Impact of Coinsurance Reduction Policy on Healthcare Utilization Among Children Under 15

The South Korean health insurance system aims to improve public health and social security by increasing access to medical services and guaranteeing opportunities for use by providing risk distribution, income redistribution, equitable cost burden, and appropriate insurance benefits.1 Universal health insurance was enabled in 1989,2 after which there has been a continuous debate regarding the increase in coverage level. Particularly, owing to the low birth rates, various medical coverage policies have been implemented for children to reduce the burden of medical expenses.

On August 9, 2017, a new policy regarding the extension of health insurance benefit coverage was announced.3 It encompassed significantly easing the burden of medical expenses for low-income families and essential medical expenses for economically and socially vulnerable groups such as the elderly, children, and women. Particularly, the policy to ease the burden of hospitalization expenses on children was implemented to secure children’s medical accessibility by minimizing situations of children receiving compulsory education not receiving proper inpatient care for economic reasons. It also aimed to improve the environment for future generations and child-rearing by preventing households from slipping into poverty owing to excessive medical expenses.4 On October 1, 2017, the coinsurance rate for children under the age of 15 was officially reduced from 10–20% to 5%5 under the amendments to subordinate statutes “Enforcement Decree of the National Health Insurance Act” and “Enforcement Decree of the Medical Care Assistance Act.”

Similar policies were established in 2006 that exempted hospitalization copayment for children under the age of six.6 However, many related studies have shown a clear increase in medical use and costs owing to the exemption policy. Additionally, controversies regarding moral hazard and fiscal soundness issues were raised in the process; ultimately, the hospitalization coinsurance rate increased from 0% to 10% in 2008.7

Since the introduction of the current coinsurance reduction policy in 2017, there has been little research on changes in health utilization and medical costs. It is relevant to review the effects and concerns of the aftermath of policy effects to ensure that this policy does not take the same route as the exemption policy. Therefore, this study explores the effect of the coinsurance reduction policy on hospitalization utilization among children under 15.

Table 1 presents the general characteristics of the study population before and after the implementation of the policy. Case group 1 (1–5 years old) included 88,390 participants before (2015–2016) and 76,614 subjects after (2018–2019) policy intervention. Case group 2 (6–10 years old) included 90,992 subjects before (2015–2016) and 89,768 subjects after (2018–2019) policy intervention. Case group 3 (11–15 years old) included 100,633 subjects before (2015–2016) and 91,404 subjects after (2018–2019) policy intervention. The control group (17 years old) included 24,661 subjects before (2015–2016) and 20,482 subjects after the (2018–2019) policy intervention.

Table 1
General characteristics of the study population by before and after the policy

• Click for larger image
• Click for full table
• Download as Excel file

Table 2 presents the results of the changes in healthcare utilization in the study population according to the application of the coinsurance reduction policy. Case group 1 exhibited (1–5 years old) an increase in inpatient cost, inpatient admission, and inpatient length of hospitalization (16.17% for inpatient cost, 8.55% for inpatient admission, and 10.67% for inpatient length of hospitalization). Both outpatient costs and visits and total costs (−9.14% for outpatient costs, −6.79% for outpatient visits) decreased. The results were statistically significant except for the total cost. In case group 2 (6–10 years old), the decrease in outpatient utilization (−9.14% for outpatient cost, −6.79% for outpatient visits) and total cost (−6.56%) were significant. Inpatient utilization differed statistically insignificantly. In case group 3 (11–15 years old), the decrease in outpatient utilization (−9.14% for outpatient cost, −6.79% for outpatient visits) and increase in total cost (6.73%) were significant. Inpatient utilization differed statistically insignificantly.

Table 2
Changes in healthcare utilizations in the study population according to application of coinsurance reduction policy

• Click for larger image
• Click for full table
• Download as Excel file

Table 3 presents the results of the subgroup analyses according to the participants’ income. In the low-income group, inpatient costs and admissions significantly increased in case group 1 (inpatient cost: 45.15%, P value ≤ 0.001; inpatient admission: 17.02%, P value = 0.003). In case group 3, inpatient cost and length of hospitalization increased significantly (inpatient cost: 30.54%, P value = 0.005; inpatient length of hospitalization: 12.14%, P value = 0.045). In case group 2, although there was an increase in inpatient costs and inpatient admissions, it was statistically insignificant.

