Abstract The growing prevalence of hypertension, a key chronic disease, amidst urbanization is analyzed through spatial similarity and dependence using correlation and spatial autocorrelation. A Multiscale Geographically Weighted Regression (MGWR) model is employed to investigate the spatial heterogeneity of environmental factors affecting hypertension prevalence, providing a deeper insight into its variation drivers. The findings reveal significant spatial non-stationarity in hypertension prevalence across New York City. Winter surface temperature, building density, and land use mix have broad and profound effects on hypertension, requiring regional policies and interventions. In contrast, factors like road network density and functional mix have more localized impacts, necessitating more targeted interventions at the community or individual behavior level. The MGWR model effectively assesses the impact of urban environments on hypertension prevalence and ranks the importance of influencing factors. With an R-squared value of 0.785, the MGWR model, outperforms both Ordinary Least Squares (OLS) and Geographically Weighted Regression (GWR) models in identifying the determinants of hypertension prevalence. This study offers valuable decision-making support for the government in formulating intervention policies and urban planning and design to reduce hypertension incidence. Future, evidence-based design research is needed to further clarify the specific parameter values ​​of the factors affecting the prevalence of hypertension.