Article Text


Medical problems presenting to paediatric emergency departments: 10 years on
  1. Rebecca Sands1,
  2. Dhurgshaarna Shanmugavadivel2,
  3. Terence Stephenson2,
  4. Damian Wood1
  1. 1Nottingham Children's Hospital, Nottingham University Hospitals NHS Trust, Queen's Medical Centre Campus, Nottingham, UK
  2. 2Faculty of Medicine and Health Sciences, University of Nottingham Medical School, Queen's Medical Centre, Nottingham, UK
  1. Correspondence to Dr Dhurgshaarna Shanmugavadivel, Nottingham Childrens Hospital Ward E17, Queen's Medical Centre, Derby Road, Nottingham NG7 2UH, UK; shaarnashan{at}


Objective To describe the common medical presenting problems of children attending a paediatric emergency department (ED) compared with 10 years previously.

Design A retrospective review of electronic patient record and comparison with previous cohort.

Setting A UK university hospital ED.

Patients A cohort of children and young people aged 0–15 years who attended the ED between 7 February 2007 and 6 February 2008 (n=39 394) compared with a historical cohort from 10 years earlier.

Main outcome measures and results Information on presenting problem, demographic data and source of referral were collected. Presenting problems were ranked and comparisons made with previous data using the difference between proportions analysis and the significance test for a difference in two proportions. A total of 39 394 children (57% boys) were seen with 14 724 medical attendances compared with 10 369 attendances from the 1997 cohort, an increase of 42%. Most (85%) ED attendances can be accounted for by the 10 most common presenting problems, including breathing difficulty (2494, 20.1%), febrile illness (1752, 14.1%), diarrhoea with or without vomiting (1731, 14.0%), rash (1066, 8.6%) and cough (835, 6.7%). Similar proportions are described to a decade earlier; however, there were fewer patients attending with breathing difficulty (−10.9%, p<0.001).

Conclusions Over a 10-year period, there has been a rise in the number of people attending the ED with medical conditions. The 10 most common presenting problems account for 85% of medical attendees. These results suggest the increasing utilisation of ED services for children with common medical presenting problems and should inform further research exploring the pathways for attendance and the thresholds in seeking medical advice in order to inform the commissioning of paediatric emergency and urgent care services.

Statistics from


Over a 10-year period, the number of child admissions to hospital has increased by 18%; the majority can be attributed to an increase in admissions via the emergency department (ED).1 The number of children being admitted via EDs across England is now over 350 000 per annum, and this represents an increase of 6.8% over the period of 2005–2008.2

Armon et al identified that, in order to plan services and develop training programmes and guidelines, high-quality data on paediatric attendances are required, and that these should be collected in a standardised and complete manner.3 They carried out the first study in the UK looking at the most common paediatric presentations to an ED, choosing to describe the presenting complaint rather than the final diagnosis with the aim of developing management guidelines for doctors. They concluded that, for children with medical complaints, six presenting problems covered 83% of presenting problems: breathing problems, feverish illness, diarrhoea and/or vomiting, abdominal pain, seizure and rash. These data informed the development of a number of presenting problem-based guidelines, which were derived from best evidence and have been shown to improve the quality of clinical care.4–6

The past decade has seen significant changes in the NHS and in particular changes in the way people access out-of-hours and emergency healthcare. The recent publication of the Emergency and Urgent Care Pathway for Children and Young People focuses on improving quality of clinical care within high-volume pathways.2 While it is clear that attendances have increased since the original study, what is unclear is whether the nature of the problems that present to paediatric EDs has also changed. A Medline search, hand searching of references, and personal contact with experts in this area revealed no other subsequent studies of a similar nature to provide an answer to this question.

We aim to describe the presentations to a paediatric ED in a busy general hospital and compare it with the data published by Armon et al.3

Patients and methods

The study was undertaken in the ED at the Queens Medical Centre, Nottingham, which provides a service to the population of Nottingham City and surrounding localities. This consists of an approximate population of 592 000, of which ∼108 000 (18.2%) are children aged 0–14 year.7 The ED at Queens Medical Centre currently sees ∼159 000 cases per year. The children and young people are seen in a separate paediatric area.

