==== Front BMC Emerg Med BMC Emerg Med BMC Emergency Medicine 1471-227X BioMed Central London 37380961 808 10.1186/s12873-023-00808-8 Research Factors related to the frequent use of emergency department services in Korea Cho Eun Deok ed.cho@nmc.or.kr 1 Kim Bomgyeol arasion12@gmail.com 2 Kim Do Hee rhkdcjf6557@gmail.com 2 Lee Sang Gyu LEEVAN@yuhs.ac 3 Jang Suk-Yong SUKYONG@yuhs.ac 4 Kim Tae Hyun THKIM@yuhs.ac 4 1 grid.415619.e 0000 0004 1773 6903 National Emergency Medical Center, National Medical Center, 245 Eulji-Ro, Jung-Gu, Seoul, 04564 Republic of Korea 2 grid.15444.30 0000 0004 0470 5454 Department of Public Health, Graduate School, Yonsei University, 50-1 Yonsei-Ro, Seodaemun-Gu, Seoul, 03722 Republic of Korea 3 grid.15444.30 0000 0004 0470 5454 Department of Preventive Medicine, College of Medicine, Yonsei University, 50-1 Yonsei-Ro, Seodaemun-Gu, Seoul, 03722 Korea 4 grid.15444.30 0000 0004 0470 5454 Department of Healthcare Management, Graduate School of Public Health, Yonsei University, 50-1 Yonsei-Ro, Seodaemun-Gu, Seoul, 03722 Korea 29 6 2023 29 6 2023 2023 23 7318 9 2022 17 3 2023 © The Author(s) 2023 https://creativecommons.org/licenses/by/4.0/ Open AccessThis article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Background Frequent Emergency Department (ED) visitors are identified by the policymakers to reduce avoidable ED visits and lessen the financial and operational burden. This study aimed to identify the factors related to the frequent use of ED services. Methods This nationwide, cross-sectional observational study was conducted using information obtained from the 2019 National Emergency Department Information System (NEDIS) database. Frequent ED users were defined as patients with four or more ED visits a year. We performed multiple logistic regression analyses to verify the relationship among sociodemographic characteristics, residential characteristics, clinical characteristics, and frequency of ED visits. Results Among 4,063,640 selected patients, 137,608 patients visited the ED four or more times a year (total number of visits = 735,502 times), which accounted for 3.4% and 12.8% of the total number of ED users and ED visits, respectively. A high ED visit frequency was associated with male sex, age < 9 or ≥ 70 years, Medical Aid (based on the insurance type), lower number of medical institutions and beds compared with that of the national average, and conditions, such as cancer, diabetes, renal failure, and mental illness. A low ED-visit frequency was associated with residence in regions vulnerable to emergency medical care and regions with high income. The possibility of frequent ED visits was high for patients with level 5 severity (non-emergent) and those with an increased need for medical treatment, including older patients and patients with cancer or mental illness. The possibility of frequent ED visits was low for patients aged > 19 years with level 1 severity (resuscitation). Conclusions Health service accessibility factors, including low income and medical resource imbalance, were associated with frequent ED visits. Future large-scale prospective cohort studies are warranted to establish an efficient emergency medical system. Keywords Emergency medical center Emergency department Frequent users Emergency service Regional differences issue-copyright-statement© BioMed Central Ltd., part of Springer Nature 2023 ==== Body pmcBackground Frequentusers of emergency department (ED) services have been generating interest in recent years [1]. Internationally, frequent ED users comprise 3–8% of all the patients visiting the ED and 67% of all ED visits over a given period (usually 1 year) [2]. In Korea, 3.1% of ED visitors were identified as frequent ED users, denoting that these patients visited EDs more than four times per year, and such visits accounted for 14% of all ED visits in 2009 [3]. Frequent ED visits lead to substantial healthcare costs [2, 4]. Moreover, they decrease ED efficiency, contribute to ED overcrowding, and can result in the redirection of services from urgent cases [2, 5]. Therefore, policymakers and researchers have been trying to find ways to improve relevant services to individuals who must use the emergency care system—not those who use EDs by choice—at a higher rate [6]. Frequent ED users may receive a suboptimal quality of care since the care provided may be fragmented, episodic, and poorly coordinated [2, 7]. Additionally, physicians may be biased and less empathetic toward frequent ED users [8]. The frequent use of ED services may sometimes be inappropriate and non-urgent [9]. Accordingly, the uncoordinated acute care received by frequent ED users can be less effective compared with the effectiveness of typical ED or primary care [2]. Several studies have been conducted on frequent ED visits worldwide. Cross-sectional studies have demonstrated that sociodemographic, clinical, and health system-level factors contribute to frequent ED visits [10]. Specifically, minority race, low educational attainment, low income, public insurance, usual sources of outpatient medical care (other than the ED), high usage of outpatient