Spatial Analysis of the Relationship between PM10 Exposure and Pediatric Atopic Dermatitis in Korea

This study aims to examine the relationship between PM10 exposure and pediatric atopic dermatitis from a spatial analytical perspective by constructing a nationwide 1 km × 1 km grid dataset. Environmental monitoring data, meteorological data, and national health insurance records were integrated to build high-resolution spatial data. PM10 concentrations were estimated using a four-dimensional inverse distance weighting method incorporating latitude, longitude, altitude, and population, while other environmental variables were interpolated using kriging techniques. Spatial autocorrelation was assessed using Moran’s I, followed by Lagrange multiplier diagnostics and a spatial lag model to account for spatial dependence and spillover effects. The results indicate significant spatial clustering in pediatric atopic dermatitis and reveal that PM10 has both significant direct effects within local grids and indirect effects on neighboring areas. These findings demonstrate that conventional non-spatial models may substantially underestimate environmental health impacts. From a knowledge management perspective, this study highlights the value of constructing advanced data infrastructures and spatial analytical pipelines that support data-driven decision making and evidence-based policy design. The proposed framework contributes to the development of analytical systems that enhance the strategic utilization of complex environmental and public health data.