Outcomes for Patients Who Contact the Emergency Ambulance Service and Are Not Transported to the Emergency Department: A Data Linkage Study

Abstract

Objectives: Emergency ambulance services do not transport all patients to hospital. International literature reports non-transport rates ranging from 3.7-93.7%. In 2017, 38% of the 11 million calls received by ambulance services in England were attended by ambulance but not transported to an Emergency Department (ED). A further 10% received clinical advice over the telephone. Little is known about what happens to patients following a nontransport decision. We aimed to investigate what happens to patients following an emergency ambulance telephone call that resulted in a non-transport decision, using a linked routine data-set. Methods: Six-months individual patient level data from one ambulance service in England, linked with Hospital Episode Statistics and national mortality data, were used to identify subsequent health events (ambulance re-contact, ED attendance, hospital admission, death) within 3 days (primary analysis) and 7 days (secondary analysis) of an ambulance call ending in nontransport to hospital. Non-clinical staff used a priority dispatch system e.g. Medical Priority Dispatch System to prioritize calls for ambulance dispatch. Non-transport to ED was determined by ambulance crew members at scene or clinicians at the emergency operating center when an ambulance was not dispatched (telephone advice). Results: The data linkage rate was 85% for patients who were discharged at scene (43,108/50,894). After removal of deaths associated with end of life care (N ¼ 312), 9% (3,861/42,796) re-contacted the ambulance service, 12.6% (5,412/42,796) attended ED, 6.3% (2,694/42,796) were admitted to hospital, and 0.3% (129/42,796) died within 3 days of the call. Rates were higher for events occurring within 7 days. For example, 12% re-contacted the ambulance service, 16.1% attended ED, 9.3% were admitted to hospital, and 0.5% died. The linkage rate for telephone advice calls was low because ambulance services record less information about these patients (24% 2,514/10,634). A sensitivity analysis identified a range of subsequent event rates: 2.5-10.5% of patients were admitted to hospital and 0.06-0.24% of patient died within 3 days of the call. Conclusions: Most non-transported patients did not have subsequent health events. Deaths after non-transport are an infrequent event that could be selected for more detailed review of individual cases, to facilitate learning and improvement.

<p>Relationship Between Nonconveyed Cases and On-Scene Time Intervals for Emergency Medical Services</p>

“…8,9 EMS providers attending the scene may decide not to convey all patients to hospital. 10 For decades, nonconveyance after dispatching EMS units has been perceived as a hindrance to the quality and efficiency of EMS operations the may jeopardize operational cost-effectiveness, leading to resource wastage. 7 However, Paulin et al and Höglund et al posited that trends of EMS might be changing, as EMS providers are increasingly assessing and treating patients without conveyance to hospital.…”

“…16 Outcomes for patients who contact EMSs and are not conveyed to an ED might change. 10 Therefore, what happens to patients after nonconveyance is an interesting area to explore. A retrospective investigation reported that within 3 days of call, 9% had recontacted the ambulance service, 12.6% attended an ED, 6.3% had been admitted to hospital, and 0.3% had died.…”

“…Rates increased within 7 days: 12% recontacted the ambulance service, 16.1% attended an ED, 9.3% were admitted to hospital, and 0.5% died. 10 Other evidence showed that within 24hours after nonconveyance, 2.5%-6.1% of patients had had recontacted EMSs and 4.6%-19.0% had presented themselves at an ED. Mortality rates vary from 0.2%-3.5% after 24 hours to 0.3%-6.1% after 72 hours.…”

Outpatient Care in Acute and Prehospital Emergency Medicine by Emergency Medical and Patient Transport Service Over a 10-year Period: A Retrospective Study Based on Dispatch Data From a German Emergency Medical Control Center (OFF-RESCUE)

“…In addition, protocols for handling low-priority patients have been successfully implemented, which lead either to the outpatient usage of a nurse or to telephone consultation [24,25]. The decision on whether a patient should be conveyed, should not be made by paramedics due to a lack of patient safety; several studies have shown inconsistent ndings [26,27,28]. Participation of general practitioners in prehospital emergency care would help [29].…”

Using deterministic record linkage to link ambulance and emergency department data: is it possible without patient identifiers?

“…Historically, in the UK, prehospital patient records from emergency ambulance services and hospital emergency department (ED) or inpatient records have not been routinely linked, though some recent research has begun to explore linkage [13,14]. This is unsurprising, as the unique identifier for each patient (known as the National Health Service [NHS] number) is not commonly collected in ambulance records and thus there is no specific unique identifier available.…”