목적: This study proposes and validates the Map Similarity Index as a novel framework to quantify the spatial relationship between environmental noise and hearing loss. Moving beyond a sole reliance on conventional regression metrics (e.g., R2), the Map Similarity Index focuses on the visual and structural concordance between noise exposure maps and patient distribution patterns. 방법:Using nationwide National Health Insurance big data, we established a reference map of hearing-loss distribution and integrated it with multi- source spatial layers—including environmental noise, topography, and land-use—onto a unified grid system (1,000 m resolution). The Map Similarity Index was calculated by integrating three key spatial dimensions: (1) the physical overlap of hotspots (Intersection over Union, IoU), (2) the structural agreement in intensity and gradients (SSIM), and (3) the similarity in spatial autocorrelation patterns (MoranSim). 결과:The Map Similarity Index-based association demonstrated explanatory power statistically equivalent to traditional regression models after adjusting for effective sample size to account for spatial autocorrelation. Furthermore, the index exhibited high scale-robustness, maintaining stable values across varying map resolutions and threshold settings. 결론:The Map Similarity Index framework offers an intuitive, map- centered analytical tool that complements or provides an alternative to traditional statistical indicators. By enabling rapid and accurate assessments of health impacts through the spatial alignment of data layers, the Map Similarity Index facilitates more accessible and evidence-based public health policymaking.