health care resources, and poor physical and mental health are associated with an increased probability of frequent ED use [3, 11–13]. Predicting and identifying the frequent ED users could help formulate target interventions for addressing unmet health and social needs while simultaneously reducing ED use [1]. The characteristics of frequent ED users may vary according to the country as well as the location, size, and role of a hospital [12]. However, most studies have used medical records or sample data from one ED, with few studies using data from multiple EDs [12, 14–16]. To the best of our knowledge, there have been no national population-based studies on factors related to frequent ED use in Korea. Accordingly, there is a need for multi-regional and multi-departmental studies to identify the characteristics and relevant factors related to frequent ED users. Therefore, we aimed to identify the characteristics of frequent ED users at the emergency medical center level and factors related to their frequent visits based on nationally representative and population-based data in Korea. Methods Aim of the study To identify the characteristics of frequent ED users at the emergency medical center level and factors related to their frequent visits based on nationally representative and population-based data in Korea. Study design and setting This nationwide, cross-sectional observational study was conducted using information obtained from the National Emergency Department Information System (NEDIS) database, which is managed by the Ministry of Health and Welfare and comprises nationwide data on ED visitors in Korea. Participants and data source We used NEDIS data collected from January 2019 to December 2019. The NEDIS contains data regarding patient demographics and clinical information, including age, sex, visit route, the Korean Triage and Acuity Scale (KTAS) level, vital signs, discharge outcomes, and diagnosis in ED [17]. The final analysis dataset was created by merging the public information from areas underserved by emergency services. Emergency medical centers in Korea are categorized into four types: regional emergency medical centers, local emergency medical centers, regional emergency medical institutions, and specialized emergency centers [18]. The structure of the system is designed to be a sequential emergency medical delivery system where the most severe emergency patients are treated in regional centers, while moderate or mild emergency patients are treated in local centers and institutions [18]. Moreover, under the structure, special emergency diseases are handled by specialized emergency centers (e.g., children, poisoning, burns) providing applicable treatment [18]. Since 2016, relevant authorities have made considerable efforts so that the system as a whole can achieve the goal of regionalization of emergency medical care [19]. Regional emergency medical centers strive to improve health outcomes for emergency patients by making better use of resources within the region [20]. Meanwhile, local emergency medical centers—usually with 35 beds and serving approximately 30,000 patients per year—aim to provide emergency services to patients living in rural or remote areas [21]. Areas with underserved emergency medical services are those in which more than 30% of the local population cannot reach the local emergency medical center within 30 min or the regional emergency medical center within 1 h. These locations were identified as those revised and promulgated by the Ministry of Health and Welfare of Korea in 2019 under Article 12, paragraphs 2 and 3 of the Public Health and Medical Services Act [22]. We included patients who visited one or more of the EDs of 38 and 124 regional and local emergency medical centers, respectively, which mainly provide medical treatment for emergency patients and have similar variables registered in the NEDIS database. We included 4,063,640 ED users after excluding 183,010 cases with missing patient residence or unknown data. Variables and measurements Dependent variables Frequent ED users are patients who visit the ED on multiple occasions [16]. We categorized patients according to their ED utilization level. We defined frequent ED users as patients with four or more ED visits a year. There is currently no established definition of high utilization; however, ≥ 4 ED visits in 1 year is a commonly used threshold [11, 13, 16]. However, given the nature of the data constructed for each institution, cases where the same patients who visited another ED may have been omitted. Independent variables Sociodemographic characteristics: Sociodemographic characteristics included sex (male and female), age (0–70 years), place of residence (Seoul, Busan, Daegu, Incheon, Gwangju, Daejeon, Ulsan, Sejong, Gyeonggi, Gangwon, Chungcheong, Jeolla, Gyeongsang, and Jeju), and insurance type (National Health Insurance [NHI; regional + employer-provided], automobile insurance, industrial accident insurance, Medical Aid [Types 1 and 2], general insurance, or others). To note, the names of the regions are original names under the Korean official administrative regional