Optimizing Deployment of Automated External Defibrillators for Out-of-Hospital Cardiac Arrest: A Province-Wide Approach

Title: Optimizing deployment of automated external defibrillators for out-of-hospital cardiac arrest: A province-wide approach

Introduction: Out-of-hospital cardiac arrest (OHCA) affects over 60,000 Canadians annually, with only 10% surviving to hospital discharge. Timely defibrillation using publicly available automated external defibrillators (AEDs) is key to improving OHCA outcomes, but AED usage for OHCA remains low. Increasing AED accessibility to OHCAs has been shown to improve survival; however, these studies have typically been done only in urban settings and not on a region-wide basis.

Objective: We sought to develop strategies to improve AED access for OHCAs across the province of Nova Scotia.

Methods: We included all OHCAs recorded by Nova Scotia Emergency Health Services between Jan. 2018 and Dec. 2023. Existing AEDs registered as of Dec. 2023 were included. We determined the number of OHCAs with an existing AED within a 3-minute (160 m) round trip walk, subject to the AED’s hours of availability. We then considered two interventions: (1) converting all existing AEDs to be accessible 24 hours a day, 7 days a week, and (2) utilizing a mathematical optimization model to determine ideal locations for future AEDs based on historical OHCA locations. We determined the number of optimized AEDs required to achieve 10% and 25% accessibility to all OHCAs then compared accessibility within each of Nova Scotia’s 54 community clusters in both cases.

Results: A total of 6,553 OHCAs and 2,052 existing AEDs were included. Under existing AED hours of availability, a total of 223 (3.4%) OHCAs had at least one AED within a 3-minute walk; this increased to 381 (5.8%) OHCAs assuming 24/7 availability of existing AEDs. Achieving 10% and 25% accessibility OHCA accessibility required deploying 40 and 300 optimally located AEDs, respectively. The median AED accessibility amongst community clusters assuming existing hours of operation was 1.4% [IQR: 0.0-2.9%], increasing to 3.0% [IQR: 1.4-5.8%] assuming 24/7 availability.

Conclusion: Low accessibility between existing AEDs and OHCAs suggests poor alignment between AED and OHCA locations. Optimization modeling can identify ideal locations for new AEDs and increase accessibility for OHCAs with relatively few AEDs required.