Studying the spatial pattern distribution of respiratory health risks and the role of spatial environmental factors can improve our understanding of pathogenic mechanisms in the spatial environment. However, previous air quality index (AQI) sensor models lacked comprehensive air quality reflection, limiting risk prediction accuracy. In this study, we improved the AQI sensor model and combined it with population density data to establish a respiratory health risk exposure assessment model. Twenty-five spatial environmental variables were selected as potential factors. Pearson correlation analysis and a geodetector were used to assess the spatial risk patterns of respiratory diseases and determine the characteristics of the selected factors. The results indicated that (1) the health risk index showed autumn < summer < spring < winter ranking, with the health risk gradually decreasing from the center outwards. (2) The positive effect of the volume ratio, a spatial morphology factor, was the strongest, and the negative effect of the sky openness was the greatest. (3) The geodetector results showed significant spatial heterogeneity in the degree of effect of the spatial environmental factors on respiratory health risks. Moreover, the explanatory power of the interaction between any two factors (except the volume ratio) far exceeded that of a single factor.