Contextualizing social-spatial determinants of health services

A renewed focus on social determinants of health has created the need for better understanding of the context in which care and health occur. However, the way we measure place is lacking a fine-scale gold-standard smaller than... [ view full abstract ]

A renewed focus on social determinants of health has created the need for better understanding of the context in which care and health occur. However, the way we measure place is lacking a fine-scale gold-standard smaller than the currently used county or zip code levels. These large coarse-scale areas are often not well-aligned with how human interactions occur. To study the effect of spatial scale on health services, we compared a catchment model to measurement of a hospital’s local service area and need for language services using a social-spatial approach at the localized hospital level and the broader county level that is the current standard among most hospitals conducting community health needs assessments. Language services were the proportion of limited English proficient population at the county level and at the individual hospital service area level.

Two major datasets were integrated for this study. First, general hospitals from the American Hospital Association 2013 (N=4514) database were used to create the hospital service areas layer and whether or not hospitals offered language services. The second dataset used the 2009-2013 American Community Survey for creating a second layer containing estimates of need (high/medium/low) for language services within each service area from the first layer and compared to county boundaries also from the ACS.

We found that this modelling approach did facilitate visualization of social-spatial differences when finer-scale methods were employed as compared to the county level measures. This is a necessary preliminary step in order to better understand the role of spatial statistics in addressing social determinants of health. The catchment area method is not without limitations but the purpose of the present analysis is to elucidate the importance of fine-scale visualization as compared to current coarse-level practices in the healthcare delivery field to facilitate the study of social-spatial determinants of health.