Introduction

The incidence of income inequality has recently increased, resulting in the decline of household income quintiles. Income quintile groups are determined by calculating the equally distributed, disposable income for each household member. According to the “Household Financial and Welfare Survey Dynamic Change Analysis Report,” 21.8% of households faced an income quintile decline between 2011 and 2014 [1].

Several factors affect household income decline. Previous studies have shown that gender, age, educational level, employment status of the household head, number of members of the household employed, initial income, and the vocational training and social insurance benefits of the household head are factors that affect income quintiles [2–5]. When household heads were paid employees (including regular and non-regular workers) and self-employed workers, 20.8% and 28.4% of those households faced a decline in income quintile in 2014 compared with 2011, respectively. In addition, the rate of age-dependent income decline in 2014 was 20.2% lower for those aged ≤ 39 years, 23.1% for those aged between 40 years and 59 years, and 20.9% for those aged ≥ 60 years, compared with income in 2011 [1].

Household members with health complications that could negatively affect their capacity to work, resulting in a decrease in their income [6]. If those households were well-equipped with social insurance schemes, such as sickness insurance or sickness allowance, compensation for reduced income may be provided, and this may not lead to a decline in household income quintile.

Although the National Health Insurance system, a social insurance system that lowers the burden of out-of-pocket health payments is in operation in Korea, the burden of out-of-pocket payments borne by individuals and their family members continues to be high. The proportion of household out-of-pocket spending based on the total national health expenditures in Korea was 35.2% in 2011, approximately 1.8 times higher than the Organization for Economic Co-operation and Development (OECD) average of 19.6% [7].

The World Health Organization define catastrophic health expenditure as a household’s health expenditure that exceeds 40% of the household’s capacity to pay. An analysis of household catastrophic health expenditure in 59 countries showed that the proportion of households facing catastrophic health expenditures in Korea was 1.73% which is higher than the proportion in all OECD countries, except Portugal (2.71%) and Greece (2.17%) [8].

Households facing excessive health expenditures may be forced to reduce spending on other subsistence costs [9–10]. Under these adverse health conditions, low income households with a high risk of catastrophic health expenditures are more likely to spend their household income on health services [11]. This may lead to a lower quality of life, economic risks [12], and an increased probability of poverty [13–16].

In the past 3 years, 20% of households have experienced a decline in income quintile. It has been speculated that the incidence of catastrophic health expenditures without sick allowance in Korea, may have had a significant impact on the decline in income quintile of middle-class households. However, there have been no studies to investigate the effects of catastrophic health expenditures on income quintile decline.

In the present study, we analyzed the proportion of households facing catastrophic health expenditures by considering households with and without income quintile decline.

Materials and Methods

1. Participants

The present study used data from the Korea Health Panel (KHP), 2012–2013 collected by the National Health Insurance Cooperation, and the Korea Institute for Health and Social Affairs. The KHP survey was conducted in 2008 on 7,866 households and 24,616 individuals. Due to loss of households for reasons such as death or survey refusal, the sample size was reduced to 5,850 households and 17,417 individuals in 2012. In 2013, 2,500 households nationwide were additionally sampled, resulting in a sample size of 7,743 households and 22,701 individuals. In the present study, there were 12,909 subjects aged 20 years to 64 years from economically active households, whose income quintile remained unchanged or declined by more than one quintile from 2012 to 2013. There were 3,489 households whose income quintile increased by one or more quintiles and were excluded from the analysis.

2. Variables

The dependent variable was income quintile decline, which was defined as income quintile that declined by more than one quintile between 2012 and 2013, and which was then compared to the households whose income quintiles remained unchanged. Income was entered in the questionnaire with the total annual income of household members, and the survey period was from January to December of the previous year. Total annual household income was calculated as follows:

Total annual household income=Total annual household income/Total number of household members

Total annual household income is composed of earned income, financial income, rental income etc. The income quintiles used in this study are divided into 5 quintiles based on the total annual household income. The 1^st quintile is the lowest 20% income bracket, and the 5^th quintile is the top 20% and the highest-income bracket.

The independent variables were determined using data from the 2012 Korea Health Panel. The independent variables included gender, age, educational level, health status (EQ-5D), presence or absence of chronic diseases, number of outpatient visits, private health insurance coverage, catastrophic health expenditures, and employment status.

