Patients and methods

The study was conducted in the paediatric accident and emergency department, located within the adult department (which sees approximately 120 000 cases per year) at the Queen's Medical Centre, Nottingham. This is a 1300 bed general and tertiary referral hospital that serves the whole of the population of the city of Nottingham and the surrounding area (approximately 745 000, of which children under 15 make up approximately 136 000). All acute child attenders aged 0–15 years, whether self or general practitioner referred, are seen within the paediatric A&E department. The department is staffed by one and a half whole time equivalent paediatric A&E consultants, and all permanent nursing staff are registered children's nurses. On average, half of the senior house officers (SHOs) working in the department have had previous paediatric experience.

Data on all 0–15 year olds attending paediatric A&E were collected prospectively over a one year period from 7 February 1997 to 6 February 1998. Demographic details at presentation were recorded on the patient administration system (PAS). Clinical details were recorded by clinicians on a standard A&E sheet and entered onto the PAS at time of discharge. Nursing staff and senior house officers were asked to complete an additional form on all medical attenders, detailing the presenting problem and grade of the most senior doctor involved with the case. These data were merged with the PAS data at the end of the study. In order to check the validity of these additional data, 16% of the forms were checked against the medical record. Univariate analysis was conducted using non-parametric techniques, χ² for categorical data and Mann–Whitney U for continuous data, as these were not normally distributed.

Results

A total of 38 982 children (58% boys) aged 0–15 years were seen during the study year. The trauma/surgical group numbered 26 756 (69%) (subsequently referred to as the “trauma” group); categories included were: accident, assault, surgical, orthopaedic, and obstetrics and gynaecology. The “medical” group numbered 10 369 (27%) and included children classified as “self inflicted”, non-accidental injury, psychiatric, and medical. Of the remaining children, 356 did not wait to see the doctor, 185 were classified as “other”, and in 1316 (3.4%) data on classification were not completed. These last three unclassified groups were excluded from subsequent analysis.

Discussion

This paper gives the first detailed description of presenting problems (rather than diagnoses) of children attending a UK paediatric A&E department. Children with medical complaints made up 27% of the total attendances and the majority of these attended once only during the year (83%). A total of 83% of medical attenders were in one of six categories of the commonest presenting problems.

The additional forms for data collection on medical cases were completed for only one third of medical attenders, which we speculate was because SHOs were too busy. The subset of medical cases on which further data were completed tended to be younger than the whole medical group (by approximately 10 months), which may influence the results towards presenting problems more prevalent in this age group.

Data collected in a Canadian emergency room on paediatric diagnoses13 showed 34% “respiratory”, 15% “otitis media”, 14% “gastroenteritis”, 7% “abdominal pain”, 8% “rash”, 5% “fever”, and 1.6% “seizure”. These show a similar spectrum of disorder to our data but differ in the inclusion of otitis media, as this is a discharge diagnosis. Data from a USA emergency room15 also found a similar range of presenting problems but in differing proportions: febrile illness in 21%, respiratory distress in 12%, vomiting in 10%, abdominal pain in 7%, rash in 6%, and seizures in 2%. Classifications used and a different health care system offer possible interpretations.

Similar presenting problems were found in paediatric admission data from five Yorkshire hospitals,17 though in different proportions. In that study admissions for diarrhoea and vomiting were 9%, for abdominal pain 3%, and for seizure 16%. This is in contrast to the percentages presenting to A&E in our study (table 1) where a higher proportion of children appear to attend for relatively minor conditions than are admitted.

In Nottingham, 78% of children (the majority with minor illness) were seen by an SHO alone. After correction for error, this would be about 75–76%; registrars, consultants, and clinical assistants were involved in 20%, 2%, and 2% of cases respectively. Consultant workload within this department is primarily in the A&E follow up clinics (not included in this data set), and for resuscitation calls, administration, and teaching. SHOs are therefore exposed to many children with common acute illness, and presenting problem based guidelines would provide a framework in which they can consolidate their experience and from which they can learn. Six such guidelines would cover over 80% of presentations.

These data are from one A&E department and may not be totally generalisable to all units. According to a survey conducted by the Children in A&E Special Interest Group in 1997,18 of 204 questionnaires returned from the 268 A&E departments in the UK, 15 reported separate children's departments (nine being within a children's hospital), and 142 reported a separate area for children within the all age department. In Nottingham, all general practitioner and self referrals are seen within this unit and many of the staff have specialist training in paediatrics; however, it is situated close to the city centre and is attended by children with higher deprivation scores than for the Nottingham Health District population,11 in keeping with adult A&E attendance studies.7 ,8 As such, it is not comparable to a “children's hospital” where the proportion of medical patients among attenders may be higher.

Trainees manage many acutely ill children in NHS hospital practice at present. In our A&E the complete episode of care for many was provided solely by SHOs. The NHS intends that in future the service should be delivered by fully trained medical and dental staff,19 but it will still be important to offer experience for trainees in clinical assessment and decision making aided by guidelines. As only one third of children with medical problems attending A&E are admitted,20 and only one quarter have investigations undertaken (Armon, unpublished data), many children could be managed in the community or by primary care teams where the ratio of fully trained doctors to trainees is much higher than in hospital. We found that a higher proportion (47%) of children referred by general practitioners were admitted than children seen after self referral (25%), suggesting a degree of clinical selection prior to attendance. Some children (with minor self limiting illness or an established diagnosis) may not need to see a doctor at all and, potentially, could be managed or triaged by a telephone advice service such as NHS Direct, community pharmacists, or emergency nurse practitioners.19 ,21