Integrating disease mapping and flexible scan statistics to identify and visualize spatial clusters.

Abstract

Accurately characterizing the spatial distribution of diseases is essential for understanding etiology and guiding targeted public health interventions. This study develops an analytical framework integrating disease mapping and spatial scan statistics to identify and validate the spatial clustering patterns of diseases. The framework follows an "exploration-validation-refinement" workflow, combining qualitative visualization with quantitative spatial analysis. Choropleth maps intuitively depict the overall distribution of diseases, providing a basis for identifying potential high-risk areas; the flexible scan statistics (FleXScan) method detects statistically significant clusters; and dot cartograms further reveal internal heterogeneity within the most likely clusters by controlling for population density. This study validated the framework using both synthetic data and real measles case data. The results indicate that the framework substantially improves the accuracy and reliability of cluster detection, facilitates the localization of local high-risk cores, and provides actionable insights for precision disease control.