division. South Korea’s health insurance system is a public and single-payer system. With the enactment of the NHI Act in 2000, all insurers were integrated under a single insurer [23]. The NHI covers 97% of the population, and the remaining 3% is covered by the Medical Aid program [24]. Unlike NHI and Medical Aid, which are provided to all Koreans through the government, automobile insurance is a private insurance service that automobile owners must subscribe to for vehicle accident coverage [25]. Industrial accident insurance provides prompt and fair compensation for employees affected by occupational accidents through industrial accident compensation insurers [26]. Place of residence characteristics: These characteristics included information regarding the status of emergency medical services (underserved or not), the number of emergency medical institutions and number of beds available in the area, and the regional income decile. Areas lacking emergency medical services were identified from those announced by the Ministry of Health and Welfare [19, 27]. Regarding the number of emergency medical institutions and beds, data were categorized using the following criteria: the number of emergency medical institutions per million residents in the city (Si), county (Gun), and district (Gu) of residence, and whether the number of beds in institutions above the hospital-clinic level was above or below that of the national average. Regarding regional income deciles, we divided the average earned income per resident of the city (Si), county (Gun), and district (Gu) into 10 deciles, from the lowest income level to the highest. Practice-related characteristics: These variables included the presence or absence of five diseases (cancer, high blood pressure, diabetes, renal failure, and mental illness) and the results of acuity classification. Disease prevalence was classified into present or absent based on whether the disease was characterized by at least 20 main symptoms presented in the NEDIS database, based on the Korean Standard Classification of Diseases-7 code. The result of the acuity classification was processed using the KTAS level [28]. The KTAS level is divided into five levels, with the lower levels indicating higher clinical severity, as follows: level 1 necessitates top priority for care and indicates life-threatening conditions, including cardiac arrest, severe respiratory failure, and loss of consciousness, requiring immediate treatment; level 2 indicates potentially life-threatening conditions, including myocardial infarction, cerebral hemorrhage, and cerebral infarction, requiring rapid treatment; level 3 indicates conditions that can eventually progress to cause serious complications; level 4 represents conditions that require treatment or reassessment within 1 to 2 h and is associated with age, pain level, and the likelihood of complications; and level 5 indicates an urgent but non-emergent condition attributable to a chronic problem or condition that is unlikely to worsen. Statistical analysis A frequency analysis was conducted to identify the status of ED visits. The results of this analysis are presented as frequencies and percentages. A chi-square test was used to examine the distribution of general characteristics according to the frequency of ED use. Multiple logistic regression was performed to examine the factors related to frequent ED use based on odds ratios (ORs) with 95% confidence intervals (CIs). All statistical analyses were performed using SAS software (version 9.4; SAS Institute Inc., Cary, NC, USA). A p-value < 0.05 was considered statistically significant. Institutional review board waiver statement This study adhered to the Declaration of Helsinki guidelines and was reviewed by the Severance Hospital Institutional Review Board (IRB number: 4–2021-0491). The requirement of written informed consent was waived by the Severance Hospital Institutional Review Board since the study used secondary anonymized data. Results Status of ED visits Figure 1 shows the status of ED visits and percentage of visits. Among the total visits, 3,050,671 patients visited the same ED once, which accounted for 75.1% and 53.2% of the total number of ED users and ED visits, respectively. Further, 875,361 patients visited two to three times (total number of visits = 1,953,498 times), which accounted for 21.5% and 34.0% of the total number of ED users and ED visits, respectively. Moreover, 137,608 patients visited the ED four or more times a year (total number of visits = 735,502 times), which accounted for 3.4% and 12.8% of the total number of ED users and ED visits, respectively (Fig. 2).Fig. 1 Emergency department visits and percentage of visits Fig. 2 Status of emergency department visits. ED, emergency department Characteristics of patients who visited EDs The patient characteristics are shown in Table 1. When it is examined according to sex and age groups, there were more male than female patients among frequent ED users. Moreover, patients aged > 70 years (4.7%) were more likely to be frequent ED users, followed by those aged < 9 years (3.9%). Regarding the insurance