysis construct comprehensive index. Finally, seasonal autoregressive integrated moving average (SARIMA) exogenous variables (SARIMAX) were developed based on monthly index (CSI). The performance these was evaluated mean absolute error (MAE), root square (RMSE), percentage (MAPE). From 2023, China cumulative total 2,949,160 cases, number 19,276. incidence is highest March each year, slight decrease February, exhibiting cycle pattern. Influenced COVID-19 pandemic, showed notable 2020 2022, but overall, it been an upward trend. After analysis, five included, lag 0 months keywords. selected SARIMA(0,1,1)(2,1,1)(12) + CSI(Lag = 0) model produced lower errors both fitting forecasting than model: MAE (706.7052 versus 759.1066 1754.05 3940.86), RMSE (973.3811 1123.343 2733.02 4846.49), MAPE (3.72% 4.08% 0.12% 0.24%). findings suggest that exhibits superior performance, maintaining accuracy even when faced emergency COVID-19. proposed SARIMAX will provide more robust support achieving 2030 dynamic zero strategy.