Preview was written by Peng Jia
Q: What is the main purpose of your study?
A: Hospital Service Area (HSA) and Hospital Referral Region (HRR), known as a hierarchical HSA system, have been used as analysis units in a growing body of large-scale studies of healthcare spending, utilization, and quality in the U.S. However, the popular Dartmouth HSAs and HRRs were produced more than two decades ago and unable to represent contemporary healthcare markets. This research uses a revised Huff model to delineate two levels of hospital service areas in Florida.
Q: What are the practical, day to day implications of your study?
A: Our method automated in Geographic Information Systems (GIS) can be easily replicated in other regions to define large-scale and consistent hierarchical Hospital Service Area (HSA) systems.
Q: How does your study relate to other work on the subject?
A: The popular Dartmouth HSAs and HRRs were produced more than two decades ago and unable to represent contemporary healthcare markets. Three elements distinguish our method from existing work. First, it strengthens the popular Huff model’s theoretical foundation in individual spatial behavior. Secondly, the hierarchical central place structure is supported by the differing spatial behavior of patients for different services. Finally, the method’s automation in Geographic Information Systems (GIS) enables its easy implementation.
Q: What are two or three interesting findings that come from your study?
A: 1. General patients experience a stepper gradient and thus a shorter average travel range that supports delineating more HSAs of smaller area size, and specialized patients exhibit a flatter gradient and thus a longer average travel range that leads to fewer HRRs of larger area size.
2. This study reveals the divergence of travel behavior of specialized patients (cardiovascular vs. neuro patients) that were grouped together in the traditional Dartmouth approach in delineating HRRs, and suggests a need to revisit the guideline for defining HRRs.
Q: What might be some of the theoretical implications of this study?
A: 1. The power function for distance decay behavior is used in the traditional Huff model. It needs to be replaced by a best-fitting function derived from the actual travel patterns of patients.
2. The hierarchical central place structure needs to be supported by the differing travel friction coefficients for different orders of services (i.e., lower-order services have a stepper travel friction gradient than higher-order ones).
Q: How does your research help us think about Geography?
A: Geography is very much relevant in public policy. Here our research advances the method for delineating Hospital Service Areas that have been widely used in assessing the health care market efficiency.