The catastrophic health expenditure was divided into percentages of household out-of-pocket health expenditure of the household’s capacity to pay, or the thresholds of catastrophic health expenditures of ≥ 40%, ≥ 30%, ≥ 20%, and ≥ 10%. The households’ capacity to pay was calculated after excluding the average food expenses from the monthly living expense and was converted into the annual unit. The household out-of-pocket health expenditure was calculated by summing up direct spending on emergency medical service, outpatient medical service, and prescription drug expenditure (emergency, and inpatient and outpatient care) per household.

The employment status was divided into regular workers and non-regular workers (temporary, day workers), self-employed workers (including unpaid family workers), and economically inactive individuals. The educational level was divided into college graduates or above, and high school graduates or below. The private health insurance coverage was divided into fixed benefit type, indemnity type, and mixed type (fixed type and indemnity type).

3. Data analysis

Logistic regression analysis was conducted to determine whether catastrophic health expenditures in 2012 were related to more than one quintile income decline in 2013. The threshold levels of catastrophic health expenditures were divided into 40%, 30%, 20%, and 10% of a household’s capacity to pay and this was analyzed.

4. Analysis model

Logit (income quintile decline) = α + β₁Gender_i + β₂Age_i + β₃Education_i + β₄Employ_i + β₅Chronic Disease_i + β₆EQ-5D_i + β₇Outpatient Visit_i + β₈Private Health Insurance_i + β₉ Catastrophic Health Expenditure_.

The proportion of households whose income declined by more than the first quintile was estimated by gender, age, education, employment status, chronic disease, EQ-5D (health status), number of outpatient visits, health insurance, and catastrophic health expenditure. The intercept of the regression equation was calculated as the median of the number of outpatient visits for all household members. Gender (Male = 0, Female = 1), Education (Less than High School = 0, Graduate more = 1), Employment status (Regular worker = 0, Non-regular worker = 1, Self-employ = 2, Non economically = 3), Chronic disease (No = 0 Yes = 1), Private Health Insurance (Fixed benefic type = 0, Indemnity type = 1, Mixed type = 2), Catastrophic health expenditure (Less than Threshold level = 0, Threshold level more = 1) is binary variables.

Results

1. Distribution of households facing catastrophic health expenditures by year and income quintile

In the present study, the proportion of households facing catastrophic health expenditures did not significantly differ between 2012 and 2013. It was rather different depending on the threshold level of catastrophic health expenditures. When the threshold levels of catastrophic health expenditures were 40%, 30%, 20%, and 10%, the proportions of households facing catastrophic health expenditures were 2.5%, 4.1%, 8.8%, and 24.3%, respectively in 2012 and 2.1%, 3.7%, 7.8%, and 20.4%, respectively in 2013 (Table 1).

The proportion of households facing catastrophic health expenditures significantly differed based on their income quintile and according to the threshold level of the catastrophic health expenditure. The proportion of households facing catastrophic health expenditures was 11.7% in the 1^st quintile and 1.1% in the 5^th quintile when the threshold of catastrophic health expenditure was ≥ 40%, whereas it was 42.8% in the 1^st quintile and 13.6% in the 5^th quintile when the threshold of the catastrophic health expenditure was ≥ 10% (Table 2).

2. Distribution of the rate of income quintile decline by explanatory variable: univariate analysis

The rate of more than one quintile income decline was higher in high school graduates or below (22.53%) in comparison to college graduates or above (19.1%) (p < 0.001).

The rate of more than one quintile income decline according to employment status was the highest with 24.54% being economically inactive individuals, followed by non-regular workers at 21.67%, self-employed workers at 20.93%, and regular workers at 16.02% (p < 0.001).

The threshold levels of catastrophic health expenditures were 40%, 30%, 20%, and 10%. The rate of more than one quintile income decline was higher in households experiencing above the threshold catastrophic health expenditures in comparison to those experiencing below the threshold catastrophic health expenditures (p<0.001) (Table 3).

