Automatic measurement of departing times in smartphone alerting systems: A pilot study

Smartphone alerting systems (SAS) alert volunteers in close vicinity of suspected out-of-hospital cardiac arrest. Some systems use sophisticated algorithms to select those who will probably arrive first. Precise estimation of departing times and travel times may help to further improve algorithms. We developed a global positioning system (GPS) based method for automatic measurements of departing times. The aim of this pilot study was to evaluate feasibility and precision of the method.

Region of Lifesavers alerting app (iOS/ Android, version 3.0, FirstAED ApS, Denmark) was used in this study. 27 experiments were performed with 9 students, who were instructed to stay in their flats during the study days. A geofence was set for each alarm in the alerting system with a radius of 10 m (8 cases), 15 m (10 cases), and 20 m (9 cases) around the GPS position at which the alarm was accepted in the app. The system logged responders as being departed when the smartphone position was registered outside the geofence. The students were instructed to manually start a stopwatch at the time of the alert and to stop the stopwatch once they had entered the street in front of their flat.

The median difference between automatically and manually retrieved times were -16 seconds [interquartile range IQR 50 seconds] (geofence 10 m), 30 seconds [IQR 25 seconds] (15 m), and 20 seconds [IQR 13 seconds] (20 m), respectively. The 20 m geofence was associated with the smallest interquartile range.

Departing times of volunteer responders in SAS can be retrieved automatically using GPS and a geofence.