A Data-Driven Method for Measuring Accessibility to Healthcare Using the Spatial Interpolation Model

The evaluation of accessibility to healthcare greatly influences public policy. However, existing approaches mainly rely on network topology technology to estimate the driving paths available to healthcare users and the spatial impedances that they may encounter; these approaches do not suffice to represent the real life travel scenarios of urban traffic. This paper proposes a data-driven method to measure healthcare accessibility. The travel records related to visiting a health facility via taxi (VHT) were identified based on the travel destination, and the travel time for each trip was accurately recorded. The spatial interpolation model converts discrete taxi trajectory data into continuous data surfaces, and an improved cumulative accessibility measure method is applied to determine the accessibility to healthcare. A case study using four months of actual Ningbo taxi data shows that the mean absolute percentage error (MAPE) is 7.1%, and the root mean square error (RMSE) is approximately 2.57min. The case study results highlight that changes in travel speed over time have a significant impact on accessibility to healthcare facilities. This method measures the spatial impedance encountered by residents visiting medical facilities by capturing real travel scenarios and presenting the methodological implications of evaluating healthcare accessibility.