Table 3
Results of the subgroup analyses according to subjects’ income

• Click for larger image
• Click for full table
• Download as Excel file

This study investigated the effect of a coinsurance reduction policy on healthcare utilization among children under 15. According to the study results, regarding individuals aged 1–5, there was a significant increase in inpatient costs, admissions, and length of hospitalization, while there was a significant decrease in outpatient costs and visits.

This study is consistent with similar research. In Vietnam,15 there was a policy change that allowed children under 6 to have free access to health services in public facilities. Compared with non-beneficiaries, the 4–5 years-old group had better health-related outcomes. In Sweden,16 there was a policy reform in 2002 that abolished copayments for those under 20. The results showed that the number of doctor visits increased by 5–10%.

Regarding those aged 1–5, there seems to be a substitution effect between inpatient and outpatient services owing to the reduction in coinsurance policy. This effect can be noted in similar policies that reduced the coinsurance rate. Lee17 observed the effect of the reduction policy among stomach cancer patients and suggested that patients who could not get hospitalized owing to medical expenses used inpatient services owing to the reduction in the coinsurance rate. Consequently, outpatient services decreased, thus showing a substitute effect. Nevertheless, this study argues that it is not necessary to note the substitute effect as a moral hazard issue. Particularly, low-income families experience a state where their medical needs are not met,18 and the reduction of the coinsurance allows them to obtain the medical care they need. However, this study analyzes the study population, which includes those who did not use inpatient services. If there is a significant increase in the population, the substitute effects should be interpreted with caution.

Regarding those aged 6–10 and 11–15, inpatient utilization showed inconclusive results. It is attributable to factors other than costs that may be considered for hospitalization. Children in Korea are obliged to attend elementary school to middle school19; attendance is critical because it determines grade retention.20 Additionally, absence from school is associated with various risk factors for child development,21 including academic outcomes.22 This fact can also be noted in a report from the United Kingdom by the Office for Standards in Education, Children’s Services and Skill, which found that a persistent coronavirus disease 2019 related absence results in students lagging behind in school.23 Additionally, children of this age group can be more negatively affected by hospitalization. As school-aged children are affected by the external environment, both physical difficulties and disruption of emotional balance during hospitalization are significantly negatively affected.24 Teenagers are also known to have worse adaptation to hospitalization than younger children.25

This study has various limitations. First, the control group used in the study may have been inappropriate. Owing to the difference in healthcare utilization behavior in the groups aged 1–5, 6–10, 11–15, and 17,26 those aged 17 may not have been the appropriate control group. However, as the policy change was implemented among children under 15, those aged 17 were the closest in age to the case groups. Second, owing to data limitations, specific diseases could not be analyzed, and their severity could not be determined. For example, influenza is a known cause of hospitalization across age.27 However, because of the masking policy, it was not possible to perform this analysis. Therefore, it will be necessary to determine the severity of each disease in future studies. Third, various policies were implemented during the period from 2017 through 2019 owing to the Moon Jae-in government’s plan for benefit expansion.28 Other health policies may have affected children’s healthcare utilization patterns. Fourth, other factors that could influence a child’s health, such as parents’ characteristics, were not considered. Future studies should consider these aspects.

This study has several strengths as well. First, it uses NHIS-NSC data. The advantage is that the database represents nationwide data. It was followed from 2002 to 2019. Therefore, underlying comorbidities are incorporated to this study, and the results may be adjusted for these important factors. Second, this study analyzes the data using the DID method. The DID method is effectively employed in the public health research field and has been developing, thereby reinforcing its credibility.

The reduction in the coinsurance rate policy in hospitalization among children under 15 showed mixed results. Regarding those aged 1–5 (case group 1), the reduction of the coinsurance rate did not fulfill its purpose as a substitute effect was noted. Regarding those aged 6–10 (case group 2) and 11–15 (case group 3), the effect of the policy was inconclusive. Every aspect of the outcomes needs to be examined as policies tend to affect policy targets and social repercussions. Further research is necessary to investigate the policy regarding long-term effects, not only among those aged 1–5, but also among those aged 6–10 and 11–15.