A New Data Driven Approach for Identifying Underserved Areas of Hospitals Accepting Medicare.

Medicare is the largest single-payer for healthcare in the USA. It is primarily for senior citizens. Health access, in terms of distance of commute, and quality of healthcare facilities, is a critical issue for senior citizens, as America has a rapidly growing aging population. The objective of this study is to examine the Locational Health Access (LHA) to hospitals accepting Medicare, across mainland USA. The key method is the computation of the LHA index. This includes road distances from population centroid to hospitals, and the facilities of the hospitals (number of beds used as a proxy). The number of census tracts is more than 84000 and the number of hospitals is more than 4000, so the matrix is very large to compute. This study devised a unique algorithm based on the map of hospitals and census tracts to find underserved areas, using a tree data structure. It demonstrated a practical implementation of data-driven solutions in this domain. Results indicate that rural counties with low LHA values experience significantly greater distances to the nearest hospital, higher proportions of elderly populations, and economic constraints that further limit access to healthcare. In contrast, urban counties with high LHA values demonstrate shorter hospital distances, larger healthcare infrastructure, and stronger economic conditions. Computational challenges in distance matrix generation and limitations in dataset completeness highlight areas for future improvement, some of which have been attempted in this paper. This study underscores the need for enhanced healthcare infrastructure in underserved areas, and expanded dataset integration to improve health accessibility assessments.