Healthcare utilization among foreign beneficiaries of the National Health Insurance program in Korea

Abstract

Background

To explore the factors related to healthcare utilization (both inpatient and outpatient services) among foreign beneficiaries of the National Health Insurance (NHI) in Korea.

Participants

The total foreign beneficiaries of the NHI in Korea 2022 were included (n = 1,386,329).

Variables

For the dependent variables, utilization and either hospital stays or count of visits were used. And the independent variables were selected guided by Andersen’s behavioral framework: predisposing characteristics (e.g., sex, age, nationality), enabling characteristics (e.g., insurance premium ranking, residency status, residency period, place of residency, beneficiary type), and need characteristics (e.g., disability, CCI, chronic diseases).

Statistical models

We conducted a two-part model analysis employing Andersen’s behavioral framework, using health insurance claims data from 2022.

Results

The results indicated that individual need characteristics, such as disability, CCI, and chronic diseases, had the most significant impact on total healthcare utilization and its associated costs. The next significant characteristic was the residence period; the longer the period, the higher the likelihood of healthcare utilization, but the lower the associated costs. Among the predisposing characteristics, individuals aged 65 years and older had the most significant impact on healthcare costs. Additionally, among the enabling characteristics, residency period, residency status, and beneficiary type were most significant.

Conclusion

Foreigners residing in Korea are eligible to use the Korean healthcare system by enrolling in NHI. As the foreign population is projected to grow even faster due to Korea’s aging society and low birth rate, this study can assist public health administrators in formulating evidence-based policies.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12913-024-11698-2.

Background

Korea achieved universal health coverage in 1989 [1]. However, the public healthcare system hasn’t fully covered foreign residents who do not hold Korean nationality, leading to unequal access to healthcare. Although legislation to extend healthcare coverage to foreigners was introduced in 1981, actual discussions began to develop in the 1980s due to labor shortages [2].

The National Health Insurance (NHI) in Korea covers employees, self-employed individuals, and their families. The employee category includes government employees, school teachers, business employers and employees. Since the self-employed category encompasses households not covered by the employee category, the Korean NHI effectively covers the entire population. Foreign employees could voluntarily enroll in the NHI since 1998 but were mandated to do so in 2006, and self-employed foreigners since 1999, with mandatory enrollment in 2019. Consequently, every foreigner could access Korean healthcare securely by 2019 [2, 3].

According to the Ministry of Justice in Korea, there were 2,245,912 foreigners residing in the country in 2022 [4]. Of these 1,343,172 (59.8%) were enrolled in the NHI [5]. The number of foreign residents is expected to increase rapidly due to the influx of foreign workers and students, the rise in international marriages, and the return of overseas Koreans, which could lead to a growth in NHI enrollees.

Foreigners in Korea utilize healthcare services less frequently, attributed to various factors such as language barriers, limited social networks, cultural differences, health-related beliefs, lack of information, economic constraints, geographical limitations, and other similar reasons [2, 6–8]. However, these findings have limitations in generalizing to the entire foreign population in Korea, as they are derived from qualitative studies or population surveys based on small sample sizes. Additionally, one study [2] that utilized administrative data was confined to voluntary enrollees as its study population. Therefore, it is essential to investigate the healthcare utilization patterns of foreigners in Korea following the implementation of the mandatory enrollment policy in 2019. We aimed to investigate healthcare utilization patterns among the entire foreign population in Korea, employing Andersen’s behavioral model framework.

Ronald M. Andersen, a U.S. medical sociologist, developed Behavioral Model of Health Services Use in 1968, which has become one of the most frequently used frameworks for explaining healthcare utilization. Andersen initially designed the model to understand how households utilized health services in the late 1960s. However, due to the heterogeneity among household members, he later adapted it to the individual level and further refined the individual-level characteristics [9].

The individual-level characteristics consist of predisposing factors, enabling factors, and need factors. Predisposing factors include demographics (e.g., age and gender), social structure (e.g., education, occupation, ethnicity, and social networks), and health beliefs (e.g., attitudes, values, and knowledge related to health and health services). Enabling factors refer to conditions that facilitate or hinder the use of health services, such as income and health insurance. Need factors are the most directly related factors to the use of health services and include perceived need and evaluated need. Perceived need refers to how individuals experience and emotionally respond to symptoms, pain, and health conditions. Evaluated need, on the other hand, represents professional judgement and objective assessment of a patient’s health status [10, 11]. Since we aimed to explore individual factors related to healthcare utilization among foreign beneficiaries, this research was designed based on Andersen’s behavioral model framework.