3. Relationship between catastrophic health expenditures and income quintile decline

Households facing catastrophic health expenditures of ≥ 40% were 1.58 times more likely to face a decline in their income quintile in comparison to households facing catastrophic health expenditures of < 40% (p < 0.003). In addition, households with a threshold of catastrophic health expenditures of ≥ 30% and ≥ 10% were 1.75 times (p < 0.001) and 1.23 times (p < 0.001) more likely to have a decline in income quintile, respectively. Non-regular workers, self-employed, and economically inactive individuals were 1.44 times, 1.46 times, and 1.42 times more likely to experience a decline in income quintile, respectively, in comparison to regular workers (p < 0.001). High school graduates or below were 1.27 times more likely to have a decline in income quintile in comparison to college graduates or above (p < 0.001). The indemnity-type private health insurance subscribers were 0.73 times less to likely to have an income quintile decline in comparison to fixed-type private health insurance subscribers (p < 0.002) (Table 4).

Discussion

The results in this study, together with those from recent studies, indicate that 1 out of 6 to 10 households in Korea experiences a decline in household income in a1 year period. In the present study, 16.4% of households experienced more than one quintile income decline from 2012 to 2013. In addition, when income quintile decline was classified in terms of the number of quintiles declined, the rate of 1, 2, 3, or 4 income quintile declines were 13.6%, 2.35%, 0.47%, and 0.03%, respectively. In addition, in another study analyzing income mobility for 3 income levels (low-income, middle-income, and high-income levels), the rate of income level decline for 1 year (between 2011 and 2012) was 10.4% [17].

When the observation period was extended to 3 to 4 years, the rate of income quintile decline was higher, with 1 out of 5 to 7 households experiencing a decline in income [1]. According to the “Household Financial and Welfare Survey Dynamic Change Analysis Report,” 21.8% of households experienced a decline in income quintile from 2011 to 2014 [1]. In a study where the income bracket was divided into 3 income brackets (low-income, middle-income, and high-income), the rate of income bracket decline over 4 years (from 2008 to 2012) was 14.8%. The middle-income bracket moved down by 5.6% into the low-income bracket, the high-income bracket moved down to the middle-income bracket by 7.9%, and the high-income bracket moved down to the low-income bracket by 1.5% [17]. In another study, in which the income bracket was divided into 4 categories based on the median income, the proportion of households with urban workers whose income bracket declined by more than two quintiles was 10.8% between 2003 and 2008 [2].

The rates of income quintile decline within 1 year in Korea were reported to be lower than that of Argentina and China. In Argentina, the income quintile decline rate was 9.1% to 22.7% between 2001 and 2002. In China, the rates of income quintile decline were 24.2% to 27.7% between 1990 and 1995 [18].

It has been reported that the rates of income quintile decline over 10 years in 6 countries were 47%–60% [19], which was higher than the rate of 1year income quintile decline of 16.4% in Korea. Comparison of income levels between 1992 and 2002 showed that the rates of income quintile decline from the highest or fifth income quintile, to 1 to 3 lower quintiles were 47% in Sweden, 48% in Demark, 50% in Norway and Finland, 55% in the United States of America (US), and 60% in the UK. In addition, the rates of income quintile decline from the highest or fifth income quintile to the lowest or first income quintile were 10% in the US, 11% in the UK, 25% in Denmark, 26% in Sweden, and 28% in Norway and Finland [19]. The reasons for the difference between the results of the aforementioned study and the results of studies in Korea are as follows. First, the former study had a longer observation period of 10 years, and secondly, the income quintile decline was based on assessments starting at the richest income quintile.

It is catastrophic if more than 40% of a household’s capacity to pay is spent on health. From the results of the present study, 2.5% to 2.1% of Korean households faced catastrophic health expenditures in 2012 or 2013. A previous study also reported that the proportion of households facing catastrophic health expenditures in Korea was 4.5% in 2015, which was the highest among the OECD countries [20]. The reason for this is that the national health insurance coverage rate, is lower than the average health insurance coverage rate in other OECD countries. In particular, Korea’s health insurance system has low coverage, with high out-of-pocket costs and many items not covered [21–23]. According to the OECD Health Data, the proportion of household direct out-of-pocket payments based on the current health expenditures was 36.9% in Korea, and this was amongst the highest of the OECD countries, after Mexico (44.7%). On the other hand, the proportion of households’ direct out-of-pocket payments based on the current health expenditures was low in European countries, such as Sweden (14.8%), Norway (14.6%), the UK (9.9%), and France (6.7%) [7].

