The spatial equity of the healthcare system is an important factor in assessing how the different medical service demands of residents are met by different levels of medical institutions. However, previous studies have not paid sufficient attention to multilevel healthcare accessibility based on both the divergence of hierarchical healthcare supplies and variations in residents' behavioral preferences for different types of healthcare. This study aims to propose a demand-driven "2R grid-to-level" (2R-GTL) method of analyzing the spatial equity in access to a multilevel healthcare system in Chengdu. Gridded populations, real-time travel distances and residents' spatial behavioral preferences were used to generate a dynamic and accurate healthcare accessibility assessment. The results indicate that significant differences exist in the spatial accessibility to different levels of healthcare. Approximately 90% of the total population living in 57% of the total area in the city can access all three levels of healthcare within an acceptable travel distance, whereas multilevel healthcare shortage zones cover 42% of the total area and 12% of the population. A lack of primary healthcare is the most serious problem in these healthcare shortage zones. These results support the systematic monitoring of multilevel healthcare accessibility by decision-makers. The method proposed in this research could be improved by introducing nonspatial factors, private healthcare providers and other cultural contexts and time periods.
Keywords: 2R grid-to-level (2R-GTL); multilevel healthcare; spatial accessibility; spatial equity.