Methods

Study design

This cross-sectional study was designed to explore factors related to both inpatient and outpatient health services among all foreign beneficiaries of NHI registered from January 1st to December 31st, 2022.

Setting

For this study, health insurance claims data were utilized. It is worth noting that Korea offers universal healthcare coverage through its national health insurance system [12]. Foreign citizens become eligible for NHI after completing alien registration. Foreign employees and their household members are covered by NHI when employed by a company that participates in the NHI program, i.e., a company with at least one employee beneficiary. Other foreigners, on the other hand, automatically become eligible for the NHI self-employed plan after residing in Korea for six months and longer. This is possible because the alien registry from the Ministry of Justice Korea is shared with the National Health Insurance Service (NHIS). The alien registry contains information on each foreigner’s nationality and residency type. With this shared information, the NHIS can issue NHI cards and send insurance contribution bills to self-employed foreign beneficiaries. Since the NHIS operates on a third-party payment system, where medical institutions submit medical bills to the NHIS, it manages various databases, including enrollees’ socioeconomic information and patients’ medical records.

Participants

The total number of foreign beneficiaries enrolled in 2022 was 1,391,418. The National Health Insurance Statistical Yearbook 2022 reported a total of 1,343,172 foreign beneficiaries, counted based on those with valid membership at the end of the year [5]. Since our study population includes all foreign beneficiaries throughout 2022, it has a greater number than reported in the yearbook.

After extracting the total foreign beneficiaries from the database, we excluded samples with missing values, such as those without nationality (n = 3,825) and those without residency type (n = 740). Additionally, samples missing starting and end date for their residency were excluded (n = 524). After these exclusions, the final sample was balanced in terms of sociodemographic and medical variables, resulting in a total sample size of 1,386,329 (Fig. 1).

Fig. 1.

Flowchart of the selection of subjects

Variables and measurement

The dependent variable was represented by separate dummy variables for healthcare utilization, distinguishing between inpatient and outpatient services. A value of 1 indicated the use of the respective service, while a value of 0 indicated non-utilization. The hospitalization period and the duration of doctors’ visits were represented by continuous variables, measured in days. Similarly, the total payment for hospitalization and outpatient services was represented by continuous variables, measured in Korean won.

The selection of variables was guided by Andersen’s behavioral framework. Firstly, the predisposing characteristics considered were sex, age, and nationality. The nationality category included China, Vietnam, Uzbekistan, USA, Japan, Canada, and Taiwan, with each nationality selected based on its ranking in at least one of the following factors: the number of enrollees, total amount of medical payment, and medical bill per person.

The enabling characteristics considered included insurance premium ranking, residency status, resident period, place of residence, and insurance beneficiary type. The insurance premium ranking serves as an indicator of one’s economic status, ranging from 1 for the lowest to 4 for the highest level, determined based on factors such as income and wealth. Residency status was classified into six groups based on ranking: overseas Korean, low-skilled employment, permanent residency, student, marriage, and other categories. Overseas Korean refers to individuals who acquired Korean nationality at birth but later acquired foreign nationality, as well as their offspring. Permanent residency refers to individuals who have resided in Korea for a certain period and whose income level is at least equal to the Gross National Income (GNI) per person. Student refers to foreign students enrolled in educational institutions in Korea. Marriage refers to either foreigners married to Korean nationals or individuals raising children born from a marriage to Korean nationals. Low-skilled employment refers to foreign blue-collar workers employed in industries such as manufacturing, construction, agricultural, livestock and fisheries, food and accommodation, and housekeeping services.

The period of residency was determined by calculating the duration between the starting and ending dates based on one’s last residence status in 2022. However, the resident period was limited to a maximum of five years because we tracked data from 2022 back to 2018. Then, the resident period was divided into four groups: less than one year, from one year to less than three years, from three years to less than five years, and more than five years.