All acute paediatric attendees, whether self, general practitioner (GP) or other referral source, are initially triaged in the ED. Patients were identified retrospectively using the electronic patient record and clinical information system—EDIS, iSOFT. In the 1997–1998 cohort studied by Armon et al, the authors used prospective data collection using the electronic patient record system, PAS, and paper proformas. In the 2007–2008 cohort, information on demographics, time and source of referral was inputted by reception staff along with the presenting complaint based on report by parent or patient. Information on diagnosis and disposal was inputted by ED medical and nursing staff. We obtained data on all patients aged between 0 and 15 years (inclusive) who had attended between 00:00 on 7 February 2007 and 23:59 on 6 February 2008. Patients were classified using a combination of presenting problem, diagnosis and disposal. We defined the children and young people attending with medical problems by excluding from the dataset those patients presenting with trauma and/or problems of a surgical, obstetric or gynaecological nature. Patients presenting with self-harm and non-accidental injury were included with the medical attendees. Direct referrals by GPs to the paediatric medical team were also excluded. The remaining cohort of undifferentiated medical attendees to the ED was then further analysed. This ED medical group were categorised by age, referral source and presenting problem. Where presenting problem was unclear from the EDIS system, the diagnosis was reviewed. If it was still unclear to which presenting problem category the patient should be allocated, then the full ED patient record was examined.

Descriptive statistics were applied to the data collected and included measurements of proportion and rank. Differences between proportions were analysed for statistical significance using the SE of the difference between two percentages and the significance test for a difference in two proportions. The accepted p value for significance was <0.05.


A total of 39 394 children (57% boys) aged 0–15 years were seen in the ED during the study year compared with 38 982 children (58% boys) 10 years ago. Of these, 14 724 (37%) were classified as medical compared with 10 369 described by Armon et al, which represents an overall increase of 42%. From the 2007–2008 cohort, 2335 (15.8%) of medical attendees at the ED were referred directly to the paediatric medical team by the GP and were therefore excluded from further analysis. The remaining undifferentiated medical group comprised 12 389 attendances by a total of 9541 children. The preschool age group of 0–4 years accounted for 69.8% of the attendances. The age distribution of the attendees is shown in table 1.

Table 1

Age distribution of medical attendees

Presenting problems could be categorised into 20 presenting complaints; this included one category, described as ‘Other’, which comprises a variety of presenting problems each with a frequency of less than 30 cases per year. After excluding the ‘Other’ category, 84.6% medical attendances to the ED can be accounted for by the 10 most common presenting problems, with the top three comprising 48.2% of the attendances. Table 2 shows the numbers and rank order for each presenting problem.

Table 2

Presenting problems of medical attendees aged 0–15 years

When compared with the data published by Armon et al, the ranking remains constant between the two studies, once the ‘Other’ category has been excluded. Table 3 shows the comparison of presenting problems with the data of Armon et al. In our study, breathing difficulty remains the most common presenting problem, accounting for 20.1% of the attendances as compared with 31% in 1997. This equates to a significant difference in proportion (10.9%; p<0.001; 95% CI 9.24% to 12.5%). The difference in proportion of each of the presenting problems between the two time periods can be seen in table 4.

Table 3

Frequency and rank of presenting problem in 2007 and 1997

Table 4

Difference in proportion of presenting problems in 2007 versus 1997

Of the medical patients seen in the ED, 8855 (71.5%) were self or parent/guardian referred. Of these, 1789 (14.5%) had been seen, or advised, by a healthcare practitioner before attending the ED. It is not clear how many of the ambulance group, which comprised 1315 (10.6%), had sought other advice before calling for assistance. Of the total number of medical patients seen in the study year, 3053 (20.7%) were advised to attend the hospital by their GP. A small proportion had been recommended to attend or brought by another agency (eg, school, police/prison or school). The sources of referral for attendees are presented in table 5.

Table 5

Source of referrals of medical attendees


Up to half of infants aged under 12 months and one-quarter of older children will attend an ED annually.8 We describe paediatric medical attendances at a large university hospital ED over a 12-month period and compare them with similar data from 10 years ago. Our data may not be applicable to all EDs, but there are some striking similarities between our data and national statistics on paediatric emergency attendances. The initial results of Armon et al3 were comparable to those of UK and overseas studies, and we could find no further published studies of a similar nature in the intervening period. This study allows direct comparison of attendance data with those collected 10 years previously. There were some differences in the methodology of the two studies. Armon et al used a combination of electronic patient database (Patient Administration System or PAS) and a data entry form completed by medical or nursing staff at the time of patient contact. In our study, the use of the EDIS (iSOFT) database allowed us to examine an entire year's data rather than a small sample. While our data analysis was retrospective, the data were captured prospectively by the ED team. Data were inputted to EDIS by both non-medical and medical staff. Demographic data and presenting problems were often recorded by non-medical staff; this has drawbacks and advantages. The presenting problem as recorded by the non-medical staff may be a better reflection of the initial complaint as described by the parent or child, but may differ from the presenting problem recorded in the clinical notes, which is often a re-interpretation by the clinician following triage.

