Abstract
INTRODUCTION: Dysentery remains a significant notifiable Class B infectious disease in China, exhibiting distinct spatial variations in incidence patterns. This persistent geographical heterogeneity necessitates a systematic investigation into the underlying influencing factors to inform targeted prevention and control strategies. METHODS: Our analytical approach incorporated Moran's I index for spatial autocorrelation analysis, multiple linear regression (MLR) for preliminary assessment, and advanced spatial regression models including spatial error model (SEM), geographically weighted regression (GWR), and multiscale geographically weighted regression (MGWR). The analysis incorporated socioeconomic, educational, healthcare, and demographic factors within a unified spatial framework. RESULTS: The analysis revealed three key findings: (1) Significant spatial clustering of dysentery incidence with identified high-risk concentration in the Beijing-Tianjin region; (2) Superior performance of MGWR modeling in capturing spatial heterogeneity compared to conventional methods; (3) Distinct regional variations in dominant factors, with economic development most influential in western China, education factors predominant in northeastern areas, and healthcare resource availability showing strongest impact in the northeast but minimal effect in southern regions. CONCLUSIONS: The study demonstrates the value of multiscale spatial analysis in understanding geographical disease patterns, revealing that dysentery incidence in China is governed by different factors across regions.