Abstract 58: Locating AEDs in an Urban City: A Geospatial View

Background: Survival from out-of-hospital cardiac arrest (OHCA) is poor and varies by geography. Variability in AED locations may be a contributing factor, but this is understudied. As publically accessible AEDs are more likely to be in commercial rather than residential areas, we sought to determine AED density in a large urban city relative to population and employment demographics. Methods: This was a retrospective analysis of MyHeartMap, a prospective cohort of AEDs in Philadelphia County. Geographic and population characteristics were obtained from US census and Local Employment Dynamics (LED) data by zipcode. Census characteristics included number of residents, percent of residents with a high school degree or above, median household income, and percent of African Americans. LED characteristics included total number of jobs reported, percent of jobs requiring a high school degree or above, percent of jobs with an annual income above $40,000, and percent of African Americans in jobs. Multivariate regression models using a Poisson distribution related the distribution of AEDs to population and employment characteristics using commercially available geographical information systems (GIS) software (ArcGIS Redlands CA) Results: AED locations (n=1295) and density per zip code (range 2-179) varied across the city. Population and employment characteristics were significantly related to the number of AEDs per zipcode (p<0.001) [R-square=0.38 and 0.74, respectively]. The incident rate ratio (IRR) for AED density was higher in zipcodes with higher-education (1.10, p<.01) and lower in zipcodes with more African American residents (0.99, p<.01). Regarding employment characteristics, the IRR for AEDs was higher in zipcodes with jobs requiring a higher education (1.03, p<.01) and jobs with higher income (1.04, p<.01) and lower in zipcodes with more African American employees (0.99, p<.01). Conclusions: Significantly fewer AEDs were located in areas with lower socioeconomic status. Employment characteristics better explained this variation across zipcodes. A more thorough study of this variability is important as optimizing the placement of AEDs in public areas may help to increase OHCA survival.