Objective To analyze the epidemiological characteristics and spatiotemporal clustering of hepatitis A in Yunnan province from 2008 to 2023 and provide a theoretical basis for the prevention and control of hepatitis A.

Methods The distributions of hepatitis A cases in Yunnan were described. Linear regression trend test was used for incidence rate and age specific incidence. Global spatial autocorrelation analysis and spatiotemporal scan analysis was conducted by using software R, and local spatial autocorrelation analysis was conducted by using ArcGIS.

Results From 2008 to 2023, a total of 27 176 hepatitis A cases were reported in Yunnan with an annual average of 1 699 cases. The annual average incidence rate was 3.62/100 000, showing a downward trend (coefficient: b= −0.45, P<0.001). The median age of the cases was 28 years, showing an upward trend (multiple R-squared: R²=0.89, P<0.01). From 2008 to 2023, the reported incidence rates of hepatitis A in age groups 0-9, 10-19, 20-29, and 30-39 years all showed downward trends (P<0.01). The results of global spatial autocorrelation analysis showed that there was global spatial correlation from 2008 to 2023 except 2011 (P<0.05). The results of local spatial autocorrelation and spatiotemporal scan analysis showed that the areas with high incidence of hepatitis A were mainly in southeastern and southern Yunnan, i.e. a belt from Qiubei to Menglian in the southwest (a high-high clustering area and a class I clustering area covering 20 districts and counties). Since 2017, most counties (districts) in Pu'er had been the high-risk areas. Since 2020, some counties (districts) in Qujing had become high-risk areas for consecutive years. In addition, as a class II clustering area, Ninglang had 535 cases being actually observed and 135 cases being predicted. It is the area with the highest clustering relative risk, indicating that this county had a relatively high risk for case clustering.

Conclusion From 2008 to 2023, the incidence rate of hepatitis A in Yunnan showed a downward trend, and the age specific incidence peak occurred later. The high-incidence population groups had changed from students and children living scatteredly to farmers, retirees, and jobless or the unemployed. In terms of spatial clustering, attention should be paid to the areas with high-risk for clustering from Qiubei to Menglian in the southwest, some counties and districts that had become high-incidence areas again in Pu'er, and some counties and districts that might become high-incidence areas in Qujing. It is necessary to explore the reasons for the abnormal increase in clustering relative risk in Ninglang in northwestern Yunnan to provide a basis for taking targeted prevention and control measures.