type, the likelihood of frequent ED use was the highest among patients receiving Medical Aid coverage (7.3%).Table 1 Comparison between frequent and non-frequent ED users Variables Total Frequent ED Users Non-Frequent ED Users p N % N % N % Total 4,063,640 100.0 3,926,032 96.6 137,608 3.4 Sex  < .0001  Male 2,002,672 49.3 1,937,177 96.7 65,495 3.3  Female 2,060,968 50.7 1,988,855 96.5 72,113 3.5 Age (years)  < .0001   ≤ 9 674,626 16.6 648,430 96.1 26,196 3.9  10–19 292,998 7.2 286,278 97.7 6,720 2.3  20–29 457,637 11.3 446,438 97.6 11,199 2.4  30–39 447,022 11.0 435,239 97.4 11,783 2.6  40–49 459,047 11.3 445,425 97.0 13,622 3.0  50–59 565,865 13.9 547,863 96.8 18,002 3.2  60–69 478,823 11.8 461,385 96.4 17,438 3.6   ≥ 70 687,622 16.9 654,974 95.3 32,648 4.7 Region  < .0001  Seoul 829,176 20.4 802,376 96.8 26,800 3.2  Busan 176,214 4.3 171,217 97.2 4,997 2.8  Daegu 152,200 3.7 148,412 97.5 3,788 2.5  Incheon 247,928 6.1 238,239 96.1 9,689 3.9  Gwangju 100,829 2.5 96,637 95.8 4,192 4.2  Daejeon 105,453 2.6 100,992 95.8 4,461 4.2  Ulsan 61,968 1.5 60,629 97.8 1,339 2.2  Sejong 11,554 0.3 11,229 97.2 325 2.8  Gyeonggi 1,133,162 27.9 1,096,411 96.8 36,751 3.2  Gangwon 138,037 3.4 133,738 96.9 4,299 3.1  Chungcheong 342,784 8.4 332,349 97.0 10,435 3.0  Jeolla 264,711 6.5 253,129 95.6 11,582 4.4  Gyeongsang 406,565 10.0 391,352 96.3 15,213 3.7  Jeju 93,059 2.3 89,322 96.0 3,737 4.0 Insurance types  < .0001  National Health Insurance 3,622,424 89.1 3,503,781 96.7 118,643 3.3  Auto insurance 162,973 4.0 159,746 98.0 3,227 2.0  Industrial accident insurance 8,556 0.2 8,334 97.4 222 2.6  Medical Aid (Types 1 and 2) 191,451 4.7 177,449 92.7 14,002 7.3  General insurance 59,510 1.5 58,458 98.2 1,052 1.8  Other insurance 18,726 0.5 18,264 97.5 462 2.5 Residence in underserved emergency Medical Services areas  < .0001  No 3,630,066 89.3 3,506,874 96.6 123,192 3.4  Yes 433,574 10.7 419,158 96.7 14,416 3.3 Number of emergency medical institutions  < .0001  Above the national average 1,306,744 32.2 1,259,050 96.4 47,694 3.6  Below the national average 2,756,896 67.8 2,666,982 96.7 89,914 3.3 Number of beds  < .0001  Above the national average 1,408,186 34.7 1,358,076 96.4 50,110 3.6  Below the national average 2,655,454 65.3 2,567,956 96.7 87,498 3.3 Regional income decile  < .0001  1st quartile (lowest) 405,600 10.0 392,447 96.8 13,153 3.2  2nd quartile 398,239 9.8 382,396 96.0 15,843 4.0  3rd quartile 395,879 9.7 382,380 96.6 13,499 3.4  4th quartile 416,699 10.3 401,957 96.5 14,742 3.5  5th quartile 428,062 10.5 410,577 95.9 17,485 4.1  6th quartile 372,942 9.2 359,150 96.3 13,792 3.7  7th quartile 433,250 10.7 419,650 96.9 13,600 3.1  8th quartile 371,584 9.1 359,746 96.8 11,838 3.2  9th quartile 439,894 10.8 427,256 97.1 12,638 2.9  10th quartile (highest) 401,491 9.9 390,473 97.3 11,018 2.7 Cancer  < .0001  No 3,929,064 96.7 3,805,592 96.9 123,472 3.1  Yes 134,576 3.3 120,440 89.5 14,136 10.5 Hypertension  < .0001  No 4,036,995 99.3 3,900,545 96.6 136,450 3.4  Yes 26,645 0.7 25,487 95.7 1,158 4.3 Diabetes  < .0001  No 4,035,224 99.3 3,899,176 96.6 136,048 3.4  Yes 28,416 0.7 26,856 94.5 1,560 5.5 Renal failure  < .0001  No 4,015,410 98.8 3,881,705 96.7 133,705 3.3  Yes 48,230 1.2 44,327 91.9 3,903 8.1 Mental illness  < .0001  No 3,996,542 98.3 3,862,282 96.6 134,260 3.4  Yes 67,098 1.7 63,750 95.0 3,348 5.0 Korean Triage and Acuity Scale  < .0001  Level 1 (Resuscitation) 42,374 1.0 40,698 96.0 1,676 4.0  Level 2 (Emergent) 216,409 5.3 208,062 96.1 8,347 3.9  Level 3 (Urgent) 1,493,106 36.7 1,438,032 96.3 55,074 3.7  Level 4 (Less urgent) 1,967,646 48.4 1,913,001 97.2 54,645 2.8  Level 5 (Non-urgent) 344,105 8.5 326,239 94.8 17,866 5.2 Statistically significant at p-value < 0.05 (shown in bold) ED Emergency department When the result was viewed by region, 3.4% and 3.3% of the patients who were non-residents and residents, respectively, of areas lacking emergency medical services were frequent ED users. Regarding the number of emergency medical institutions and beds, there was a higher likelihood of frequent ED visits when the numbers were higher than that of the national average (3.6%) compared with when the number was below. In terms of regional income deciles, among patients living in areas with low-income deciles, 2.7% and 2.9% of patients in the 10th and 9th deciles, respectively, were frequent ED users. According to the type of chronic diseases, patients with chronic diseases were more likely to be frequent ED users. The highest frequency of ED use was found among patients with cancer (10.5%), hypertension (4.3%), diabetes (5.5%), renal failure (8.1%), and mental illness (5.0%). Additionally, the likelihood of frequent ED use was the highest in patients with level 5 acuity (non-emergency) with the lowest severity. Factors related to frequent ED use Table 2 shows the results of the multiple logistic regression analysis of factors related to frequent ED use. According to the sex and age groups, males and patients aged < 9 years had a higher likelihood of being frequent ED users than females and other age groups. Patients residing in the non-capital areas had a higher range (OR 