GIS and area deprivation: An approach to neighborhood level population health management

Background: It has been shown that neighborhood-level characteristics can often influence health outcomes. This suggests that primary care providers should also be cognizant of a patient’s neighborhood environment in order to provide the best treatment, inclusive of preventive care. However this can be a challenge with respect to the ease of access, interpretation, operationalization, and generalizability of this neighborhood data.

Objective/Purpose: To determine the relationship between an area deprivation index and selected health outcomes at the regional (Hudson Valley, NY) and local (e.g., county) levels and to propose a simple method of disseminating this data along with potential resources interactively.

Methods: Using a geographic information science (GIS) framework, we examined the Area Deprivation Index (ADI), developed by Singh and updated by Kind, which is freely accessible and available for the entire United States. The measure represents the socioeconomic deprivation of these areas using 17 census markers. Regional (Hudson Valley, NY), and local (e.g., county-wide) analyses were performed by correlating ADI values with selected hospitalization rates (cardiovascular disease, substance abuse, and asthma). This multi-scale look was done in order to examine local variation in correlations as well as to better reflect primary care and county social services availability. Spatial data was then transformed into easily accessible Google Earth files to provide interactive ability.

Results: Regional ADI value rankings concealed some important local variation in the ADI/health relationships.

Discussion/Conclusions: Presenting neighborhood-level characteristics may help providers understand the relative deprivation and risk experienced by the patient based on potential influences of the social and built environments, otherwise referred to as the broader determinants of health. Incorporating this information with health data, utilization, and costs available from claims databases will provide even richer datasets for analytics, public health surveillance, allocation of community resources, and overall population health management.