Spatial Analysis of Prehospital Emergency Medical Services Accessibility: A Comparative Evaluation of the GAUSS-Probability Two-Step Floating Catchment Area Model in Handan City

Feng Tian ^1,2,3 Saicong Lu ^2,3,4 Zhenjie Yang ⁵ Tingting Zhao ^1,2,3 Penghui Li ^2,3,6* Haifang Zhang ^5*

• ¹ School of Medicine, Hebei University of Engineering, Handan, Hebei Province, China
• ² Hebei Key Laboratory of Medical Data Science, Handan, Hebei Province, China
• ³ Institute of Biomedical Informatics, Handan, Hebei Province, China
• ⁴ School of Information and Electrical Engineering, Hebei University of Engineering, Handan, China
• ⁵ Handan Emergency Rescue Command Center, Handan, China
• ⁶ Hebei University of Engineering, Handan, China

The evaluation of the accessibility of prehospital emergency medical facilities plays a vital role in the rational allocation of urban medical resources. Proper emergency medical services (EMS) planning based on accessibility evaluations can facilitate a more equitable distribution of healthcare services, thereby reducing disparities in emergency care availability across different urban areas.Prehospital EMS is often highly urgent, requiring effective intervention for patients in the shortest possible time. To meet this need, the model incorporated distance-based selection probabilities to optimize decision making by considering both geographical location and the distribution of emergency resources, making it more compatible with the practical demands of prehospital EMS. In this study, the GAUSS-Probability Two-step Floating Catchment Area (GP2SFCA) method was applied to evaluate the spatial distribution of access to prehospital EMS in Handan City and to assess accessibility differences by comparing it with various models. The results demonstrated that the GP2SFCA model achieved significantly improved performance, with an average correlation coefficient of 0.7017, indicating a notably higher predictive accuracy. In contrast, other models, including the Cumulative-Opportunity Rectangular (CUMR), Kernel Density (KD), and GAUSS models, showed a much lower average correlation coefficient of 0.1542. This comparison showed that the GP2SFCA model achieved optimized accuracy, increasing its applicability to real world scenarios. Our study significantly improved the evaluation and optimization of the spatial distribution of medical resources in the Handan region, offering crucial decision support for enhancing the reachability of regional medical services via its meticulous analysis of accessibility and resource allocation.