The place of residence was divided into two groups due to the concentration of foreigners in Seoul, Incheon, and Gyeonggi Province: Metropolitan area and other areas. The insurance beneficiary type was categorized into four groups: employees, family members of employees, self-employed individuals, and household members of the self-employed.

The need characteristics considered included disability status, the Charlson Comorbidity Index (CCI), and the presence of chronic diseases. Because households with a registered disabled member receive a reduction benefit on insurance contributions (e.g., 20% for mild disabilities, and 30% for severe disabilities), the NHIS maintains a disabled registry. For this study, the sample was divided into two groups: 1 for those with either a mild or severe disability, and 0 for those without a disability.

Since a patient’s disease severity can impact healthcare utilization by contributing to complications, increased length of stay, higher medical bills, and mortality, diagnosis information from claims data is valuable for constructing the analysis dataset. We used ICD-10 codes from each patient’s claims data over the past two years to calculate the CCI score (Supplementary Material 1). After calculating the CCI score, the study sample was divided into two groups: 0 and more than 1 [13].

The presence of chronic diseases was defined if an individual had a medical record of any of the twelve specific diseases within the past two years. These diseases, selected based on the Statistical Year Book of the NHIS [5], include hypertension, diabetes mellitus, heart disease, cerebrovascular disease, malignant neoplasm, liver disease, mental and behavioral disorders, tuberculosis, diseases of the nervous system, disorders of the thyroid gland, chronic kidney disease, and musculoskeletal diseases and arthropathies (Supplementary Material 2).

Statistical modeling

For the descriptive statistics of continuous variables, we utilized mean ± standard deviation (SD) and median with interquartile range (IQR) because the median is less sensitive to outliers in skewed distributions, such as medical bills. For categorical variables, we presented the percent distribution of the enrollees. We applied a two-part model to analyze the factors influencing healthcare utilization and the extent of utilization, assuming that the factors related to utilization and its amount differ.

A two-part model is useful when most medical bill values are zero (0) with heterogeneous variance [14]. Therefore, we conducted the analysis in two steps. Firstly, logistic regression was applied to identify the factors related to healthcare utilization. Then, a generalized linear model with a log link function was applied to individuals who utilized healthcare services, assuming a gamma distribution.

Because some ICD-10 codes were used both to define the presence of chronic diseases and to calculate the CCI score, the two variables may exhibit multicollinearity. However, the variance inflation factor (VIF) between the two variables were 1.3, indicating that multicollinearity was not an issue in this case.

Results

A summary of characteristics from the study population can be found in Table 1. In 2022, male foreigners (52.1%) were enrolled in NHIS more than females (47.9%), and the largest age group among foreigners was 20–34 years old (36.2%). The socially active age group of 20–64 years old comprised the majority of the sample population at 85.1%, with an average age of 39.7 years old. The majority nationality was Chinese, comprising 50.9%, while the most frequent residency status was overseas Korean, comprising 29.9%. The majority duration of residency was more than five years, comprising 42.8%, and 65.8% of the sample population resided in the Metropolitan area. The third quartile of the insurance premium ranking accounted for 46.5% of the study population, and the largest beneficiary type was self-employed individuals, comprising 41.2%, followed by employees, comprising 33.5%. Disabled individuals comprised only 0.4% of the sample population, so the majority of foreign enrollees were non-disabled. Individuals with a CCI score higher than 1 comprised 38.2% of the population, while 32.1% of the study population had at least one chronic disease out of the 12 diseases considered.

Table 1.