1.06–1.67) use of ED than those residing in other parts of Korea. Regarding the insurance type, patients with Medical Aid coverage were more likely to be frequent ED users than those with the NHI.Table 2 Multiple logistic regression analysis of factors associated with frequent ED use Variables OR 95% CI p Sex  Female ref  Male 1.06 1.05 1.08  < .0001 Age (years)   ≤ 9 ref  10–19 0.55 0.54 0.57  < .0001  20–29 0.60 0.59 0.61  < .0001  30–39 0.64 0.62 0.65  < .0001  40–49 0.67 0.65 0.68  < .0001  50–59 0.68 0.67 0.69  < .0001  60–69 0.74 0.72 0.75  < .0001   ≥ 70 0.94 0.93 0.96  < .0001 Region  Seoul ref  Busan 0.96 0.91 1.01 0.1450  Daegu 0.77 0.74 0.80  < .0001  Incheon 1.25 1.22 1.29  < .0001  Gwangju 1.36 1.28 1.44  < .0001  Daejeon 1.33 1.27 1.38  < .0001  Ulsan 0.74 0.70 0.79  < .0001  Sejong 0.99 0.88 1.11 0.8798  Gyeonggi 1.06 1.04 1.08  < .0001  Gangwon 1.04 0.98 1.11 0.1867  Chungcheong 1.16 1.12 1.20  < .0001  Jeolla 1.57 1.48 1.65  < .0001  Gyeongsang 1.44 1.36 1.51  < .0001  Jeju 1.67 1.57 1.78  < .0001 Insurance types  National Health Insurance ref  Auto insurance 0.70 0.68 0.73  < .0001  Industrial accident insurance 0.90 0.79 1.03 0.1312  Medical Aid (Types 1 and 2) 2.12 2.08 2.16  < .0001  General insurance 0.60 0.57 0.64  < .0001  Other insurance 0.84 0.77 0.92 0.0002 Residence in Underserved Emergency Medical Services areas  No ref  Yes 0.79 0.77 0.81  < .0001 Number of emergency medical institutions  Above the national average ref  Below the national average 1.20 1.15 1.25  < .0001 Number of beds  Above the national average ref  Below the national average 1.04 1.01 1.06 0.0015 Regional income decile  1st quartile (lowest) ref  2nd quartile 1.25 1.22 1.28  < .0001  3rd quartile 1.07 1.05 1.10  < .0001  4th quartile 1.12 1.09 1.15  < .0001  5th quartile 1.37 1.33 1.40  < .0001  6th quartile 1.22 1.19 1.25  < .0001  7th quartile 1.01 0.98 1.04 0.4939  8th quartile 1.05 1.02 1.08 0.0006  9th quartile 0.95 0.93 0.98 0.0002  10th quartile (highest) 0.95 0.92 0.98 0.0002 Cancer  No ref  Yes 3.33 3.26 3.39  < .0001 Hypertension  No ref  Yes 0.87 0.82 0.93  < .0001 Diabetes  No ref  Yes 1.17 1.11 1.24  < .0001 Renal failure  No ref  Yes 1.91 1.84 1.98  < .0001 Mental illness  No ref  Yes 1.60 1.54 1.66  < .0001 Korean Triage and Acuity Scale  Level 1 (Resuscitation) ref  Level 2 (Emergent) 1.04 0.99 1.10 0.1161  Level 3 (Urgent) 1.08 1.03 1.14 0.0029  Level 4 (Less urgent) 0.92 0.88 0.97 0.0013  Level 5 (Non-urgent) 1.78 1.69 1.88  < .0001 Statistically significant at p-value < 0.05 (shown in bold) OR Odds ratio, CI Confidence interval, Ref Reference In terms of the insurance type, there was a lower likelihood of frequent ED visits among residents than among non-residents of areas lacking emergency medical services. Regarding the number of emergency medical institutions and beds, there was a lower likelihood of frequent ED visits when the numbers were less than the national average. Regarding the regional income decile, compared with the first income decile, the second, third, fourth, fifth, sixth, seventh, and eighth deciles had higher likelihoods of frequent ED visits, while the ninth and tenth deciles had lower likelihoods of frequent ED visits. Among them, the seventh decile was not statistically significant. There was relatively high probability of frequent ED visits among patients with cancer, diabetes, renal failure, and mental illness; contrastingly, patients with hypertension had a relatively low probability of frequent ED visits. Regarding the severity classification results, there was higher likelihood of frequent ED visits for level 2 (severe; OR = 1.04), level 3 (emergency), and level 5 (non-emergency) compared with level 1 (resuscitation), while level 4 (quasi-emergency) had lower likelihood of frequent ED visits. Among them, level 2 was not statistically significant. Discussion This study investigated the characteristics and relevant factors related to frequent ED visits (≥ 4 ED visits in a year). We found that among all patients who visited the ED in 2019, 3.4% of them were frequent ED users, which accounted for 12.8% of the total number of ED visits. Among studies conducted outside Korea, frequent ED users accounted for 3.5%–4.5% and 13.9%–18.1% of the total number of ED visitors and visits, respectively [12, 29]. In Korean studies, frequent ED users accounted for 2.7%–3.1% and 11.9%–14.0% of the total number of ED visitors and visits, respectively, which is consistent with our findings [3, 12]. Our findings contribute to the emerging literature on frequent ED users in Korea. First, we observed that patients with Medical Aid coverage showed a higher frequency of ED use than those with NHI coverage, which is consistent with previous reports that frequent ED use has various causes, including relatively low health and access to medical care as well as moral hazards [12, 30]. Patients with Medical Aid coverage have both poor physical health and low socioeconomic status [9]. Since these patients appear to use the ED as an alternative to other sources of primary care, they could use multiple medical resources, including the ED, to address their unmet healthcare