Spatiotemporal AED optimization is generalizable.

Area of Science:

• Public Health
• Emergency Medicine
• Spatial Analysis

Background:

• Automated external defibrillator (AED) placement optimization can improve out-of-hospital cardiac arrest (OHCA) coverage.
• The generalizability of existing AED placement optimization methods remains under-investigated.

Purpose of the Study:

• To assess the performance and generalizability of a spatiotemporal AED placement optimization methodology in Copenhagen, Denmark.
• To evaluate if a model developed for Toronto, Canada, is applicable to a different urban setting.

Main Methods:

• Utilized data on public OHCAs (1994-2016) and registered AEDs (2016) in Copenhagen.
• Calculated coverage loss due to AED temporal accessibility limitations.
• Applied a spatiotemporal optimization model to quantify potential coverage gains.
• Compared spatiotemporal deployment gains against spatial-only solutions using 10-fold cross-validation.

Main Results:

• Identified 2149 OHCAs and 1573 AED locations in Copenhagen.
• Found a coverage loss of 24.4% under actual AED availability.
• Demonstrated a 15.3% coverage gain using the spatiotemporal model over a spatial-only approach.
• Observed temporal and geographical coverage gain trends similar to those in Toronto.

Conclusions:

• A spatiotemporal AED optimization approach developed for Toronto is effective in Copenhagen without modification.
• The study highlights the generalizability of this optimization methodology across diverse cities.
• Findings underscore the critical role of temporal AED accessibility and the potential for improved placement strategies.