The results of the present study found that households facing catastrophic health expenditures in Korea were 1.58 times more likely to face income decline, when catastrophic health expenditures were ≥ 40% of a household’s ability to pay. The probability of household income quintile decline differed according to the definition of catastrophic health expenditure. More specifically, households facing catastrophic health expenditures of ≥ 10% and ≥ 30%, they had a 1.23 times and 1.75 times higher chance of income quintile decline, respectively.

To date, there is no domestic study that analyzes the relationship between catastrophic health expenditures and income quintile decline. However, 2 domestic studies have analyzed the relationship between catastrophic health expenditures and the incidence of poor households, and have shown that households facing catastrophic health expenditures that spent ≥ 10% of their capacity to pay, were 1.42 or 1.88 times more likely to fall into poverty. Poverty, is an independent variable, defined as < 50% of the median income, based on disposable income [13–14].

If the proportion of households facing catastrophic health expenditures can be lowered, the probability of a decline in income quintile can be reduced. Various policies, such as increasing the national insurance coverage rate, are required to reduce the burden of the rate of household out-of-pocket payments. If the level of national health insurance coverage is low, people with low incomes may not be able to receive the required medical services, or may face economic risks due to the heavy burden of health expenditure, even if they used medical services [24]. In other words, to lower catastrophic medical expenditures, various policies that reduce the number of non-covered items are needed, along with lower out-of-pocket costs.

Another factor related to income quintile decline was economic activity. Maintaining a stable form of employment, such as regular work, could lower the likelihood of income quintile decline. In the present study, the rate of more than one quintile income decline was highest among economically inactive individuals, followed by non-regular workers, such as daily workers, temporary workers, self-employed, and regular workers (Table 3). A previous study also found that the rate of transition from non-poverty to poverty was highest when household heads were economically inactive, followed by non-wage workers, wage workers, and unemployed workers [25].

Furthermore, even after adjusting for catastrophic health expenditures, self-employed workers faced the highest probability of decline in income quintile in comparison to regular workers, followed by non-regular workers and economically inactive workers (Table 4). A previous study reported that the income quintiles of regular workers were less likely to move down to lower income quintiles [3], and the household heads of urban laborers who were temporary or day workers faced a higher probability of decline in income quintile in comparison to regular workers [2].

In the present study, the reasons why regular workers were less likely to face a decline in income quintile were not analyzed. However, it is judged that maintaining a stable form of employment, such as regular work, can prevent poor households moving toward income quintile decline. In addition, it is considered necessary to introduce more active social security systems to prevent such a decline in the income quintiles of non-regular and self-employed workers. That is to say, a sickness allowance system, which ensures income even if one becomes unemployed due to an illness, is needed. European countries, which feature relatively low household out-of-pocket costs, guarantee a certain amount of income through a sickness allowance system even when individuals cannot work due to an illness [26].

Unlike advanced welfare countries which are well-equipped with a social security system, such as sickness insurance, the sickness allowance system in Korea is not adequate.

A decline in income quintile caused by disease and medical care costs may lead to a reduction in national economic activity. The independent variable of this study, the 2012 catastrophic health expenditure experience, was significantly associated with the dependent variable, income quintile decline. However, the limitation of the present study is that the observation period of catastrophic health expenditures and income quintile decline was short. Continued spending on catastrophic health expenditures may lead to poverty due to declines in household income. Further studies are required to determine whether continuous catastrophic health expenditures may lead to income quintile decline and poverty. In addition, in the current situation where various types of social security systems, such as child allowance, basic old-aged pension reinforcement, and minimum wage increase are introduced, studies are required to investigate the effects of such policies on income quintile decline. Furthermore, although income quintile decline is a rank-type of variable, we could not apply an ordered logit model due to the rate of income quintile decline across income quintiles. Additional studies are needed to analyze income quintile decline across income quintiles using an ordered logit model.

Article information

Conflicts of Interest

All authors have no conflicts of interest to declare.

