1. Introduction
Medical overuse is characterized by healthcare interventions that provide no overall benefit to patients while being associated with significant costs and potential harm. Despite improvements in diagnostics, as well as diagnostic tools and protocols, pediatric patients are highly affected by medical overuse [
1,
2]. Unlike adults, children are more vulnerable, exhibit fewer specific symptoms, and experience rapid changes in disease progression [
3]. Other factors, such as the reduction of uncertainty or expectation from the patient’s family, are associated with overdiagnosis and overtreatment in pediatric patients [
4]. Those factors contribute to unnecessary care, resource utilization, and, in some cases, may even harm the patient [
5].
Upper respiratory tract infections (URTIs) constitute a significant portion of non-urgent visits to the pediatric emergency department (PED) [
6]. Since most URTIs are viral and self-limiting, advanced testing or specific treatment is generally unnecessary [
7]. Patients presenting to PED with URTIs may exhibit a broad spectrum of symptoms that overlap with other diseases. While most viral infections manifest with mild symptoms, the presentation varies among individuals [
8,
9]. The literature suggests that most physicians may fear missing serious conditions, thus leading to excessive testing and treatment [
1]. Harmless intent has been proven to result in harmful outcomes, including a prolonged disease course, more serious issues, and the labeling of chronic diseases [
10]. In contrast, presenting to PED too early, or in cases of repeated visits, places URTI patients at risk of undergoing excessive examination, hospitalization, and inappropriate antibiotic use [
11].
In this study, we aimed to elucidate the frequency of laboratory tests, such as complete blood count (CBC), C-reactive protein (CRP), urinalysis, chest X-ray, and other diagnostic tests performed in our PED for children diagnosed with URTI. Additionally, we aimed to investigate the factors influencing a physician’s decision to perform diagnostic tests and the factors affecting outcomes (discharge, hospitalization, antibiotic prescription, etc.) for children diagnosed with URTIs at PED.
2. Materials and Methods
2.1. Study Design and Study Population
A single-center, cross-sectional study was conducted in the pediatric emergency department of the Lithuanian University of Health Sciences Kaunas Clinic’s tertiary hospital. This hospital typically sees approximately 40,000 pediatric visits annually. Data from all the patients diagnosed with upper respiratory tract infections between 1 December 2021 and 31 January 2022 were retrieved from the local electronic healthcare data record system. Newborns, children with chronic diseases, congenital pathologies, psychiatric disorders, and repeated visits were excluded from the study.
2.2. Data Collection
Further data were collected, including demographics (age and gender), triage categories, time of arrival from the onset of fever, symptoms, auscultation findings, and physiological parameters, such as respiratory rate (RR), body temperature (t°), heart rate (HR), and blood oxygen saturation (SpO2). Data were also collected from conducted laboratory tests, such as complete blood count (CBC), C-reactive protein (CRP), urinalysis, blood, tonsil, cerebrospinal fluid cultures (if any), and the results of rapid strep test (if performed). Instrumental tests, such as chest X-ray, abdominal, and/or chest ultrasound were also documented. Final diagnoses included acute upper respiratory tract viral infection, such as nasopharyngitis or pharyngitis, tonsillitis, and laryngitis. Outcomes were classified as discharge, short stay (up to 24 h), hospitalization, and treatment details, including antibiotics and others.
Additionally, tachycardia and tachypnoea were evaluated in accordance with the European Resuscitation Council Guidelines 2021: Paediatric Life Support [
12]. Complete blood count reference values were determined based on Pediatric Hematology edited by Robert Arceci et al. [
13]. Patient groups were categorized by age, diagnosis, triage category, and the origin of the infection (bacterial and non-bacterial (i.e., viral and non-identified)). Unspecified etiology infections were excluded from the further analysis of the antibiotic prescription rates. Age groups were defined as <6 months old, 6–11 months old, 1–4 years old, 5–11 years old, and 12–18 years old. The origin of the infection was reassessed based on clinical presentation and investigations (if tests were performed). The triage categories were assigned according to vital signs, primary complaint, pain level, severity of the condition, and medical history upon arrival to PED, all in accordance with the Manchester Triage System [
14]. Category number was allocated based on urgency, with the first category indicating the highest level of urgency and the fourth signifying the lowest.
2.3. Statistical Analysis
Data analysis was conducted using Microsoft Excel and IBM SPSS Statistics, version 29.0 software (SPSS Inc., Chicago, IL, USA) for Windows. The Shapiro–Wilk test was used to assess for a normal distribution. The non-Gaussian distributed data were described by the median and interquartile range (IQR). The descriptive statistics of the qualitative variables are presented in absolute numbers (n) and percentages (%). A chi-square (χ2) test or a Fisher’s exact test were conducted to assess the statistical significance of the observed association. A multivariant logistic regression was performed to calculate the relationship between the odds ratio and the coefficient. Spearman’s correlation coefficient was used to evaluate the statistical strength of the association between variables. A p-value of <0.05 was considered significant.
2.4. Ethical Consent
Permission to conduct the study was obtained from Kaunas Regional Biomedical Research Ethics Committee (BE-2–27). This study was conducted in accordance with the guidelines detailed in the Declaration of Helsinki and Good Clinical Practice Guidelines.
