Modeling the impact of public access defibrillator range on public location cardiac arrest coverage

Auyon Siddiq, Steven C. Brooks, Timothy C. Y. ChanResuscitation 2012 · DOI

Abstract

Public Access Defibrillation (PAD) programs aim to reduce mortality from out-of-hospital cardiac arrests (OHCAs) by placing Automated External Defibrillators (AEDs) in public spaces. However, these devices are used in fewer than 3% of cases. This study introduces the concept of "effective range"—the distance a bystander can travel to retrieve an AED before emergency services arrive—and models its impact on coverage. Using spatial optimization on historical cardiac arrest data from Toronto, we demonstrate that increasing the effective range through improved visibility and accessibility is often more impactful and cost-effective than increasing the total number of devices.

Introduction

Despite the widespread deployment of AEDs in cities worldwide, their impact on survival rates remains limited by low utilization. Current placement strategies often focus on high-traffic areas but fail to account for the "effective range" of a device. Effective range is not merely a physical distance but a composite metric reflecting bystander awareness, device visibility, and accessibility. If an AED is nearby but unknown or inaccessible (e.g., behind a locked door), its effective range is essentially zero. This study seeks to quantify the sensitivity of city-wide cardiac arrest coverage to changes in both the number of AEDs and their effective range.

Methodology

The study utilizes the Maximal Covering Location Problem (MCLP) framework, a classic operations research challenge, to optimize AED placement. The objective was to select a subset of locations that maximizes the coverage of historical cardiac arrests.

Data Sources

Two primary datasets from Toronto, Canada, were analyzed:

  1. ROC Epistry: A prospective registry of 1,310 public-location, non-traumatic, EMS-attended cardiac arrests occurring between 2005 and 2010.
  2. Toronto Employment Survey (2009): A database of 25,851 non-residential buildings, which served as the candidate sites where AEDs could theoretically be placed.

Simulation Design

The researchers simulated 600 unique scenarios, varying:

  • Quantity: 10 to 200 AEDs.
  • Effective Range: 10 to 300 meters.

Additionally, an "idealized" scenario was calculated to determine the maximum possible coverage if an AED were placed in every single non-residential building in the city.

Results

The analysis revealed critical insights into the spatial dynamics of PAD programs and the diminishing returns of hardware-only expansion.

The Limits of Saturation

Even in an ultra-optimistic scenario where all ~26,000 public buildings in Toronto were equipped with an AED, coverage remained highly dependent on range. If the effective range was limited to 50 meters (roughly line-of-sight), only 50.9% of historical arrests would have been covered.

Range vs. Quantity Trade-offs

There is a significant trade-off between the number of devices deployed and their individual reach. For example, to increase coverage from 15% to 20% at a short effective range (30m), a city would need to deploy 70 additional AEDs. However, if the effective range is improved to 100m, that same 5% boost in coverage requires only 15 additional AEDs.

Diminishing Returns

While increasing range improves coverage, the gains are non-linear. For a deployment of 200 AEDs, increasing the range from 100m to 200m adds 15% coverage. A further increase to 300m adds only 10%, as cardiac arrests are spatially clustered and high-traffic "hotspots" become saturated.

Discussion

The findings suggest that "range-extending" interventions—such as improved signage, public AED registries, and bystander notification apps—can be more impactful than hardware procurement. By increasing the probability that a bystander finds and retrieves a device, these technologies effectively expand the spatial reach of existing infrastructure without the cost of new hardware.

Limitations

The model employs Euclidean (straight-line) distance, which may overestimate coverage by ignoring physical barriers like walls or traffic. Furthermore, the 2D nature of the model does not account for vertical travel time in high-rise buildings, a significant factor in dense urban environments like Toronto. Future work should include formal economic assessments to compare the cost-effectiveness of buying more AEDs versus investing in digital range-extending technologies.

Conclusion

Optimizing PAD programs requires a shift in focus from quantity to accessibility. Improving the effective range of a small, well-placed fleet of AEDs can yield coverage comparable to much larger, less visible deployments. Strategic investment in visibility and bystander engagement is essential to maximizing the life-saving potential of public defibrillators.

Summary updated

BibTeX
@inproceedings{siddiq2012modeling,
  author = {Auyon Siddiq and Steven C. Brooks and Timothy C. Y. Chan},
  title = {{Modeling the impact of public access defibrillator range on public location cardiac arrest coverage}},
  booktitle = {Resuscitation},
  year = {2012},
  doi = {10.1016/j.resuscitation.2012.11.019},
}
Figure