References

• 1. Statistics Korea. Household financial and welfare survey dynamic change analysis report [Internet] Daejeon: KOSIS; 2014 [cited 2016 Sep 23]. Available from: http://kosis.kr/ups/ups_01List.jsp?pubcode=HF.
• 2. Kang SW. Income Mobility of Urban Laborer Household of Korea since 1990. Korea Development Economics Association 2010;17:1−28.
• 3. Kim ST, Chun YJ, Lim BI. An Analysis on the Income Mobility and its Determinants in Korea. The Korean Economic Association 2012;60:5−43. PMID: 10.5124/jkma.2017.60.1.5.Article
• 4. Shi X, Liu X, Nuuetah A, et al. Determinants of Income Mobility in Rural China. China & World Econ 2010;18:41−59. PMID: 10.1111/j.1749-124X.2010.01188.x.Article
• 5. Woolard I, Klasen S. Determinants of Income Mobility and Households Poverty Dynamics in South Africa. J Dev Stud 2005;41:865−97. PMID: 10.1080/00220380500145313.Article
• 6. Marmot M. Social determinants of health inequalities. Lancet 2005;365(9464). 1099−104. PMID: 10.1016/S0140-6736(05)74234-3. PMID: 15781105.ArticlePubMed
• 7. Korea Institute for Health and Social Affairs, Ministry of Health & Welfare in Korea. OECD Health Data 2015. Sejong: Korea Institute for Health and Social Affairs, Ministry of Health & Welfare; 2015.
• 8. Xu K, Evans DB, Kawabata K, et al. Household catastrophic health expenditure: a multicountry analysis. Lancet 2003;362:111−7. PMID: 10.1016/S0140-6736(03)13861-5. PMID: 12867110.ArticlePubMed
• 9. Kim Y, Yang B. Relationship between catastrophic health expenditures and household incomes and expenditure patterns in South Korea. Health Policy 2011;100:239−46. PMID: 10.1016/j.healthpol.2010.08.008.ArticlePubMed
• 10. Wagstaff A. Measuring financial protection in health. Policy Research Working Paper 4554. The World Bank; Washington DC, USA: 2008. pp 1−31.Article
• 11. Pannarunothai S, Mills A. The poor pay more: health-related inequality in Thailand. Soc SciMed 1997;44:1781−90.Article
• 12. Song EC, Shin YJ. The Effect of Catastrophic Health Expenditure on the Transition to Poverty and the Persistence of Poverty in South Korea. J Prev Med Public Health 2010;43:423−35. PMID: 10.3961/jpmph.2010.43.5.423. PMID: 20959713.ArticlePubMedPDF
• 13. Woo KS, Shin YJ. The Effect of Catastrophic Health Expenditure on Household Economy: Focusing on Financial Coping and Poverty. Health Soc Welf Rev 2015;35:166−98. http://dx.doi.org/10.15709/hswr.2015.35.3.166. PMID: 10.15709/hswr.2015.35.3.166.Article
• 14. Lee HJ. Determinants of household catastrophic health expenditure and impacts on poverty. Seoul National University; 2015. pp 1−178.
• 15. Cheon BY. Growing Inequality and its’ Drivers in Korea. Democr Soc and Policy Stud 2013;23:15−40.
• 16. Knaul FM, Arreola-Ornelas H, Mendez-Carniado O, et al. Evidence is good for your health system: policy reform to remedy catastrophic and impoverishing health spending in Mexico. Lancet 2006;368:1828−41. PMID: 10.1016/S0140-6736(06)69565-2. PMID: 17113432.ArticlePubMed
• 17. Jeon BM. Income Shifting Between Personal and Corporate Income Tax Bases. Research paper. Sejong: Korea Institute of Public Finance; 2013. December;Report No. 13-3.
• 18. Fields GS. What We Know About Earnings Mobility in Developing Countries Cornell University ILR School; 2011 http://digitalcommons.ilr.cornell.edu/workingpapers.
• 19. Julia BY. International Comparisons of Economic Mobility. The Brookings Institution; 2006.
• 20. Xu K, Jeong HS, Saksena P, et al. Financial risk protection of National Health Insurance in the Republic of Korea: 1995–2007. World Health Report (2010). Background Paper, 23.
• 21. Park IH. Analysis on the Level of National Health Expenditure and Associated Factors in the OECD Countries. Korean J Health Policy Adm 2012;22(4). 538−60. PMID: 10.4332/KJHPA.2012.22.4.538.ArticlePDF
• 22. Shin YS. Policy Measures of Medical Safety Net for Low-income and Underprivileged Brackets. Health Soc Welf Rev 2012-03 6−14.
• 23. Song EC, Shin YJ. The Comprehensive Health Expenditure Ceiling System to Prevent Catastrophic Health Expenditure: Focusing on Applicability Using Cost Estimation. Health Soc Welf Rev 2015;35(2). 429−56. PMID: 10.15709/hswr.2015.35.2.429.Article
• 24. Huh SI, Kim CB. An evaluation of benefit extension strategies of the Korea National Health Insurance. Korean J Health Policy Adm 2009;19:142−65. https://doi.org/10.4332/KJHPA.2009.19.3.142.PMID: 10.4332/KJHPA.2009.19.3.142.ArticlePDF
• 25. Seok SH. Trend and causes of household income mobility in Korea. The Korean Social Security Association 2009;25:25−44.
• 26. Moon SY, Choi EH, Kim SH, Lee JW. A Study on the Status and Implications of the Sickness Allowance System in Major Countries. Kangwon: National Health Insurance Corporation Health Insurance Policy Research Institute; 2015. Report No. 2015-1-0006.