Descriptive statistics for the study population

Variables			N	(%)
Total			1,386,329	(100%)
Predisposing characteristics	Sex	Male	722,845	(52.1%)
		Female	663,484	(47.9%)
	Age, years	Under 20	97,503	(7.0%)
		20–34	501,559	(36.2%)
		35–49	367,141	(26.5%)
		50–64	311,274	(22.5%)
		65 or older	108,852	(7.9%)
		Mean age	39.7(± 16.7)
	Nationality	China	705,247	(50.9%)
		Vietnam	143,781	(10.4%)
		Uzbekistan	59,157	(4.3%)
		USA	52,226	(3.8%)
		Japan	23,457	(1.7%)
		Canada	16,178	(1.2%)
		Taiwan	14,971	(1.1%)
		Etc.	371,312	(26.8%)
Enabling characteristics	Insurance premium ranking	1Q(low)	221,789	(16.0%)
		2Q	404,062	(29.2%)
		3Q	645,134	(46.5%)
		4Q(high)	115,344	(8.3%)
	Residency status	Overseas Korean	415,040	(29.9%)
		Low-skilled employees	284,394	(20.5%)
		Permanent residency	157,699	(11.4%)
		Foreign students	148,169	(10.7%)
		Marriage	129,526	(9.3%)
		Etc.	251,501	(18.1%)
	Residency period, years	Under 1	148,367	(10.7%)
		1–3	249,042	(18.0%)
		3–5	396,089	(28.6%)
		5 or more	592,831	(42.8%)
	Place of residency	Outside of metropolitan area	474,815	(34.3%)
		Metropolitan area(Seoul, Gyeonggi, Incheon)	911,514	(65.8%)
	Beneficiary type	Employees	464,631	(33.5%)
		Family members of employees	212,514	(15.3%)
		Self-employed	571,011	(41.2%)
		Household members of self-employed	138,173	(10.0%)
Need characteristics	Disability	Non-disabled	1,381,040	(99.6%)
		Disabled	5,289	(0.4%)
	CCI	0	856,376	(61.8%)
		1 or more	529,953	(38.2%)
	Chronic disease	None	941,009	(67.9%)
		Present	445,320	(32.1%)

The average amount of health care utilization of the study population can be found in Table 2. Only 9.2% of the study population used inpatient services during the year, while 77.5% used outpatient services. Among those who used inpatient services, the average duration of hospitalization was 11.7 days, with a median of 4.0 days, and the average total medical bill amounted to $3,206, with a median of $1,409. Among those who used outpatient services, the average duration of an outpatient visit was 13.5 days, with a median of 8.0 days, and the average total medical bill amounted to $565, with a median of $281.

Table 2.

Utilization status of medical services of the study sample

Dependent variable			N	(%)
Inpatient services	Utilization	No	1,258,677	(90.8%)
		Yes	127,652	(9.2%)
	Hospital staysa	Mean ± SD	11.7 ± 35.6
		Median (IQR)	4.0 (6.0)
	Costsac	Mean ± SD	3,206 ± 7515
		Median (IQR)	1,409 (2,229)
Outpatient services	Utilization	No	312,639	(22.6%)
		Yes	1,073,690	(77.5%)
	Count of visitsb	Mean ± SD	13.5 ± 16.3
		Median (IQR)	8.0 (15.0)
	Costsbc	Mean ± SD	565 ± 1327
		Median (IQR)	281 (560)

^aonly for the hospitalized patients (n = 127,652)

^bonly for those utilized outpatient services (n = 1,073,690)

^c1 USD = 1291.95 KRW (in 2022)

Factors related to inpatient services

The results of the two-part model analysis for inpatient services can be found in Table 3. The need characteristics, such as disability, CCI, and the presence of chronic disease, significantly influenced the probability of hospitalization, duration of inpatient service, and the total amount of services provided.

Table 3.

