What happened
In April 2011 Michael Sweeney, a sporadic cocaine user, entered a public house with a knife. He was extremely agitated. Officers identified almost immediately that he was unwell, took his condition to be the medical emergency their training called excited delirium, and asked control to arrange an ambulance.
The London Ambulance Service categorised the call as C1 Amber, not Red One or Red Two. No paramedics were working in ambulance control at the time who could have recognised how serious his condition was and upgraded it, and no ambulance arrived within the target time.
Twenty minutes after the request, police took him to the Royal London Hospital in a police van. There he stayed violently agitated, trying to hurt himself and resisting help. He was restrained prone until sedation took effect, and then turned over. He arrested within a minute of being turned, and died less than two hours later.
The outcome
The inquest concluded on 18 September 2013. The jury found that he died by accident, giving the medical cause as the acute toxic effects of cocaine, with restraint and struggling in association with acute behavioural disturbance. Ten times the usual recreational level of cocaine was in his blood.
The jury found that the delay in getting him medical assistance resulted in over-exertion during his struggling and restraint. On 23 September 2013 the Senior Coroner, Mary Hassell, issued a Regulation 28 report to the Metropolitan Police Service and the London Ambulance Service, requiring a response within 56 days.
What went wrong
The failure sat between two services, and no one owned the space.
Police used the term their training gave them. The coroner found it was not widely used in this country: neither ambulance, nursing nor some medical staff had heard of it. The label did not translate into a category the receiving service acted on, so the call went in as C1 Amber and stayed there. Nobody with clinical training was in ambulance control to hear it and upgrade it.
The coroner added a second warning. Treating excited delirium as an apparent diagnosis risks missing organic causes: a bleed on the brain, sepsis, a diabetic coma.
What it teaches
Call-taking is a clinical intervention. A request describing active restraint in an agitated patient has to reach someone who can upgrade it, and the words used have to map onto a category the receiving service acts on. A label nobody else recognises is not a handover.
The restraint carries the second lesson. It was face down, in a hospital, and it lasted until the sedation worked. The inquest put the over-exertion with the delay, not the hold alone: the hold lasted as long as it did because the ambulance had not come.
Questions for the room
- You are the officer at the pub door. What do you say down the radio that cannot be filed as C1 Amber?
- Excited delirium was in police training and unknown in ambulance control. Whose job is it to check that a shared vocabulary is actually shared?
- Who owns a call that describes restraint, and would they know to upgrade it?
Evidence and limits
Confidence: probable. The coroner’s Regulation 28 report was read directly, but no independent second source was found, so the two-source test is not met. deathlessons.org reproduces the same document and is not independent. The inquest concluded on 18 September 2013, so the findings may be cited with that date.
The London Ambulance Service replied on 14 November 2013, declining the term “extreme agitation” in its response.
No member of staff is named.
Our vulnerable and volatile persons training covers acute behavioural disturbance and what has to happen before hands are used.
Sources
- Courts and Tribunals Judiciary, Regulation 28 report 2013-0236, 23 September 2013
- deathlessons.org, verbatim text and LAS response, 2013
Reviewed 20 October 2026.