needs [16]. Second, Korea has policies for reducing the burden of medical expenses on patients who live in areas without emergency medical services. Health insurance coverage to emergency management fees for non-emergency patients at the same level as emergency patients is applied when these patients visit EDs without symptoms to request emergency medical treatment [31]. Underserved emergency medical service areas are defined as areas where more than 30% of the local population cannot access the local emergency medical center within 30 min or the regional emergency medical center within 1 h. Nonetheless, patients living in areas lacking emergency medical services were less likely to be frequent ED users than patients residing outside such areas. Our findings suggest that living in an area underserved by emergency medical services is a greater limitation to ED use compared with other factors, such as age and income. Furthermore, given the reported low relevance index of areas lacking emergency medical services, there is a need to elucidate ED use in regions to establish emergency medical service provision systems within regions [32] and identify areas lacking emergency medical services. In addition, the research results should be utilized as foundational data for regional health care plans, such as public health care plans for medically vulnerable areas, to devise necessary policies for each region. Third, we examined the relationship of frequent ED visits with the prevalence of cancer, hypertension, diabetes, renal failure, and mental illness. Patients with cancer were found using the ED frequently, which can be attributed to ED visits for continuous pain relief treatment [12, 33]. Patients with diabetes and renal failure have a high tendency to visit the ED due to acute exacerbation of chronic diseases [3, 17, 34], with patients with renal failure showing a relatively high number of ED visits due to renal dialysis [35]. Chronic diseases are conditions for which adequate ambulatory care can prevent deterioration or complications requiring ED visits or hospitalizations [36]. Primary care should be organized to meet the needs of patients with chronic diseases and a high illness burden [36]. Moreover, patients with mental illness showed a relatively high likelihood of frequent ED visits. This is consistent with previous reports showing that frequent ED use is associated with drug use, alcohol addiction, depression, self-harm, and suicide [3, 17, 37]. We observed no frequent visits among patients with hypertension, which is inconsistent with previous studies [13, 16]. There is an underestimated prevalence of diseases measured by vital signs, including blood pressure, since the NEDIS does not record medical history, such as medical records, which should be considered when interpreting our findings. Finally, there have been inconsistent reports regarding the acuity of frequent ED users. For example, Vinton et al., [13] Moore et al., [11] and Han et al. [38] reported that the health status of frequent ED users was poor, and the acuity was higher in frequent ED users than in non-frequent ED users. Contrarily, Choe et al., [14] Shin et al., [12] and Uscher-Pines et al. [39] reported no significant difference in acuity between frequent and non-frequent ED users. In our study, there was a higher probability of frequent ED visits in level 5 cases (non-emergency) than in level 1 cases (resuscitation). This suggests that frequent ED visits are contributing to inefficiencies in the medical system, including increasing medical costs and overcrowding, which have been consistently identified in some Korean studies [12, 40]. Limitations This study has several limitations. First, since we only used NEDIS data, we only included EDs at the emergency medical center level or higher. However, the NEDIS database is an emergency medical data registration system that is commonly adopted by emergency medical institutions nationwide. It collects medical treatment data from EDs across Korea without restrictions on hospitals and insurance types to enhance the quality of emergency medical services and provide basic data for informing policy and decision-making. Accordingly, it is an excellent source of data with government-managed quality. Second, frequent ED use was defined as a patient using the same ED multiple times since we used patient registration numbers registered for each ED. Third, the prevalence of underlying diseases, including hypertension and diabetes, could have been underestimated since the NEDIS does not provide patient history data, including medical records. Finally, since this was a retrospective cross-sectional study based on 1-year medical data, it cannot demonstrate a causal relationship. Nonetheless, this study has important significance and implications. Specifically, since we used nationwide data recorded in the NEDIS from multiple EDs, it resolves the limitation of existing studies on data from a single ED. Conclusion We found that patients with high medical needs, including those with cancer, chronic disease, and mental illness, were more likely to visit the ED multiple