JMIR Publications

ABSTRACT

Background:

The World Health Organization declared Mpox an international public health emergency. Since January 1, 2022, China has been ranked among the top 10 countries most affected by the Mpox outbreak globally. However, there is a lack of spatial epidemiological studies on Mpox, which are crucial for accurately mapping the spatial distribution and clustering of the disease.

Objective:

This study aimed to provide geographically accurate visual evidence to determine priority areas for Mpox prevention and control.

Methods:

Locally confirmed Mpox cases were collected between June and November 2023 from 31 provinces of mainland China, excluding Taiwan, Macao, and Hong Kong. Spatio-temporal epidemiological analyses, including spatial autocorrelation and regression analyses, were conducted to identify the spatio-temporal characteristics and clustering patterns of Mpox attack rate and its spatial relationship with sociodemographic and socioeconomic factors.

Results:

From June to November 2023, 1,610 locally confirmed Mpox cases were reported in 30 provinces in mainland China, resulting in an attack rate of 11.40 per 10 million people. Global spatial autocorrelation analysis showed that in July (Moran’s I = 0.0938, P = .08), August (Moran’s I = 0.1276, P = .08), and September (Moran’s I = 0.0934, P = .07), the attack rates of Mpox exhibited a clustered pattern and positive spatial autocorrelation. The Getis-Ord Gi* statistics identified “hot spots” of Mpox attack rates in Beijing, Tianjin, Shanghai, Jiangsu, and Hainan. Beijing and Tianjin were consistent “hot spots” from June to October. No “cold spots” with low Mpox attack rates were detected by the Getis-Ord Gi* statistics. Local Moran’s I statistics identified a high-high clustering of Mpox attack rates in Guangdong, Beijing, and Tianjin. Guangdong Province consistently exhibited high-high clustering from June to November, while Beijing and Tianjin were identified as high-high clusters from July to September. Low-low clusters were mainly located in Inner Mongolia, Xinjiang, Xizang, Qinghai, and Gansu. OLS regression models showed that the cumulative Mpox attack rates were significantly and positively associated with the PUP (t = 2.4041, P = .02), PCGDP (t = 2.6955, P = .01), PCDI (t = 2.8303, P = .008), PCCE (t = 2.7452, P = .01), and PCCEH (t = 2.5924, P = .01). The GWR regression models indicated a positive association and spatial heterogeneity between cumulative Mpox attack rates and the PUP, PCGDP, PCDI, and PCCE, with high R2 values in North and Northeast China.

Conclusions:

Hot spots and high-high clusterings of Mpox attack rates identified by local spatial autocorrelation analysis should be considered key areas for precision prevention and control of Mpox. Specifically, Guangdong, Beijing, and Tianjin provinces should be prioritized for Mpox prevention and control. These findings provide geographically precise and visualized evidence to assist in identifying key areas for targeted prevention and control.