Regional disparities in lung cancer surgery and survival in South Korea: a nationwide population-based retrospective cohort study using linked cancer registry and national health insurance data

Introduction

Lung cancer is a major and increasingly prevalent cancer in Korea, with age-standardized incidence steadily rising over the past decade (1,2). Despite this increase, clinical outcomes have gradually improved, with 5-year relative survival increasing from 30.2% in 2006–2010 to 37.7% in 2017–2021 owing to advances in surgery, targeted therapies, and immunotherapy (3).

However, regional disparities in lung cancer outcomes persist. In Korea, cancer care resources—including specialized personnel, facilities, and clinical trials—are heavily concentrated in capital-region hospitals and perceptions of superior care in capital-region hospitals drive inter-regional patient migration, causing many non-capital area residents to travel to metropolitan hospitals, potentially disrupting timely coordinated care (4). Analyses of the Korean National Health Insurance Service (KNHIS) claims and national surveys indicate that cross-regional travel is associated with longer referral and scheduling intervals, higher out-of-pocket spending, and more fragmented follow-up. These factors often delay the initiation of systemic therapy and the management of time-sensitive complications. Such patterns reflect patient concentration in metropolitan tertiary hospitals, imposing substantial personal and systemic burdens. Long-distance travel for surgery and follow-up causes physical exhaustion, financial strain, and discontinuity of care, while overcrowding in metropolitan centers delays access for severe cases and contributes to loss of patients and medical professionals from regional hospitals (5,6). Centralization across multiple high-income countries (e.g., the United States and Germany) involves consolidating thoracic surgery in high-volume metropolitan centers, increasing inter-regional patient travel, and downgrading or closing low-volume units, with multidisciplinary services and early-phase trials concentrated in designated cancer centers (7-9). In Korea, cancer care has increasingly become concentrated in capital-region hospitals. Approximately 40% of patients undergoing lung cancer surgery receive treatment in these metropolitan centers, with the highest rates among those living in rural areas. This centralization has intensified over time and is more common among younger and higher-income patients (10).

Structurally, cancer care in Korea is organized around a small number of tertiary general hospitals, most of which are located in the capital region (Seoul, Incheon, and Gyeonggi). To counterbalance this concentration, the government established the Regional Cancer Center Program in 2004 to strengthen local capacity for cancer treatment, research, and prevention; however, sustained underinvestment has limited its impact, and specialized thoracic oncology resources, systemic therapies, and clinical trials remain disproportionately concentrated in metropolitan centers (11). As a result, a substantial and growing share of non-capital residents travel to capital-region hospitals for lung cancer surgery, particularly younger and higher-income patients (10).

These patterns raise an unresolved controversy that current evidence has not adequately addressed. On one hand, non-capital patients may face genuine treatment disadvantages arising from insufficient local surgical and multidisciplinary capacity; on the other, excessive concentration of care in the capital region may generate resource inefficiency, longer referral and travel times, and avoidable financial and logistical burdens without a corresponding survival benefit for all patients. Prior nationwide analyses have largely documented the phenomenon of patient migration to the capital region or reported an overall survival advantage of metropolitan hospitals, but they seldom examined whether this advantage is uniform across disease stages, and often relied on sample-based claims data lacking detailed clinical variables (10,12,13). Whether centralization benefits all surgical patients equally, or only those with specific disease characteristics, therefore remains unclear. Clarifying this distinction is essential for designing a resource-allocation strategy that neither under-serves non-capital regions nor over-concentrates care unnecessarily.

Patients treated in capital-region hospitals have considerably better survival than those treated in non-capital regions (12). Moreover, recent evidence shows centralization of cancer surgery reduces postoperative mortality; however, many existing studies have limited methodological rigor (13). In particular, previous nationwide analyses focused on specific cancer types such as lung and gastric cancers, relied on National Health Insurance Service (NHIS) sample data lacking detailed clinical variables, and used broad regional classifications that obscured within-capital-area quality differences. To address these limitations, our study used linked nationwide data from the Korea Central Cancer Registry (KCCR) and NHIS to analyze all patients who underwent lung cancer surgery, enabling detailed adjustment for cancer stage and patient characteristics while evaluating the effect of hospital location on survival outcomes.

In this study, we aim to (I) analyze trends in hospital surgical use among non-capital residents with resectable lung cancer, (II) identify factors influencing the selection of capital-region hospitals, and (III) compare surgical outcomes according to hospital location. We present this article in accordance with the STROBE reporting checklist (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-0533/rc).

Methods

Study design and data sources

This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Institutional Review Board of Chungbuk National University (CBNU-2024-A-0048). The requirement for informed consent was waived because this study used de-identified, secondary administrative data. This nationwide population-based retrospective cohort study used linked data from the KCCR (https://ncc.re.kr/main.ncc?uri=english/sub04_ControlPrograms02) and the KNHIS (https://www.nhis.or.kr/english/index.do). The KCCR and KNHIS data have complementary strengths. The KCCR provides nationwide population-based information on cancer diagnosis, SEER stage, and histologic type and covers more than 90% of newly diagnosed cancer cases in Korea. The KNHIS provides near-universal longitudinal claims data on demographic and socioeconomic characteristics, treatment, comorbidities, healthcare utilization, mortality, and medical institution attributes. Linking the two databases enabled the integration of cancer-specific clinical information with nationwide longitudinal healthcare utilization and outcome data.

KNHIS

We used data from the KNHIS, which provides near-universal health coverage in Korea. The KNHIS database contains nationwide health insurance claims information generated through its fee-for-service reimbursement system. It includes diagnostic codes, inpatient and outpatient healthcare utilization, diagnostic examinations, procedures, treatments, and prescription medications. Information on comorbid conditions is also available. Mortality data are obtained from national administrative records and cross-validated against health insurance qualification termination records. The database additionally provides demographic and socioeconomic information, including age, sex, residential area, and income level based on insurance premium tiers. The KNHIS database is widely used in epidemiological and health policy research (14).