Result of two-part model analysis of inpatient services

Variable		1st part		2nd part
		Utilization		Hospital stays		Costs
		β	odds ratio	β	exp(β)	β	exp(β)
Sex (ref.=male)	Female	0.087*	1.09	-0.197*	0.82	-0.144*	0.87
Age (ref.=under 20)	20–34	-0.387*	0.68	-0.035	0.97	0.161*	1.17
	35–49	-0.586*	0.56	0.006	1.01	0.242*	1.27
	50–64	-0.525*	0.59	0.163*	1.18	0.419*	1.52
	65 or older	-0.201*	0.82	0.790*	2.2	0.723*	2.06
Nationality (ref.=etc.)	China	0.031	1.03	-0.004	1	0.012	1.01
	Vietnam	-0.110*	0.9	0.067*	1.07	0.080*	1.08
	Uzbekistan	0.179*	1.2	-0.025	0.98	0.066	1.07
	USA	-0.141*	0.87	-0.088*	0.92	0.001	1
	Japan	-0.179*	0.84	0.046	1.05	0.110*	1.12
	Canada	-0.163*	0.85	-0.089	0.92	0.054	1.06
	Taiwan	-0.112	0.89	0.457*	1.58	0.277*	1.32
Insurance premium rank (ref.=1Q)	2Q	0.028	1.03	-0.031	0.97	-0.014	0.99
	3Q	0.031	1.03	-0.082*	0.92	-0.025	0.98
	4Q	0.024	1.02	-0.192*	0.83	-0.079*	0.92
Residency status (ref.=etc.)	Overseas Korean	0.163*	1.18	0.065*	1.07	0.049*	1.05
	Low-skilled employment	-0.046	0.95	0.235*	1.26	0.060*	1.06
	Permanent residency	0.219*	1.24	-0.037	0.96	-0.016	0.98
	Student	-0.638*	0.53	-0.157*	0.86	-0.215*	0.81
	Marriage	0.466*	1.59	-0.078*	0.93	0.053*	1.05
Residency period (ref.=under 1 year)	1–3	0.330*	1.39	-0.113*	0.89	-0.409*	0.66
	3–5	0.119*	1.13	-0.065*	0.94	-0.394*	0.67
	5 or more	0.017	1.02	-0.164*	0.85	-0.531*	0.59
Place of residency (ref.=outside)	Metropolitan area	-0.026	0.97	-0.147*	0.86	-0.005	1
Beneficiary type (ref.=employees)	Family members of employees	0.285*	1.33	0.369*	1.45	0.243*	1.28
	Self-employed	-0.013	0.99	0.354*	1.42	0.211*	1.24
	Household members of self-employed	0.258*	1.29	0.379*	1.46	0.279*	1.32
Disability (ref.=none)	Disabled	0.748*	2.11	1.447*	4.25	0.975*	2.65
CCI (ref.=0)	1 or more	0.878*	2.41	0.276*	1.32	0.379*	1.46
Chronic disease (ref.=none)	Present	0.692*	2	0.463*	1.59	0.532*	1.7

* p < 0.0001

Among the predisposing characteristics, females exhibited a higher likelihood of hospitalization compared to males (OR = 1.09), while also experiencing shorter periods (-18%) and paying smaller medical bills (-13%). Compared to the age group younger than 20 years old, individuals aged 20 years and older exhibited a lower likelihood of hospitalization. Additionally, as age increased, medical bills also increased, with this trend being more significant in the age group of 65 years and older.

Compared to individuals of other nationalities, Vietnamese enrollees exhibited a lower likelihood of hospitalization (OR = 0.9), while also experiencing longer periods of hospitalization (+ 7%) and paying larger medical bills (+ 8%). While Taiwanese enrollees did not show a significant likelihood of hospitalization, once hospitalized, they experienced longer periods of hospitalization (+ 58%) and incurred larger medical bills (+ 32%).

Among the enabling characteristics, individuals with a residency status of marriage exhibited a higher likelihood of hospitalization (OR = 1.59), while shorter hospital stays (- 7%) and larger medical bills (+ 5%) compared to other residency statuses. Conversely, individuals with a type of low-skilled employees did not show a significant likelihood of hospitalization; however, once hospitalized, they experienced longer hospital stays (+ 26%) and incurred larger medical bills (+ 6%). Overseas Koreans exhibited a higher likelihood of hospitalization (OR = 1.18), as well as longer hospital stays (+ 7%) and larger medical bills (+ 5%). However, foreign students exhibited a lower likelihood of hospitalization (OR = 0.53), as well as shorter hospital stays (-14%) and smaller medical bills (-19%).

When compared to those with less than one year of residency, individuals who resided one year or more but less than three years showed the highest likelihood of hospitalization (OR = 1.39). Furthermore, the odds ratios tend to decrease as the duration of residency increases. As the duration of residency increases, both the length of hospital stays and the amount of medical bills decrease as well.

Compared to employee beneficiaries, family members of employees showed a higher likelihood of hospitalization (OR = 1.33), with longer hospital stays (+ 45%) and larger medical bills (+ 28%). Similarly, household members of self-employed individuals showed a higher likelihood of hospitalization (OR = 1.29), with longer hospital stays (+ 46%) and larger medical bills (+ 32%). While self-employed individuals did not show a significant likelihood of hospitalization, once hospitalized, they experienced longer periods of hospitalization (+ 42%) and incurred larger medical bills (+ 24%).

