Time to train all doctors to look after seriously ill patients

Objectives: To determine whether patients brought to the Accident and Emergency (A&E) Department by 999 ambulance genuinely need to attend hospital.

Setting and Subjects: 314 patients transported to the A&E department of the Southern General Hospital by 999 ambulances during the 3-week study period.

Complete data was available for 290/314 (92.4%) patients. 126 (40.1%) of 999 ambulances arrived between 9am and 5pm, 127 (40.4%) between 5pm and 1am and 61 (19.4%) between 1am and 9am. 202 patients (73.1%) had one or more investigations carried out in the A&E department. 32 (11.0%) had a dressing, 118 (40.7%) had drugs prescribed, 27 (9.3%) had sutures and 4 (1.4%) had a plaster cast applied. 179 (57%) patients were referred to inpatient specialties and 95 (30.3%) were discharged without follow up, 23 (7.9%) of these having required no investigations or treatment. 16 (5.1%) were discharged to outpatient follow up, 6 (1.9%) were transferred to another hospital and 2 patients (0.6%) died. 16 patients (5.1%) discharged themselves against medical advice.

The majority of patients currently brought to hospital by 999 ambulance attend appropriately. Over 90% of patients receive investigations or treatment or are admitted. The introduction of prioritisation and telephone advice is unlikely to reduce the number of patients requiring transport to hospital.

Calls to the Ambulance Service are increasing. Between 1998 and 1999 the Scottish Ambulance Service reported a 9.4% increase in emergency responses across Scotland, taking the total to 276 753. This increase was greatest in areas of high population density (10.8%).¹ One proposed solution to this increased demand is the introduction of priority-based ambulance dispatch to tailor ambulance responses to the seriousness of calls. It is also hoped that the availability of telephone advice via NHS24 may reduce the number of inappropriate ambulance calls.

Studies into the appropriateness of ambulance calls suggest that 30% to 52% of patients do not require a 999 ambulance.^2,3 These studies used the judgement of senior clinicians considering final diagnosis to determine appropriateness. This disregards the suspected diagnosis at the time of calling 999 and the lack of clinical expertise of callers. Our study examined the need for hospital attendance in patients arriving by 999 ambulance using a “process of care” model including investigations, treatments undertaken and discharge disposition.⁴

The Accident and Emergency department of the Southern General Hospital serves a population of approximately 220,000 with approximately 43,000 new patients annually. This population includes areas of high mortality - Glasgow Pollok (standardised mortality ratio (SMR) 187) Glasgow Govan (SMR 172) ranked 4^th and 8^th worst areas in the United Kingdom.⁵ 30% of men aged 16-64 in Glasgow Govan, and 38.3% in Glasgow Pollok, are permanently sick and inactive or unemployed compared with the overall British figure of 19.6%. 30% of the population in Glasgow Govan and 37% in Glasgow Pollok were among the poorest 10 % in Britain.⁶

All arrivals at the department by 999 ambulance were identified prospectively. “GP Urgent” calls (initiated by a general practitioner) were excluded. Additional data came from A&E and in-patient records.

314 patients were identified over a three-week period. Complete data was available for 290 patients (92%). All but one of the patients with missing data were either admitted to hospital or referred to an outpatient clinic. 126 (40.1%) of patients arrived between 9am and 5pm, 127 (40.4%) between 5pm and 1am and 61 (19.4%) between 1am and 9am. 212 patients (73.1%) had one or more investigations carried out in the A&E Dept (figure 1). 32 (11.0%) had a dressing, 118 (40.7%) had drugs prescribed, 27 (9.3%) had sutures and 4 (1.4%) had a plaster cast applied. Discharge disposition for all 314 patients is shown in table 1.

Disposal	N	%
Referred Medical Specialty	112	35.7
Referred Orthopaedics	40	12.7
Referred Surgical Specialty	23	7.3
Referred Psychiatry	3	1
Referred Obs Gyn	1	0.3
Transferred to another Hospital	6	1.9
Discharged to Return Clinic	16	5.1
Discharged Without Follow-up	95	30.3
Irregular Discharge	16	5.1
Died	2	0.6

Patients referred to surgical specialties included referrals to General Surgery (10), Neurosurgery (9), Urology (2), Ophthalmology (1), and Plastic Surgery (1). Those referred to medical specialties comprised 108 General Medicine and 4 Acute Stroke Unit referrals. Discharges to Return Clinics included referrals to the Fracture (6), A&E (4), Hand Injuries (2), ENT (1), Diabetic (1), Knee Injuries (1) and Surgical (1) Clinics.

95 (30.3%) patients were discharged from the department with no follow up, 23 (7.9%) of these having required no investigations or treatment.

This study demonstrates that almost all patients who arrive at hospital by 999 ambulance attend appropriately.

Although 35.4% of patients were discharged without follow-up, this does not equate with inappropriate attendance as X-rays, the majority of blood tests and, in our area, 12-lead ECGs are not practicable for pre-hospital use. 23 (7.3%) patients were discharged without follow up having had no investigation or treatment beyond history-taking and examination. Arguably, these patients were transported to hospital inappropriately.^7,8 However, in Gardner’s study, 29 of 64 patients in this group had a potentially life-threatening conditionsuch as angina or epilepsy warranting an ambulance call.⁹

There would therefore appear to be a group of patients, particularly in the deprived, urban setting of this study, whose main need is for transportation. A universal response by a fully equipped accident and emergency ambulance and a paramedic crew represents an inappropriate use of a scarce resource.

The NHS already provides transport for non-acute hospital services. Our local out-of-hours primary care service employs such ambulances to transport patients. No equivalent service exists for patients in need of acute care, resulting in many of these patients calling 999. Telephone advice to these callers may not influence the need for attendance and therefore ambulance transportation. Also it should be noted that the majority of 999 calls are not actually made by the patient.⁹

We suggest that if priority-based dispatch can identify patients for whom urgent treatment and immediate transport is not necessary, these patients may be better served by being transported by alternative means, freeing up emergency ambulances, and reducing their response times.

Further work is needed firstly to separate those who genuinely need a 999 ambulance from those who require transport to hospital for legitimate non-emergency reasons and secondly, to examine alternative methods of transport or alternatives to hospital attendance.

2. Wilson, Edwards, Cooke. Inappropriate ambulance use is a retrospective diagnosis. EMJ 1999; 16: 75.

3. Snooks H, Wrigley H, George S et al. Appropriateness of use of emergency ambulances. EMJ 1998; 15: 212-215.

4. Lowey, Kohler Nicholl. Attendance at Accident and Emergency departments: unnecessary or inappropriate? EMJ Medicine 1994 16(2):134-40

6. “The Widening Gap” Shaw, Dorling, Gordon et al, The Policy Press November 1999.

7.Pennycook AG, Makower RM, Morrison WG. Use of the emergency ambulance service to an inner city Accident and Emergency Department – a comparison of general practitioner and “999” calls. J R Soc Med 1991; 84: 726-727.

8. Morris D L, Cross A B. Is the emergency ambulance service abused. BMJ 1980; 3: 121-3.

9.Gardner G J. The use and abuse of the emergency ambulance service: some of the factors affecting the decision whether to call an emergency ambulance. EMJ 1990; 7, 81-89.