We analyzed the insurance claims records of the Korean NHI claims database from January 2010 to December 2014. Through this retrospective population-based study, we estimated the prevalence of asthma in children in Korea and investigated their asthma-related health care utilization, prescription and test patterns. Asthmatic patients in this study were defined as children who were younger than 18 years who visited hospital with an asthma code of J45 or J46 for the principal or within 5 additional diagnoses based on the 10th Revision of the International Classification of Diseases. Additionally, they required a prescription for asthma maintenance medication such as inhaled corticosteroid (ICS), ICS/long-acting beta2-agonist (LABA) combination inhaler or leukotriene receptor antagonist (LTRA) on the same date. Population data by age with health insurance were used to estimate the prevalence of asthma by age. In terms of hospitalization and emergency department (ED) visits for asthma, asthmatic patients were defined as the use of short-acting beta₂-agonist (SABA) during the hospital visit under the asthma definition in the current study. This study was approved by the Institutional Review Board of Severance Hospital (protocol no. 4-2015-0891). The requirement for informed consent was waived due to the retrospective nature of the study.

Asthma-related medications included SABA, ICS, ICS/LABA combination inhaler, LTRA, aminophylline/theophylline and systemic steroids (Supplementary Table S1). Analysis of drug prescribing frequency in asthmatic patients only included outpatient prescriptions. Asthma-related tests included spirometry with/without bronchodilator and allergy tests such as multiple-allergen simultaneous test, serum total or allergen-specific Immunoglobulin E levels and skin prick test.

In a prevalence and cost-of-illness study, the cost of an illness in a population is measured over a defined period, usually a year.14 The economic costs of a disease can be classified as direct, indirect or intangible costs. Direct costs comprise all direct medical cost, of which include expenditures for diagnosis, treatment, continuing care, rehabilitation, terminal care for an illness15 and direct non-medical costs. Direct medical costs are defined as formal expenses for the diagnosis and treatment of asthma within the official health care system and consist of inpatient and outpatient care costs. In the current study, asthma-related cost was defined as the sum of the direct medical costs, which consisted of asthma-related hospitalizations, outpatient visits, ED visits and the costs of prescribed medications. The outpatient visits were divided into primary hospital, secondary hospital and tertiary hospital visits. Expenses not covered by insurance were excluded in the calculation. All costs were calculated in KRW then converted into USD using the exchange rate of 1 USD at 1,107 KRW, which is the average exchange rate between 2010 and 2014.

The number of asthmatic children (from 1 month to 18 years old) on record was 1,172,807 in 2010 and increased steadily to 1,590,228 in 2014 in Korea (Table 1). In 2014, 16,110 (male 16,720; female 15,010) patients per 100,000 children had asthma compared to 10,780 (male 11,370; female 9,930) patients per 100,000 in 2010 (Table 1, Supplementary Table S2). The annual trend of increasing prevalence was noted for all ages (Fig. 1A). Asthma prevalence decreased with increasing age and the peak prevalence was mostly noted in the age group 2–4 years (33.1%–41.5%) (Fig. 1B, Table 1).

The distribution of asthma-related health service use according to age during the study period was investigated (Fig. 2, Supplementary Table S3). The proportions of outpatient clinic visit among the types of healthcare facilities increased with age (Fig. 2A, Supplementary Table S3). The frequency of outpatient clinic visits among children with asthma was highest in the 2–4 years age group, and decreased with age (Fig. 2B, Supplementary Table S3). The proportion and frequency of ED visits decreased with age (Fig. 2, Supplementary Table S3). Annually, approximately 68% of children younger than 2 years visited the ED and about 18% of them were hospitalized at least once due to asthma (Supplementary Table S3). While both proportions decreased with increasing age, the proportion of outpatient clinic visits remained steady in the range 95%–99%, through all age groups (Supplementary Table S3).

Spirometry for pulmonary function test (PFT) was performed in 0.4%–5.7% of asthmatic patients older than 4 years and allergy test was performed in 0.9%–4.4% of them. A higher proportion of those for whom asthma related tests were performed were older children (Fig. 3, Supplementary Table S4).

Fig. 4 shows the prescription patterns of asthma medication according to age in outpatients. LTRAs were the most commonly prescribed asthma medications for all ages (Fig. 4A). After 2 years of age, the majority of asthmatic patients received prescriptions for LTRAs, and this proportion of patients increased with increasing age (Fig. 4B). The overall proportion of patients who were prescribed ICS or ICS/LABA combination inhalers was less than 15%. While the proportion of asthmatic children prescribed ICS decreased by age and fell to 4.2% in the 15–19 year age group, the proportion of patients prescribed ICS/LABA combination inhalers increased to 10% among those aged 18 years (Fig. 4B, Supplementary Table S5). The number of systemic steroid prescriptions declined with age (Fig. 4A, Supplementary Table S6). Systemic steroids were prescribed at least once in 25%-42% of the asthmatic patients and the proportion of patients for whom it was prescribed slightly decreased with age (Fig. 4B, Supplementary Table S5). The proportion of patients for whom xanthine derivatives were prescribed at least once and declined with age (Fig. 4B, Supplementary Table S5). Between 2010 and 2014, the proportion of patients for whom LTRAs were prescribed and the number of LTRA prescriptions per person increased over time for all the ages, while the proportion of patients for whom ICS or ICS/LABA combination inhalers were prescribed decreased (Fig. 4C and D, Supplementary Tables S5 and S6). The proportions of patients for whom xanthine derivatives were prescribed and the number of prescriptions per person decreased gradually during the study period (Fig. 4C and D, Supplementary Tables S5 and S6). There was no significant change in the number of prescribed systemic corticosteroids.

The direct medical costs of asthma during the years 2010–2014 are presented in Supplementary Table S7 and Fig. 5. While there was an increase of 35.6% (from 1,172,807 to 1,590,228) in the number of children with asthma between 2010 and 2014, there was no significant increase in the annual total direct medical costs for asthma or the annual direct medical expenses per person decreased. The total age- and year-specific asthma-related costs ranged from 109 USD (120,747 KRW) in the 15–19 years age group in 2013 to 508 USD (562,747 KRW) in those aged 1 month–2 years in 2010. Over the entire 5 years, the mean direct medical cost per patient decreased with increasing age, from 447 USD (495,173 KRW) in the age group of less-than-2-year-olds to 121 USD (134,040 KRW) in the 15-19 years age group.

The total cost of medications prescribed for patients with asthma younger than 18 years amounted to 105 million USD in 2011, but declined to 70 million USD in 2014. The cost of asthma medication per person each year also decreased from 87 USD in 2011 to 53 USD in 2014. The total direct medical cost for the treatment of pediatric asthma showed a declining trend during the recent 5 years.

In terms of the proportion of composition in total direct medical costs, hospitalizations and ED visits made up more than half of the total cost in the age group younger than 4 years, while medications and hospital outpatient visits constituted the most costs in those older than 12 years (Fig. 5B). The total cost per person in each subcategory showed a decreasing trend with age and the number of ED visits and hospital admissions showed the greatest difference with increasing age (Supplementary Fig. S1).