Work

Temporal and spatial dynamics of dengue fever in mainland China from 2004 to 2020 and the impact of COVID-19 public health interventions

Year: 2025

Type: article

Abstract: BACKGROUND: Dengue fever (DF), a mosquito-borne viral disease, presents critical public health issues worldwide, especially in tropical and subtropical areas. This study analyzed temporal and spatial patterns of DF in mainland China from 2004 to 2020 and evaluated the association between COVID-19-related public health interventions and DF incidence. METHODS: The dataset was sourced from China's Public Health Science Data Center. Joinpoint regression identified changes in incidence and mortality rates. Local Indicators of Spatial Association identified high-incidence clusters. To forecast DF incidence rates, time series models were employed, including Seasonal Autoregressive Integrated Moving Average (SARIMA), Exponential Smoothing (ETS), and Bayesian Structural Time Series (BSTS). Model performance was evaluated by RMSE, MASE, and SMAPE. Counterfactual forecasts for 2020 (assuming no COVID-19 public interventions) were generated to quantify reductions in DF incidence. RESULTS: Between January 2004 and December 2020, a total of 94,725 DF cases and 13 related fatalities were documented in China. The incidence showed multi-stage trends with joinpoints in 2011 and 2014 and phase-specific APCs of -3.07 % (95 % CI -73.07 to 53.87) for 2004-2011, +301.54 % (38.41-697.77) for 2011-2014, and -25.60 % (-78.89 to 18.10) for 2014-2020. High-incidence clusters were in Guangdong, Guangxi, Fujian, Zhejiang, and Hainan. Global Moran's I was 0.0298 (P = 0.62), indicating no significant global spatial autocorrelation, but Local Moran's I identified significant high-high clusters in Fujian, Jiangxi, Guangdong, and Guangxi. Age-specific analysis showed the highest incidence rates in the 15-64 age group. On the national test set, ETS performed best (RMSE = 0.81, MASE = 0.82, SMAPE = 0.71), while province-specific optimal models varied. Observed 2020 incidence was 96.5 % lower than counterfactual predictions overall, with Guangdong and Guangxi showing reductions of 99.5 % and 49.3 %, respectively. CONCLUSIONS: DF incidence in mainland China exhibited significant temporal and spatial variations from 2004 to 2020, with high-risk areas mainly in the south. The multi-model forecasting framework yields robust, interpretable province-level predictions and reveals a substantial 2020 decline associated with COVID-19-related public interventions.

Source: Travel Medicine and Infectious Disease

DOI: 10.1016/j.tmaid.2025.102938

Authors: Zhiqun Lei, Zhiqun Lei, Zhixin Lei, Zhixin Lei, Qi Wang

Institutions: Wuhan University of Technology, Gannan Medical University, Huazhong University of Science and Technology

Language: English

Topic: Mosquito-borne diseases and control

Subfield: Public Health, Environmental and Occupational Health

Field: Medicine

Domain: Health Sciences

SDG: Good health and well-being

Open Access status: gold

Estimated APC paid: $3,710