<p>The COVID-19 pandemic has underscored the significance of spatial context in understanding patterns of disease transmission, particularly in rapidly urbanizing regions of the Global South. Despite growing research on the topic, medium-sized cities remain underrepresented in high-resolution spatial analyses that integrate detailed land use characteristics. This study addresses this gap by examining the spatial determinants of COVID-19 infections across Kozhikode City, Kerala, with a specific focus on residential land-use intensity derived from building-height data. Using the Global Human Settlement Layer (GHSL) and local zoning maps, population distribution was derived from vertical residential land-use typologies, enabling a more refined representation of urban density. By incorporating building height to capture vertical residential intensity, the height-informed population proportion metric provides a more realistic representation of population distribution than conventional density measures. Clustering analysis, followed by Ordinary Least Squares (OLS) and Geographically Weighted Regression (GWR) models, was applied to assess the influence of variables, including residential density, essential employment, public transport dependency, socio-economic vulnerability, and the stringency of public health measures. The results highlight strong spatial correlations between COVID-19 incidence and areas characterized by high-rise residential buildings, transit access, and economic precarity. This research contributes to policy-relevant urban health planning by demonstrating how built form and vertical urban structure are associated with spatial patterns of health vulnerability, advocating for more inclusive, resilience-oriented planning strategies in emerging urban regions.</p>