Spatio-temporal dynamics, trend analysis, and incidence prediction of echinococcosis in Xinjiang, China (2005–2024)

OBJECTIVE: Given the persistent transmission and marked spatial heterogeneity of echinococcosis in endemic settings, a comprehensive spatiotemporal assessment is warranted. This study aimed to characterize the spatiotemporal epidemiology of echinococcosis in the Xinjiang Uygur Autonomous Region from 2005 to 2024, develop a time-series forecasting framework, and inform region-specific prevention and control strategies. METHODS: Spatiotemporal clustering analysis was conducted to delineate geographic heterogeneity and dynamic transmission patterns at the county level. An interrupted time series framework was employed to estimate changes in the level and slope associated with the implementation of the universal free health examination policy in September 2016. Subsequently, a seasonal autoregressive integrated moving average (SARIMA) model was fitted to characterize temporal autocorrelation structures and seasonal fluctuations and to produce short-term forecasts of incidence trends. RESULTS: From 2005 to 2024, a total of 23,813 echinococcosis cases were reported in Xinjiang, with an average annual incidence of 5.48 per 100,000 population. Incidence increased significantly from January 2005 to August 2016 (β = 0.0026 per 100,000 per month, P < 0.001), with an immediate increase observed in September 2016 (0.2265 per 100,000, P < 0.05), and then declined significantly through December 2024 (β = -0.0069 per 100,000 per month, P < 0.001). Spatial clustering was observed in the Ili and Altay regions. Incidence increased and then decreased across sex and age groups, with larger declines among males. Significant reductions were observed in the 15-44 and ≥ 45 year groups, but not in the 0-14 year group. The optimal first-stage model was SARIMA(1,1,0)(2,0,0)₁₂, and the second-stage model was SARIMA(0,1,1)(1,0,2)₁₂. The second-stage model produced a relative prediction error of - 14.06% and projected 1,040 cases in 2025. CONCLUSION: Echinococcosis incidence in Xinjiang demonstrated a temporal increase followed by a sustained decline, with pronounced spatial heterogeneity across regions. Despite inherent uncertainty, SARIMA-based projections offer evidence to strengthen early warning systems and guide region-specific control strategies.