Response-time variability
A machine learning-based analysis of how workload, priority, geography, weather, call handling, travel, and resource availability shape EMS response times.
Response time as emergent system behaviour
The study examines EMS response time as a systems metric shaped by interacting operational, environmental, geographic, and priority-related factors.
Why response time varies across the EMS system
Multiple drivers interacted
Response time was shaped by call priority, workload, geography, weather, call handling, travel time, and resource availability rather than by any single determinant.
Priority changed system behaviour
High-priority calls were protected differently from lower-priority calls, showing how allocation rules redistribute delay across priority groups.
Tail delays matter
Lower-priority response-time distributions showed long-tail delay patterns, supporting governance approaches that look beyond averages and median targets.
From response-time target to system diagnosis
Study IV supports a shift from single-point response-time benchmarking toward distribution-aware EMS governance.
The study shows that response time reflects the interaction between clinical priority, system strain, geography, weather, and operational intervals. This makes response time useful as a systems signal, but risky as a simplistic performance target if variability and tail delays are ignored.
Interpretation note
This study is observational and explanatory. It supports system understanding and governance of response-time variability, but it should not be interpreted as a directly deployable real-time prediction system. Operational use would require prospective validation, integration with live data streams, human-factors design, and governance safeguards.
