Emergency Stroke Calls: Obtaining Rapid Telephone Triage (ESCORTT) - a programme of research to facilitate recognition of stroke by emergency medical dispatchers

Abstract

Background: Rapid access to emergency stroke care can reduce death and disability by enabling immediate provision of interventions such as thrombolysis, physiological monitoring and stabilisation. One of the ways that access to services can be facilitated is through emergency medical service (EMS)dispatchers. The sensitivity of EMS dispatchers for identifying stroke is < 50%. Studies have shown that activation of the EMSs is the single most important factor in the rapid triage and treatment of acute
stroke patients.

Objectives: To facilitate recognition of stroke by emergency medical dispatchers (EMDs).

Design: An eight-phase mixed-methods study. Phase 1: a retrospective cohort study exploring stroke diagnosis. Phase 2: semi-structured interviews exploring public and EMS interactions.
Phases 3 and 4: a content analysis of 999 calls exploring the interaction between the public and EMDs.
Phases 5–7: development and implementation of stroke-specific online training (based on phases 1–4).
Phase 8: an interrupted time series exploring the impact of the online training.

Setting: One ambulance service and four hospitals.
Participants: Patients arriving at hospital by ambulance with stroke suspected somewhere on the stroke pathway (phases 1 and 8). Patients arriving at hospital by ambulance with a final diagnosis of stroke (phase 2). Calls to the EMSs relating to phase 1 patients (phases 3 and 4). EMDs (phase 7).

Interventions: Stroke-specific online training package, designed to improve recognition of stroke for EMDs.

Main outcome measures: Phase 1: symptoms indicative of a final and dispatch diagnosis of stroke.
Phase 2: factors involved in the decision to call the EMSs when stroke is suspected.
Phases 3 and 4: keywords used by the public when describing stroke and non-stroke symptoms to EMDs.
Phase 8: proportion of patients with a final diagnosis of stroke correctly dispatched as stroke by EMDs.

Results: Phase 1: for patients with a final diagnosis of stroke, facial weakness and speech problems were
significantly associated with an EMD code of stroke. Phase 2: four factors were identified – perceived seriousness; seeking and receiving lay or professional advice; caller’s description of symptoms and emotional response to symptoms. Phases 3 and 4: mention of ‘stroke’ or one or more Face Arm Speech Test (FAST) items is much more common in stroke compared with non-stroke calls. Consciousness level
was often difficult for callers to determine and/or communicate. Phase 8: there was a significant difference
(p = 0.003) in proportions correctly dispatched as stroke – before the training was implemented 58 out of 92 (63%); during implementation of training 42 out of 48 (88%); and after training implemented 47 out of 59 (80%).

Conclusions: EMDs should be aware that callers are likely to describe loss of function (e.g. unable to grip) rather than symptoms (e.g. weakness) and that callers using the word ‘stroke’ or describing facial weakness, limb weakness or speech problems are likely to be calling about a stroke. Ambiguities and contradictions in dialogue about consciousness level arise during ambulance calls for suspected and confirmed stroke. The online training package improved recognition of stroke by EMDs. Recommendations
for future research include testing the effectiveness of the Emergency Stroke Calls: Obtaining Rapid Telephone Triage (ESCORTT) training package on the recognition of stroke across other EMSs in England; and exploring the impact of the early identification of stroke by call handlers on patient and process outcomes.

Funding: The National Institute for Health Research Programme Grants for Applied Research programme.