Factors related to outpatient services

The results of the two-part model analysis for outpatient services can be found in Table 4. Among the predisposing characteristics, females exhibited a higher likelihood of using outpatient services compared to males (OR = 1.66), with longer periods (+ 30%) and larger medical bills (+ 31%). As age increases, individuals tend to have a lower likelihood of clinic visits compared to the age group younger than 20 years old. Furthermore, while the duration of outpatient services tends to increase with age, it is noteworthy that the longest durations were observed in individuals younger than 20 years old. Compared to individuals younger than 20 years old, those aged 50 to 64 years old tend to spend larger medical bills (7%), while those aged 65 years and older spend even larger amounts (43%).

Table 4.

Result of two-part model analysis of outpatient services

Variable		1st part		2nd part
		Utilization		Count of visits		Costs
		β	odds ratio	β	exp(β)	β	exp(β)
Sex (ref.=male)	Female	0.504*	1.66	0.261*	1.3	0.272*	1.31
Age (ref.=under 20)	20–34	-0.554*	0.57	-0.553*	0.58	-0.018	0.98
	35–49	-0.697*	0.5	-0.545*	0.58	-0.052*	0.95
	50–64	-0.714*	0.49	-0.400*	0.67	0.066*	1.07
	65 or older	-1.255*	0.29	-0.236*	0.79	0.361*	1.43
Nationality (ref.=etc.)	China	0.318*	1.37	0.257*	1.29	0.199*	1.22
	Vietnam	-0.104*	0.9	-0.074*	0.93	-0.053*	0.95
	Uzbekistan	0.051*	1.05	0.088*	1.09	0.123*	1.13
	USA	0.241*	1.27	0.214*	1.24	0.255*	1.29
	Japan	-0.015	0.99	0.083*	1.09	0.132*	1.14
	Canada	0.475*	1.61	0.239*	1.27	0.264*	1.3
	Taiwan	-0.033	0.97	0.217*	1.24	0.186*	1.2
Insurance premium rank (ref.=1Q)	2Q	0.322*	1.38	0.017*	1.02	0.006	1.01
	3Q	0.134*	1.14	0.019*	1.02	0.011	1.01
	4Q	0.274*	1.31	0.041*	1.04	0.045*	1.05
Residency status (ref.=etc.)	Overseas Korean	0.572*	1.77	0.079*	1.08	0.107*	1.11
	Low-skilled employment	-0.051*	0.95	-0.092*	0.91	-0.121*	0.89
	Permanent residency	0.780*	2.18	0.161*	1.17	0.147*	1.16
	Student	-0.273*	0.76	-0.143*	0.87	-0.163*	0.85
	Marriage	1.043*	2.84	0.215*	1.24	0.262*	1.3
Residency period (ref.=under 1 year)	1–3	1.073*	2.92	0.394*	1.48	0.327*	1.39
	3–5	1.075*	2.93	0.409*	1.5	0.300*	1.35
	5 or more	1.237*	3.44	0.478*	1.61	0.349*	1.42
Place of residency (ref.=outside)	Metropolitan area	0.039*	1.04	0.059*	1.06	0.122*	1.13
Beneficiary type (ref.=employees)	Family members of employees	0.114*	1.12	0.070*	1.07	0.130*	1.14
	Self-employed	-0.869*	0.42	-0.053*	0.95	-0.013*	0.99
	Household members of Self-employed	0.021	1.02	0.041*	1.04	0.070*	1.07
Disability (ref.=none)	Disabled	-0.350*	0.7	0.489*	1.63	1.207*	3.34
CCI (ref.=0)	1 or more	2.084*	8.04	0.507*	1.66	0.560*	1.75
Chronic disease (ref.=none)	Present	1.839*	6.29	0.540*	1.72	0.611*	1.84

* p < 0.0001

Compared to individuals of other nationalities, such as etc. nationality (the reference group), Chinese, Uzbekistani, American, and Canadian enrollees exhibited a higher likelihood of using outpatient services, longer duration, and higher medical bills. However, Vietnamese enrollees exhibited a lower likelihood of using outpatient services (OR = 0.9), with a shorter period (-7%) and smaller medical bills (-5%).