times. Additionally, factors regarding medical access, including low income and disparity in medical resources across regions, were associated with frequent ED use. Future large-scale prospective cohort studies are warranted to establish an efficient emergency medical system because there is variability in geographic, socioeconomic, individual, clinical, and medical-systemic differences. Such studies will achieve a higher level of relevance in emergency medical resource use, the performance of life-saving interventions, admission rates, and mortality in EDs according to the needs of each patient. Abbreviations ED Emergency department NEDIS National Emergency Department Information System KTAS Korean Triage and Acuity Scale NHI National Health Insurance OR Odds ratio CI Confidence interval Acknowledgements Not applicable. Authors’ contributions EDC, and THK conceptualized and designed this study. EDC performed data acquisition. BK and S-YJ performed the statistical analyses. EDC, and BK drafted the manuscript. All authors finalized the manuscript. SGL, and THK supervised the entire process. The authors read and approved the final manuscript. Funding No funding was received for conducting this study. Availability of data and materials The data that support the findings of this study are available from National Emergency Medical Center under the Ministry of Health and Welfare in Korea, which were used under license for the current study, and so are not publicly available. Informed consent was waived because of the retrospective nature of the study. The datasets which were analyzed during the current study are available from the corresponding author [Tae Hyun Kim] on reasonable request at THKIM@yuhs.ac. Declarations Ethics approval and consent to participate All procedures contributing to this work comply with the ethical standards of the relevant national and institutional guidelines on human experimentation (Finland) and with the Helsinki Declaration of 1975, as revised in 2008. Permission from the Institutional Review Board committee of the Severance Hospital was obtained (IRB number: 4–2021-0491). The IRB committee waived the individual consents of the patients in our data analysis because this study used a method of retrospective data analysis where data were collected in the ED for patient care and administrative purposes. During the data collection process, all information identifiable to the patient was deleted and the collected data were used for research purposes only. The requirement of written informed consent was waived since the study used secondary anonymized data. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Eun Deok Cho and Bomgyeol Kim contributed equally to this work. ==== Refs References 1. Kanzaria HK Niedzwiecki MJ Montoy JC Raven MC Hsia RY Persistent frequent emergency department use: core group exhibits extreme levels of use for more than decade Health Aff (Millwood) 2017 36 1720 1728 10.1377/hlthaff.2017.0658 28971916 2. Krieg C Hudon C Chouinard MC Dufour I Individual predictors of frequent emergency department use: scoping review BMC Health Serv Res 2016 16 594 10.1186/s12913-016-1852-1 27765045 3. Woo JH Grinspan Z Shapiro J Rhee SY Frequent users of hospital emergency departments in Korea characterized by claims data from the national health insurance: cross sectional study PLoS ONE 2016 11 e0147450 10.1371/journal.pone.0147450 26809051 4. Sandoval E Smith S Walter J Schuman SA Olson MP Striefler R comparison of frequent and infrequent visitors to an urban emergency department J Emerg Med 2010 38 115 121 10.1016/j.jemermed.2007.09.042 18462906 5. Lee A, Lau FL, Hazlett CB, Kam CW, Wong P, Wong TW, et al. Measuring the inappropriate utilization of accident and emergency services? Int J Health Care Qual Assur Inc Leadersh Health Serv. 1999;12:287–92. 6. Kanzaria HK Niedzwiecki M Cawley CL Chapman C Sabbagh SH Riggs E Frequent emergency department users: focusing solely on medical utilization misses the whole person Health Aff (Millwood) 2019 38 1866 1875 10.1377/hlthaff.2019.00082 31682499 7. Ettinger WH Casani JA Coon PJ Muller DC Piazza-Appel K Patterns of use of the emergency department by elderly patients J Gerontol 1987 42 638 642 10.1093/geronj/42.6.638 3680884 8. Survey: ED physicians report burnout, desire help for dealing with frequent users. ED Manag. 2011;23(9):104–5. https://pubmed.ncbi.nlm.nih.gov/21916320/. 