LO
PilotCognitive automation2021-01-04
Machine-learning alerts during emergency calls did not help dispatchers recognise more cardiac arrests, though the model alone was more sensitive, a Copenhagen randomised trial found
Emergency dispatcheroccupation page →
Event date / reported
2021-01-04
Evidence stage
PilotSmall-scale trial in a real setting. Tells us the deployment conditions are being tested, not that they hold — so one pilot is never enough on its own; two independent ones are.
Tasks this bears on
Answering and triaging calls
Answering emergency and non-emergency calls, working out what is happening and deciding how urgent it is.
Being augmented✓ Evidence-backed
Giving instructions until help arrives
Talking callers through CPR, bleeding control or getting to safety, and staying on the line with them.
Still human-led✓ Evidence-backed
Where this applies
Copenhagen, Denmark, live emergency calls. Of 169 049 calls examined, dispatchers with machine-learning alerts recognised 93.1 percent of confirmed cardiac arrests against 90.5 percent without, not a significant difference; the model alone was more sensitive than dispatchers (85.0 vs 77.5 percent) but had a far lower positive predictive value. The dispatchers in Copenhagen are nurses and paramedics; the trial tests alerts, not automated call handling.
What this means
An AI can flag cardiac arrests on calls, but flagging them did not make dispatchers better at recognising them.
What it does not yet show
One trial in one city, where dispatchers are clinicians; later systems may differ.
What you can check
Open PubMed 33404620 (JAMA Network Open, 2021) and find "did not find any significant improvement".
Does it change the assessment?
No. The impact index is never moved by a single event, and this stage does not move one on its own: a "Pilot" record counts toward a judgement but needs a second, independent record before the judgement rests on evidence. This one is counted; on its own it changed nothing.
Source
Blomberg SN et al. — Effect of Machine Learning on Dispatcher Recognition of Out-of-Hospital Cardiac Arrest During Calls to Emergency Medical Services: A Randomized Clinical Trial, JAMA Network Open (published 4 January 2021; PubMed 33404620) · verified 2026-10-01 · Claude (VOLO agent) · interpreted 2026-10-01 · Claude (VOLO agent)
Primary source — published by the party that did this, or the authority of record. No co-signature needed.