Among the enabling characteristics, individuals with higher rankings of insurance premium exhibited a greater likelihood of using outpatient services for longer durations and incurring larger medical bills compared to those with the lowest rankings. Individuals with residency statuses of marriage, overseas Korean, and permanent residency exhibited a higher likelihood of using outpatient services for longer durations and incurring larger medical bills compared to the reference group (etc.). While individuals classified as low-skilled employees and foreign students exhibited a lower likelihood of using outpatient services for shorter durations and incurring smaller medical bills. Individuals residing in Metropolitan areas were more likely to use outpatient services (OR = 1.04) for longer durations (+ 6%) and incur larger medical bills (+ 13%) compared to those residing outside of Metropolitan areas. Compared to employee beneficiaries, family members of employees showed a higher likelihood of using outpatient services (OR = 1.12), with longer durations (+ 7%) and larger medical bills (+ 14%). Meanwhile, self-employed individuals were less likely to use outpatient services (OR = 0.42), with shorter durations (-5%) and smaller medical bills (-1%).

Among the need characteristics, individuals with a CCI greater than 1 and chronic diseases were more likely to use outpatient services for longer durations and incur larger medical bills compared to those without them. While disabled individuals were less likely to use outpatient services (OR = 0.77), once they used them, they tended to use them for longer durations and spend more on medical bills compared to non-disabled individuals.

Discussion

This study investigated factors influencing healthcare utilization, including both inpatient and outpatient services, for foreign beneficiaries of NHI in Korea by employing Andersen’s behavioral model framework. The results showed that need characteristics, such as disability, CCI, and chronic disease, were the most influential factors for both inpatient and outpatient services.

More specifically, individuals aged 65 years and older were less likely to use healthcare services, but once they utilized them, they tended to spend more on medical bills. Senior foreigners appear to face greater barriers in accessing healthcare services compared to the younger generation, as they may experience dual challenges associated with aging and being foreigners. Therefore, as the population of senior foreigners in Korea is expected to grow in the near future [2], access to healthcare services for them needs to be improved so that they can treat their symptoms early enough.

Female foreigners tended to utilize both inpatient and outpatient services more frequently than males, whereas males had longer inpatient stays and higher medical bills. This suggests that females may seek medical attention more proactively, potentially averting serious illnesses. This aligns with prior research indicating higher healthcare utilization among females without an associated increase in medical expenses [8]. Variations in healthcare utilization by sex among foreign employees may stem from differences in workplace environments. Statistics Korea reports that male foreign workers are more prevalent in industries such as mining, manufacturing, and construction, where they face heavier labor burdens and higher risks of work-related injuries and mortality [15] compared to females, who are more commonly employed in wholesale/retail, food, and accommodation sectors [16].

Vietnamese individuals exhibited a lower propensity for using both inpatient and outpatient services; however, once hospitalized, they showed prolonged stays and incurred higher medical expenses. This phenomenon can be attributed to Vietnam’s cultural context, characterized by limited healthcare access, high medical costs, and substantial out-of-pocket payments, particularly after the privatization of Vietnam’s healthcare system [17, 18]. Vietnamese wives residing in Korea tend to visit pharmacies rather than doctor’s office, further indicating cultural influences on healthcare-seeking behavior [19].

The length of residency significantly influences outpatient service utilization among enabling characteristics. Longer residency is associated with increased outpatient service usage and higher medical expenses. Conversely, longer residency is associated with decreased inpatient service utilization, resulting in reduced medical costs. Residency duration reflects foreign residents’ adaptation to Korean culture, with longer residency indicating better integration [20]. Therefore, greater integration correlates with higher healthcare service utilization, facilitating access to outpatient services and potentially preventing severe illness requiring hospitalization.

Residency status such as overseas Koreans, permanent residents, and marriage are associated with increased utilization of outpatient services, longer durations, and higher medical expenses. This trend extends to inpatient services. Permanent residency is typically granted to foreigners residing for at least 5 years with income surpassing a certain threshold. Overseas Korean status is granted to those who acquired foreign nationality after being Korean or to descendants of Korean nationals living abroad, though proficiency in Korean is required for eligibility.