9. Ruger JP Richter CJ Spitznagel EL Lewis LM Analysis of costs, length of stay, and utilization of emergency department services by frequent users: implications for health policy Acad Emerg Med 2004 11 1311 1317 10.1197/j.aem.2004.07.008 15576522 10. Colligan EM Pines JM Colantuoni E Howell B Wolff JL Risk factors for persistent frequent emergency department use in Medicare beneficiaries Ann Emerg Med 2016 67 721 729 10.1016/j.annemergmed.2016.01.033 26947801 11. Moore L Deehan Seed P Jones R Characteristics of frequent attenders in an emergency department: analysis of 1-year attendance data Emerg Med. 2009 26 263 7 12. Shin TG Song JW Song HG Hong CK Characteristics of frequent users of emergency department J Korean Soc Emerg Med 2011 22 86 92 13. Vinton DT Capp R Rooks SP Abbott JT Ginde AA Frequent users of US emergency departments: characteristics and opportunities for intervention Emerg Med 2014 31 526 532 14. Choe MSP Seo KS Kam S Seo JS Lee JH Seol DH Clinical analysis of frequent attenders of emergency department J Korean Soc Emerg Med 2003 14 560 568 15. Locker TE Baston S Mason SM Nicholl J Defining frequent use of an urban emergency department Emerg Med 2007 24 398 401 16. Lee JH Park GJ Kim SC Kim H Lee SW Characteristics of frequent adult emergency department users: A Korean tertiary hospital observational study Medicine (Baltimore) 2020 99 e20123 10.1097/MD.0000000000020123 32358403 17. Kim S Kang H Cho Y Lee H Lee SW Jeong J Emergency department utilization and risk factors for mortality in older patients: an analysis of Korean National Emergency Department Information System data Clin Exp Emerg Med 2021 8 128 136 10.15441/ceem.20.098 34237818 18. Han KS Kim WY Kim SJ Jeong J Kang H Lee C Research for improvement of the national evaluation program for emergency medical center in Korea J Korean Med Assoc 2020 63 227 234 10.5124/jkma.2020.63.4.227 19. Ministry of Health and Welfare The 3rd National Emergency Medical Plan (2018–2022) 2019 Sejong Ministry of Health and Welfare 20. Kim Y Yeom S Ryu J Jeon YJ Regionalization of emergency medical system and re-establishment of regional emergency medical plan J Korean Soc Emerg Med 2020 31 255 258 10.3346/jkms.2020.35.e255 21. Lee GW Lee KW Jang TC Kim GM Seo YW Ko SH Relevance of emergency level assessment by the Korean Triage and Acuity Scale for adult patients in a local emergency medical center J Korean Soc Emerg Med 2018 29 595 602 22. Ministry of Health and Welfare. Public Health and Medical Services Act. https://url.kr/qnksbj. Accessed 16 Feb 2022. 23. Health Insurance Review & Assessment Service. Total Solution for Value-based Healthcare Purchasing HIRA System. https://www.hira.or.kr/eng/ebook/00_Page_img/extra/00.pdf. 24. Hong JS, Kang HC. Regional differences in treatment requency and casefatality rates in Korean patients with cute myocardial infarction using the Korea national ealth insurance claims database: findings of a large retrospective cohort study. Medicine. 2014;93:e287. 25. Kim H, Kim Y. Factors influencing the use of health services by trauma patients according to insurance type and injury severity score in South Korea: Based on Andersen’s behavioral model. PLoS ONE. 2020;15:e0238258. 26. Industrial Accident Compensation Insurance Act. https://elaw.klri.re.kr/kor_service/lawView.do?hseq=51148&lang=ENG. 27. Jo M Oh H Jang SY The effect of residence in underserved emergency medical services areas on awareness of myocardial infarction symptoms in Korea J Health Info Stat 2021 46 8 18 10.21032/jhis.2021.46.1.8 28. Park J Lim T Korean triage and acuity scale (KTAS) J Korean Soc Emerg Med 2017 28 547 551 29. Hansagi H Olsson M Sjoberg S Tomson Y Göransson S Frequent use of the hospital emergency department is indicative of high use of other health care services Ann Emerg Med 2001 37 561 567 10.1067/mem.2001.111762 11385324 30. Kim S Lee J Utilization of emergency medical services according to the medical aid benefit Health Serv Manage Res 2017 11 35 47 31. Kim BK. Applying health insurance to emergency rooms and non-emergency patients in areas vulnerable to emergency medical care. Yonhap News. 2015. https://www.yna.co.kr/view/AKR20151126192500017. Accessed 19 May 2022. 32. Oh M Jeon B Lee J Jeong T Heo T Inflow and outflow type analysis of emergency department patients of the Honam region J Korean Soc Emerg Med 2019 30 348 354 33. Acosta AM da Silva Lima MAD Frequent users of emergency services: associated factors and reasons for seeking care Rev Lat Am Enfermagem 2015 23 337 344 10.1590/0104-1169.0072.2560 26039306 34. Ustulin M Woo J Woo JT Rhee SY Characteristics of frequent emergency department users with type 2 diabetes mellitus in Korea J Diabetes Investig 2018 9 430 437 10.1111/jdi.12712 28686322 35. Komenda P Tangri N Klajncar E Eng A Di Nella M Hiebert B Patterns of emergency department utilization by patients on chronic dialysis: population-based study PLoS ONE 2018 13 e0195323 10.1371/journal.pone.0195323 29664922 36. Hudon C Courteau J Krieg C Vanasse A Factors associated with chronic frequent emergency department utilization in a population with diabetes living in metropolitan areas: a population-based retrospective cohort study BMC Health Serv Res 2017 17 1 9 10.1186/s12913-017-2453-3 28049468 37. Behr JG Diaz R Emergency department frequent utilization for non-emergent presentments: Results from a regional urban trauma center study PLoS ONE 2016 11 e0147116 10.1371/journal.pone.0147116 26784515 38. Han JO Kang KH Yim J The usual source of healthcare and frequent visits to emergency departments J Korean Soc Emerg Med 2017 21 97 107 39. Uscher-Pines L Pines J Kellermann A Gillen E Mehrotra A Emergency department visits for nonurgent conditions: Systematic literature review Am J Manag Care 2013 19 47 59 23379744 40. Lee I The limit of emergency medical services: emergency services public goods HIRA Res 2020 14 19 25