Common diagnoses among pediatric attendances at emergency departments

Singapore is an urban city with a land area of 728 km² and a population of 5686800 [12]. Persons under the age of 20 years old make up 14.1% of the population. The healthcare system consists of private and public institutions. At the nine public acute hospitals where more than 80% of all hospital beds are located, healthcare for Singapore citizens and permanent residents are provided at a subsidised rate with up to 80% of the costs covered by the government. These public hospitals are grouped into three healthcare clusters based on geographical locations.

This study was carried out in a pediatric tertiary hospital and two adult tertiary hospitals which belong to the same healthcare cluster. The 830-bed pediatric hospital has an ED with an annual census of 142,000. It is supported by inpatient and outpatient pediatric specialties. The two adult hospitals, one 1000- and another 1700-bed hospital, have general EDs with annual census of 157,000 and 165,000 respectively. There are no pediatricians in these EDs and the hospitals have no pediatric inpatient units. The distances between these three hospitals range from 4 to 20 km.

Primary care for pediatric patients is provided at about 1700 clinics run by private general practitioners and 20 polyclinics which are ‘one-stop’ healthcare centers offering primary care services owned by the government. Onward referral to the acute hospitals may be made by the primary care providers for urgent conditions. However, primary care consult is not required before a patient can attend the ED of an acute hospital.

Pre-hospital emergency medical services are provided by a single national provider, the Singapore Civil Defence Force. When activated, patients will be conveyed to the public hospitals. Pediatric patients less than 16 years old will be conveyed to the pediatric EDs. However, if they are critically ill with imminent airway compromise, profound shock or cardiopulmonary arrest, they will be conveyed to the nearest ED for assessment and stabilization, regardless of whether it is a general or pediatric ED. After stabilization, the patients will then be transported to the public pediatric hospitals by either the ED team or the pediatric hospital’s transport team for further management.

In this retrospective study, all patients less than 16 years old who attended the three EDs in the healthcare cluster between 1 January to 31 December 2018 were included. Their electronic medical records were accessed and the following data were extracted: patient demographics, triage category, case type (trauma or non-trauma) and SNOMED Clinical Terms diagnoses codes.

This study was approved by Institutional Review Board at SingHealth, Singapore (CIRB reference 2019/2360), with waiver of informed consent. All methods were carried out in accordance with relevant guidelines and regulations.

There were 159,040 pediatric attendances during the study period, of which 3477 (2.2%) were seen at the general EDs and 155,563 (97.8%) were seen at the pediatric ED (Table 1). The median age of pediatric patients seen in the general EDs (11 years, IQR 5 to 14) was higher than those seen in the pediatric ED (3 years, IQR 1 to 7). There was a greater proportion of emergent (P1) cases at general EDs (N = 233, 6.7%) than the pediatric ED (N = 3821, 2.5%; p < 0.01). Non-traumatic conditions were most prevalent in both general (N = 1933, 55.6%) and pediatric EDs (N = 128,415, 82.5%). However, there was a greater proportion of trauma related conditions at the general EDs (N = 1544, 44.4%) than the pediatric ED (N = 27,148, 17.5%; p < 0.01).

Three out of the five most common diagnoses were non-trauma related diagnoses at the general EDs while all five were non-trauma related diagnoses at the pediatric ED (Table 2). At the general EDs, the five most common diagnoses accounted for 47.2 and 39.4% of the pediatric attendances (Table 3). In comparison, the five most common diagnoses accounted for 44.0% of the attendances at the pediatric ED.

For non-trauma related diagnoses, upper respiratory tract infection, (URTI) unspecified musculoskeletal pain and gastroenteritis were the three most common diagnoses seen at the general EDs (N = 330, 9.5%; N = 177, 5.1%, N = 176, 5.1% respectively), while URTI, unspecified fever and gastroenteritis were the three most common diagnoses seen at the pediatric ED (N = 31,013, 19.9%; N = 11,114, 7.1%; N = 10,151, 6.5% respectively). For trauma related diagnoses, wound, fracture and contusion were the three most common diagnoses seen at the general EDs (N = 419, 12.1%; N = 401, 11.5%; N = 173, 5.0% respectively), while fracture, wound and head injury were the three most common diagnoses seen at the pediatric ED (N = 5756, 3.7%; N = 5350, 3.4%; N = 3776, 2.4% respectively).

Among the critically ill pediatric patients in the emergent (P1) category, febrile (N = 41, 17.6%) and non-febrile (N = 27, 11.6%) seizure disorders, as well as allergic conditions (N = 15, 6.4%) were the three most common emergent (P1) diagnoses at the general EDs, while bronchiolitis (N = 452, 11.8%), asthma (N = 417, 10.9%), and bronchitis (N = 320, 8.4%) were the three most common emergent (P1) diagnoses seen at the pediatric ED (Table 4).

This retrospective study involved pediatric attendances at the EDs over a 1 year period, resulting in a limited sample size and any trends over time in the common diagnoses could not be established. We did not present collective data on the common diagnoses across all EDs in the healthcare cluster as the number of patients at the pediatric ED was significantly larger than the general EDs and this could skew the data leading to misrepresentation. Using patients’ electronic medical records, the diagnoses were captured based on the SNOMED Clinical Terms diagnoses codes entered by the doctor during consultation. The accuracy of the diagnoses was not verified against the clinical details of the consult due to missing information and inconsistencies in documentation by various medical personnel. We did not analyse the initial presenting complaints because these were entered as free text in our hospitals due to the lack of a widely accepted categorization system [23]. We also did not ascertain the reasons behind the observed pattern of ED utilization by paediatric patients. This study was carried out in a single healthcare cluster consisting of only tertiary hospitals linked to academic centres in urban areas. The common diagnoses identified may differ in other hospitals such as community and non-academic centres, and settings such as those with differing population density and without a pediatric ED located nearby. A collaborative effort at an international or national level can be built upon this work to allow for a more in-depth understanding of the common diagnoses among pediatric patients at EDs, and how it varies across healthcare institutions and settings. This knowledge can help reshape and define the role of EDs in its delivery of pediatric care.