Since the cultural and communication issues have been reported to hinder migrants’ access to healthcare [21], it can be inferred that foreigners may find it easier to use healthcare services when they are well integrated into Korean society. This study’s result supports this idea, as extended residency, elevated income, and proficient language abilities signify enhanced assimilation into Korean culture, leading to greater healthcare access. Marriage residency is awarded to those wedded to Korean nationals or raising children from such marriages, with pregnancy and delivery-related healthcare needs comprising a notable portion of their requirements. Moreover, marriage facilitates foreign spouses’ adaptation and integration into Korean society, streamlining their access to healthcare services.

Low-skilled employees usually exhibit reduced demand for general health check-ups or treatment, primarily due to their youth and overall good health [22]. Similarly, foreign students tend to have lower healthcare service demand as they are typically in their twenties, a life stage associated with decreased healthcare needs across the lifespan [8].

Another notable factor among enabling characteristics was beneficiary type. Family members of employees and household members of self-employed individuals exhibited a heightened likelihood of utilizing both inpatient and outpatient services, leading to extended durations and increased medical expenses. However, family members of employees had a more pronounced effect on duration and medical costs compared to household members of the self-employed. This discrepancy can be attributed to eligibility differences between the two groups: household members of the self-employed are restricted to spouses and underage children, whereas family members of employees can include parents and siblings.

While this study delved into factors impacting healthcare utilization among foreign beneficiaries using Andersen’s behavioral model framework, there are several limitations to acknowledge. Firstly, certain crucial factors such as Korean proficiency and social network were not accounted for in the insurance claims data. Secondly, the analysis was limited to a one-year timeframe, preventing a comprehensive understanding of utilization patterns. However, this cross-sectional study was the most suitable option given the impact of the COVID-19 pandemic following the implementation of mandatory enrollment in 2019. Although foreigners were required to enroll in the NHI program in 2019, the strict quarantine policies during the pandemic significantly reduced the number of incoming foreigners, making it difficult to design a longitudinal study. Thirdly, because self-employed foreign beneficiaries have a six-month delay before enrolling in the NHI, short-term stay foreigners were not included in this study population.

Conclusion

It is worth noting that this cross-sectional study utilized comprehensive administrative data from the implementation of the mandatory enrollment policy for foreigners residing for at least six months, thereby representing the entire population of foreign beneficiaries in 2022. Additionally, by separately examining the likelihood and extent of healthcare service utilization, the results offer valuable insights into the healthcare behavior of foreigners, particularly given the expected growth of the foreign population in Korea.

The health insurance policy for foreigners in Korea shifted from voluntary enrollment for those residing for three months or longer to mandatory enrollment for a six-month stay, addressing issues of adverse selection and fiscal deficit. However, following this change, some self-employed foreigners may face unmet healthcare needs during the six-month waiting period. Therefore, it is essential to consider measures to improve access to healthcare services for these individuals. Additionally, further research is needed to analyze whether foreigners use more or fewer healthcare services compared to the amount they contribute to the NHI. Based on these significant findings, the Korean government will be better positioned to discuss foreign health insurance enrollment. This study could serve as a foundation for future research on this topic.

Abbreviations

NHI

National Health Insurance

NHIS

National Health Insurance Service

GNI

Gross National Income

CCI

Charlson Comorbidity Index

SD

Standard deviation

IQR

Interquartile range

OR

Odds ratio

Authors’ contributions

J.L. and H.G. both contributed equally to conceiving the study, analyzing the data, and writing this paper. All authors reviewed the manuscript.

Funding

None.

Availability of data and materials

The data that support the findings of this study are available from National Health Insurance Service – National Health Information Database (NHIS – NHID), but restrictions apply to the availability of these data, which were used under license for the current study and so are not publicly available. The data are, however, available from the authors upon reasonable request and with the permission of NHIS.

Declarations

Ethics approval and consent to participate

Approval for this research was obtained from the Institutional Review Board of National Health Insurance Service, approval number 2023-HR-06-001. Given that the study involved secondary data analysis with all personal identifiers removed, the need for informed consent was waived. The study methods were performed in full compliance with the relevant ethical standards and guidelines.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.