Calculating real-world travel routes instead of straight-line distance in the community response to out-of-hospital cardiac arrest

Background: Using straight-line distance to estimate the proximity of public-access Automated External Defibrillators (AEDs) or volunteer first-responders to potential out-of-hospital cardiac arrests (OHCAs) does not reflect real-world travel distance. The dierence between estimates may be an important consideration for bystanders and first-responders responding to OHCAs and may potentially impact patient outcome. Objectives: To explore how calculating real-world travel routes instead of using straight-line distance estimates might impact the community response to OHCA. Methods: We mapped 4355 OHCA (01/04/2016-31/03/2017) and 2677 AEDs in London (UK), and 1263 OHCA (18/06/2017-17/06/2018) and 4704 AEDs in East Midlands (UK) using ArcGIS mapping software. We determined the distance from OHCAs to the nearest AED using straight-line estimates and real-world travel routes. We mapped locations of potential OHCAs (London: n = 9065, 20/09/2019-22/03/2020; East Midlands: n = 7637, 20/09/2019-17/03/2020) for which volunteer first-responders were alerted by the GoodSAM mobile-phone app, and calculated response distance using straight-line estimates and real-world travel routes. We created Receiver Operating Characteristic (ROC) curves and calculated the Area Under the Curve (AUC) to determine if travel distance predicted whether or not a responder accepted an alert. Results: Real-world travel routes to the nearest AED were (median) 219 m longer (623 m vs 406 m) than straight-line estimates in London, and 211 m longer (568 m vs 357 m) in East Midlands. The identity of the nearest AED changed on 26% occasions in both areas when calculating realworld travel routes. GoodSAM responders' real-world travel routes were (median) 222 m longer (601 m vs 379 m) in London, and 291 m longer (814 m vs 523 m) in East Midlands. AUC statistics for both areas demonstrated that neither straight-line nor real-world travel distance predicted whether or not a responder accepted an alert. Conclusions: Calculating real-world travel routes increases the estimated travel distance and time for those responding to OHCAs. Calculating straight-line distance may overestimate the benefit of the community response to OHCA.