4. Discussion
Upper respiratory tract infections (URTIs) represent a primary driver of unnecessary medical visits, thus leading to the overuse of diagnostic tests and inappropriate management decisions, such as hospitalization or antibiotic prescriptions. Our study revealed that the majority of URTI cases were non-urgent, viral, and often subjected to unnecessary testing. Clinician decisions to perform laboratory tests were influenced by the duration of the fever and triage categories in the PED. Common unnecessary diagnostic procedures included CBC, CRP, and urine tests. In addition, radiological testing, such as chest X-ray and abdominal ultrasound, was performed on pediatric patients with URTIs, despite these medical tests typically providing no diagnostic value in such cases as none presented urgent or life-threatening conditions.
Studies have indicated that 20–65% of PED visits are likely to be non-urgent or low acuity cases, thus contributing to emergency department overcrowding, which adversely affects the quality of care and increases healthcare system costs [
15,
16,
17,
18,
19]. In our study, we observed that most URTI cases were triaged as a category 4, indicating non-urgent visits. While we did not measure overcrowding directly, when considering daily statistics, annual visits, staffing level, and referrals to pediatric emergency department, we believe these cases could potentially contribute to the overcrowding of pediatric emergency departments. Parental concerns contribute to the increasing use of emergency departments, thus making the pediatric population the main users of emergency department facilities [
15,
20,
21,
22,
23,
24]. Addressing this issue necessitates parental education, improved availability of primary care, and healthcare provider training in effective communication to reduce unnecessary PED visits.
Clinical decisions regarding URTI etiology and treatment should typically rely on patient symptoms, but high-income settings often respond to laboratory investigations [
25]. Our study found that over two thirds of patients with URTIs underwent CBC and CRP to determine the origin of infection. In our study, it was observed that even 26.7% of laryngitis cases had CBC and CRP tests performed, despite diagnosis traditionally being reliant solely on clinical signs. Similar tendencies were noted in tonsillitis cases within our study, wherein 97% cases underwent CBC and CRP tests, even though it was contrary to recommendations advocating for a preference of rapid tests [
26]. Excessive laboratory testing, ranging from 36–69% in various regions [
27,
28,
29], is often attributed to physician experience, cognitive bias, parental concerns, and a perceived risk of a rapid deterioration in a child’s condition [
28,
30,
31].
Furthermore, our study identified urinalysis as the third most common laboratory test among patients with URTIs. Urinary tract infections (UTIs) pose a challenge, as young children with UTIs often present with non-specific symptoms shared with a wide range of diseases [
32], including URTIs. In our study, 110 (26.8%) patients had a urine test performed, and none of the test results showed pathological findings. Interestingly, three urine cultures were obtained, and all of them were evaluated as normal or contaminated. Our findings indicated that age, gender, triage category, and duration of fever are not associated with the incidence of urinalysis, thus suggesting that urinalysis are performed against the guidelines and latest recommendations on urinary tract infection diagnostics [
33,
34,
35].
Using a rapid strep test alongside the Centor Criteria for diagnosing group A beta-hemolytic
Streptococcus pharyngitis can prevent unnecessary antibiotic prescriptions [
36], thus proving convenient and acceptable to both patients and clinicians while preventing economic damage [
37]. Despite recommendations advocating the avoidance of blood test performance, our study revealed that merely 19.7% of tonsillitis cases underwent a rapid strep test. Additionally, none of the physicians documented the results of the Centor score criteria (data not shown), especially in those 17 cases (4.2%) that underwent a rapid strep test. In contrast to our findings, studies performed in European countries reported high rapid test performance rates in patients diagnosed with pharyngotonsillitis [
38,
39]. Employing rapid antigen tests for group A beta-hemolytic
Streptococcus pharyngitis is a cost-effective method; it enables physicians to decide on a treatment plan in one appointment, and it could be used in each PED.
We observed that the overall antibiotic prescription rate was 20.1%, which is comparable to the rates in other European countries [
40,
41]. Overall, 4.9% of the antibiotic prescriptions were associated with viral diseases, thereby representing unnecessary cases, and these were found to be at a much higher rate than reported in other studies [
42]. Inappropriate antibiotic prescriptions do not improve patient outcomes but may increase antimicrobial resistance, thus threatening our ability to treat infectious diseases successfully in the future. Persistently high rates of inappropriate antibiotic prescriptions likely result from multifactorial factors, including patient sociodemographic data, signs and symptoms at presentation, patient expectations, and even individual factors of a physician [
41].
There is a need to establish local guidelines that emphasize the judicious use of laboratory investigations to effectively utilize limited healthcare resources in our country. Cautious use of laboratory investigations, radiological imaging, and antimicrobials will help reduce hospital management costs.
Certainly, our study has limitations. First, in data records, some physicians did not define the origin of the infection, thus leaving 15.1% (n = 62) of cases unclear with respect to whether a physician suspected bacterial or viral infection. Therefore, the antibiotic prescription rates in the non-bacterial infection group could be identified as too high and impact our conclusion of inappropriate antibiotic use in our setting. Second, our data reflect medication mentions and not antibiotics consumed, and the information if the parents or general practitioners (GP) discontinued antibiotics was limited. We did not analyze if those patients who did not receive antibiotics in PED were further prescribed any at the GP’s office as we wanted to clarify the situation in PED. Moreover, it is important to note that our study period coincided with the COVID-19 pandemic, which could potentially influence our findings. Given the broad spectrum of pathogenesis associated with COVID-19, which affects various organ systems beyond the respiratory system, individuals with a main diagnosis of COVID-19 disease were not included in this study. Consequently, certain data pertaining to supplementary laboratory investigations, instrumental tests, and rates of unnecessary antibiotic prescriptions were not captured, thus potentially resulting in a slightly skewed assessment of medical overuse within our study.