Strategic automated external defibrillator deployment based on geospatial analysis of historical cardiac arrest...

We hypothesized that historical out-of-hospital sudden cardiac arrest (SCA) locations over multiple years can inform deployment of automated external defibrillators (AEDs).

Consecutive SCA events attended by EMS were ascertained prospectively in two identically designed community-based studies: PRESTO (Ventura County, California, population ∼ 850,000; 2015-2023; SCA n = 2,935) and SUDS (Multnomah County, Oregon, population ∼ 800,000; 2012-2020; SCA n = 2,213). Clusters of ≥ 3 historical SCA events within 100-meters were plotted using geospatial analysis (QGIS Desktop 3.23.3). Proposed AED locations were positioned within 200-meters. An existing AED registry in Ventura County (n = 998) was also plotted using geospatial analysis. Distances between SCAs and proposed AEDs were calculated using Euclidean distances.

In PRESTO, 97 clusters of ≥ 3 SCAs were identified, amounting to 22% (n = 653) of total SCAs, covered by 90 proposed AEDs. The mean distance between individual SCA events and the nearest AED was 99.7 (SD = 53.1) meters, representing a proposed bystander AED response time of 3.0 (SD = 1.6) minutes. Proposed bystander AED response times were significantly shorter than actual EMS response times [mean difference: 3.1 (95% CI: 2.9-3.4), p < 0.0001]. Similar results were observed in SUDS. Compared to the existing AED registry in Ventura County, 18.9% (n = 17) of our 90 proposed AEDs were within 200 m of an existing AED. Of total SCAs in Ventura County,11.4% (n = 334) were within 200 m of an existing AED.

16-22% of historical SCA events in Ventura and Multnomah Counties could potentially be covered by 78-90 AEDs. Outcomes studies that compare this geospatial approach to current AED location strategies are warranted.