Jocelyn Hunyadi ^1,2 Lara S. Savas ³ Kehe Zhang ^1,2 Jeanette E Deason ² Ryan Ramphul ⁴ Melissa Peskin ³ Erica L Frost ³ Cici Bauer ^1,2*

• ¹ Center for Spatial-Temporal Modeling for Applications in Population Sciences, School of Public Health, the University of Texas Health Science Center at Houston, Houston, United States
• ² Department of Biostatistics and Data Science, School of Public Health, University of Texas Health Science Center at Houston, Houston, Texas, United States
• ³ Department of Health Promotion and Behavioral Sciences, School of Public Health, University of Texas Health Science Center at Houston, Houston, Texas, United States
• ⁴ Department of Epidemiology, Human Genetics and Environmental Sciences, School of Public Health, University of Texas Health Science Center at Houston, Houston, Texas, United States

Healthcare resources are often crucial but limited, requiring careful consideration and informed allocation based on population needs and potential healthcare access. In resource allocation settings, availability and accessibility of resources should be examined simultaneously. The two-step floating catchment area (2SFCA) method has been previously used to evaluate spatial accessibility to healthcare resources and services, and to address health-related disparities. The 2SFCA methods have regained significant popularity during the COVID-19 pandemic, as their application proved crucial in addressing priority public health data analysis, modeling, and accessibility challenges. However, comprehensive comparisons of the 2SFCA method input parameters in the context of public health concerns in Texas are lacking. Our study aims to (a) perform a comparative analysis of 2SFCA input parameters on patterns of spatial accessibility and (b) identify a 2SFCA method to guide evaluation of equitable allocation of scarce mental health resources for children and adolescents in Texas. We used the Texas Child Psychiatry Access Network (CPAN) data to assess county-level, regional patterns in access to pediatric psychiatric care, and to identify areas to expand CPAN to mitigate access-related disparities. We further compared accessibility patterns across two kernel density distance decay functions for 10 catchment area specifications. As expected, spatial accessibility measures, such as the spatial accessibility ratio (SPAR), are sensitive to input parameters, particularly the catchment area. However, across all catchment area thresholds, two clusters of counties in southern and central Texas had particularly low accessibility, highlighting the opportunity for expanding the provider network in these areas. Identifying areas with low accessibility can help public health initiatives prioritize regions in need of improved services and resources. The incorporation of additional data on supply capacity and care-seeking behavior would aid in the refinement of estimates for spatial accessibility at the regional level and within larger urban centers.