Our results suggest that, in the 10 years since Armon et al first described the presenting problems of children attending a UK paediatric ED, there has been an increase in the number and proportion of children attending with medical problems. While the overall number of children attending the ED with all conditions has remained fairly similar (39 394 vs 38 982), we show a 42% increase in medical attendees. There are a number of possible explanations, which this study is not powered or designed to answer. It may be that this is a true reflection of disease epidemiology, with trauma and surgical presentations reducing through prevention strategies. While an increase in the proportion of medical presentations may reflect better recognition of potentially serious medical symptoms by parents and other healthcare professionals, such as the meningitis awareness campaign,9 it could also be due to a change in healthcare-seeking behaviour, with parents more likely to seek urgent paediatric care via the ED rather than from primary care. This may be a consequence of organisational changes within the NHS including NHS Direct, the GP contract and reduction in out-of-hours service provided by the patient's usual family doctor, and the development of walk-in centres.10 We do not have information on the proportion of medical patients referred by GPs from the 1997 dataset; hence it is not possible to determine whether any of the increase in medical attendances is due to a rise in referrals from primary care. Interestingly, recent national data from the Department of Health report a decrease of 16% in the proportion of emergency admissions via GP surgeries over the past decade.1

We describe a broadly similar range of presenting complaints to Armon et al, with the most frequent presenting problems being ranked in a similar order with similar proportions. One notable difference is the reduction in the number and proportion of children presenting with breathing difficulties. Although it remains the most common presenting complaint, the proportion has fallen from 31% in 1997 to 20% in 2007. This may reflect the downward trend seen internationally in acute asthma presentations.11–13 An alternative explanation is that differences in methodology and categorisation between the two studies resulted in a number of patients who would have been categorised as having breathing difficulty being recategorised as having a cough or another complaint. Even if these two groups (breathing difficulty and cough) are combined, there remains a difference of ∼5%, which, although smaller, remains statistically significant.

Further substantial reform of health services for England is now underway, and a key principle of the reforms is ‘the service must be focused on outcomes and the quality standards that deliver them’.14 In his recent report Getting it Right for Children and Young People,15 Sir Ian Kennedy notes that ‘Accident and emergency (A&E) has become the default option for the treatment of children and young people’, and our data would support this. He also recommends that ‘those commissioning health services for children and young people should (sic) require providers to design services around the needs of children or young people, establish a single portal of access, ensure that care is delivered in line with the normal pathway of care, and require the collection, analysis and dissemination of information’. Focusing on a limited range of high-volume pathways is a key concept in making the maximum impact on improving the quality and value of care for NHS patients as recognised in the Department of Health document Delivering Quality and Value. Focus on: Children and Young People Emergency and Urgent Care Pathway.2 We have shown that, at a time when there is a drive to keep children and young people out of hospital, the number of children attending the ED with medical problems continues to rise. Children and young people are attending with the same problems as 10 years ago, so further research is needed to explore the pathways for attendance and the thresholds and behaviour of parents in seeking medical advice, as this is the information that will inform the commissioning of services based on the needs of children, young people and their families.

In addition, undergraduates and postgraduate trainees new to paediatrics often feel overwhelmed by the breadth of conditions that they may encounter during their attachments.16 Sir Ian Kennedy also acknowledged the limited training for GPs in child health and the impact on ED attendances.15 The knowledge that 85% of medical attendees can be categorised into the 10 most commonly presenting problems helps the trainee or student focus their learning.


Over a 10-year period, attendances to the paediatric ED have remained similar; however, there has been a disproportionate rise in the number attending with medical conditions. The presenting problems also remain similar, although there has been a significant reduction in those presenting with difficulty in breathing. We describe the 10 most common presenting problems, which account for 85% of medical attendees. Care pathways focusing on this limited range of high-volume presenting problems will make the maximum impact on improving quality and value of care for children. Further research exploring the pathways for attendance and the thresholds in seeking medical advice is needed to inform the commissioning of paediatric emergency and urgent care services.

What is already known on this subject

  • High-quality data on paediatric attendances is required in order to develop guidelines and training programmes.

  • 83% of paediatric admissions 10 years ago were due to six main medical complaints including breathing problems, fever, diarrhoea and vomiting, abdominal pain, seizure and rash.

What this study adds

  • This study shows a 42% rise in medical emergency department attendees over the last 10 years.

  • The most frequently presenting medical complaints are the same, although there has been a 10.9% reduction in those presenting with breathing difficulties.

  • Understanding frequent presenting problems can inform commissioning and pathway provision and undergraduate and postgraduate medical education.


View Abstract


  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.