Figure & Data

References

Citations

Citations to this article as recorded by  

• Association of Depressive Symptoms With Health Service Use and Catastrophic Health Expenditure Among Middle-Aged and Older Chinese Adults: Analysis of Population-Based Panel Data
  Rui Yan, Lifeng Li, Xiaoran Duan, Jie Zhao
  Journal of the American Medical Directors Associat.2023; 24(5): 664.     CrossRef
• Out-of-pocket expenditure among patients with diabetes in Bangladesh: A nation-wide population-based study
  Zakir Hossain, Moriam Khanam, Abdur Razzaque Sarker
  Health Policy OPEN.2023; 5: 100102.     CrossRef
• Analysis of Factors Affecting the Awareness of the Health Care System
  Hyojeong Lee, Jihye Lim
  Journal of Health Informatics and Statistics.2023; 48(3): 242.     CrossRef
• Laos’ Social Health Insurance (SHI) program’s impact on older people’s accessibility and financial security against catastrophic health expense
  Somdeth Bodhisane, Sathirakorn Pongpanich
  BMC Health Services Research.2023;[Epub]     CrossRef
• Gastos catastróficos em saúde: análise da associação com condições socioeconômicas em Minas Gerais, Brasil
  Jéssica de Brito Macedo, Alexandra Crispim Boing, Juliana Mara Andrade, Helton Saulo, Rodrigo Nobre Fernandez, Fabíola Bof de Andrade
  Ciência & Saúde Coletiva.2022; 27(1): 325.     CrossRef
• What Policy Approaches Were Effective in Reducing Catastrophic Health Expenditure? A Systematic Review of Studies from Multiple Countries
  HyunWoo Jung, Kwang-Soo Lee
  Applied Health Economics and Health Policy.2022; 20(4): 525.     CrossRef
• Elderly chronic diseases and catastrophic health expenditure: an important cause of Borderline Poor Families’ return to poverty in rural China
  Xiaocang Xu, Haoran Yang
  Humanities and Social Sciences Communications.2022;[Epub]     CrossRef
• Does Elderly Chronic Disease Hinder the Sustainability of Borderline Poor Families’ Wellbeing: An Investigation From Catastrophic Health Expenditure in China
  Xiaocang Xu, Haoran Yang
  International Journal of Public Health.2022;[Epub]     CrossRef
• The Relationship between Unmet Healthcare Needs Due to Financial Reasons and the Experience of Catastrophic Health Expenditures
  Jeong-Hee Kang, Chul-Woung Kim
  Journal of Korean Academy of Community Health Nurs.2021; 32(1): 95.     CrossRef
• Effect of financial stress on self-rereported health and quality of life among older adults in five developing countries: a cross sectional analysis of WHO-SAGE survey
  Rui Huang, Bishwajit Ghose, Shangfeng Tang
  BMC Geriatrics.2020;[Epub]     CrossRef