Cancer Registration System in the Republic of Korea

The Korean Central Cancer Registry (KCCR), a nationwide government-sponsored cancer registry, was established in 1980. It is a population-based registry that currently captures more than 90% of newly diagnosed cancer cases nationwide, providing broad coverage of the Korean cancer population (15). The KCCR provides information on patient characteristics and cancer-specific variables, including patient sex, age at diagnosis, date of diagnosis, cancer site, and summary stage. The Surveillance, Epidemiology, and End Results (SEER) stage information has been available through the KCCR since 2006. SEER stage at diagnosis is classified as localized (invasive cancer limited to the organ of origin), regional (tumor extension beyond the organ of origin), distant (spread to distant areas from the primary tumor), and unknown (16).

Study patient identification

This study is part of a research project investigating differences in lung cancer treatment facility use and survival among patients residing in non-capital regions based on whether they received care in capital or non-capital regions. To construct a project database, we linked the KNHIS and KCCR data using personal identification numbers. Patients newly diagnosed with lung cancer [International Classification of Diseases, 10th Revision (ICD-10) code C34] were identified from the KCCR between January 1, 2012, and December 31, 2019 (n=188,176). Incident lung cancer cases were defined using the earliest lung cancer diagnosis date recorded in the KCCR. The KCCR data consisted of one patient-level record per individual; therefore, duplicate patient records did not arise. For individuals with multiple primary cancer diagnoses, information on each cancer was recorded in separate fields within the same patient record.

We excluded patients aged <19 years at diagnosis or the index date (n=42) and those without complete KNHIS eligibility or claims information required to ascertain baseline covariates, including residential area, income level, insurance type, and comorbidity burden (n=3,623). The resulting nationwide lung cancer cohort included 184,511 patients, of whom 52,521 underwent lung cancer surgery. Lung cancer surgeries were identified using KNHIS reimbursement codes O1400–O1405, O1410, O1421–O1424, and O1431–O1432. Patients who underwent exploratory thoracotomy alone (O1360) were not classified as having undergone qualifying lung cancer resection. Surgical procedures were classified as pneumonectomy, lobectomy, segmentectomy, or wedge resection, encompassing both anatomic and limited pulmonary resections. For patients with multiple qualifying lung cancer surgery records, the earliest surgery was defined as the index surgery. The hospital region was classified according to the location of the institution where the index surgery was performed. To examine hospital utilization and survival outcomes according to the region of the index surgical hospital, we restricted the surgical cohort to patients residing in non-capital regions (n=27,482). Of these patients, 13,893 underwent index surgery at capital-region hospitals and 13,589 underwent index surgery at non-capital-region hospitals (Figure S1).

Variables and outcome measures

Sex, age at diagnosis, residential area, income level, and insurance type were obtained from the KNHIS data for the year of lung cancer diagnosis. Age was categorized as <50, 50–59, 60–69, or ≥70 years, and income level was classified as Medical Aid or health insurance premium quintiles. The Charlson comorbidity index (CCI) was calculated from ICD-10 main diagnosis codes in inpatient and outpatient claims during the year of diagnosis, excluding the age component, and was categorized as 0, 1, 2, 3, or ≥4. SEER stage and histologic type at diagnosis were obtained from the KCCR. Primary treatment was identified from KNHIS claims within one year after diagnosis and classified according to combinations of surgery, radiotherapy, and chemotherapy, including no treatment. Although distant-stage lung cancer is generally not considered surgically treatable, we included patients classified as distant stage who underwent surgery as recorded in the KNHIS database to reflect real-world practice and the scope of national surgical care.

In this study, we defined the capital region as Seoul, Incheon, and Gyeonggi, with all other areas (e.g., Gangwon, Chungbuk, Gyeongbuk, etc.) classified as non-capital regions. The location and type of the surgical hospital were determined using KNHIS medical institution information for the hospital where the index surgery was performed. Medical institution type was classified as a tertiary general hospital or general hospital.

Our analyses included two dependent variables. The first was the use of medical facilities located in capital regions. Among patients residing in non-capital regions, capital-region care utilization was defined as undergoing index lung cancer surgery at a hospital located in Seoul, Incheon, or Gyeonggi. This classification was based on the location of the hospital where the index surgery was performed, rather than the hospital of initial diagnosis or overall healthcare utilization records. The second outcome was all-cause mortality within 1, 3, and 5 years after lung cancer diagnosis. Mortality was ascertained using the date-of-death variable in the KNHIS database, which is based on national death registration records provided by the Ministry of the Interior and Safety. Treatment continuity was not evaluated as a study outcome. As cause-of-death information was not available in the linked dataset, overall mortality was used instead of cancer-specific mortality.

Statistical analyses

Descriptive statistics summarized patient demographic and clinical characteristics and examined trends in capital-region hospital use for lung cancer surgery from 2012 to 2019. To identify factors associated with capital-region hospital utilization, we used multivariable logistic regression, adjusting for sex, age, residence, income, CCI, cell type, and cancer stage (SEER), using data from 2017–2019. Restricting analyses to this recent period captured current healthcare utilization patterns and reduced potential heterogeneity from earlier changes in hospital access and data reporting. Among patients residing in non-capital regions, the hazard ratio analyses were restricted to patients diagnosed with lung cancer between 2017 and 2019 who underwent index lung cancer surgery within one year after diagnosis. The Cox regression analysis examined the association between the region of the index surgical hospital and overall mortality. Hazard ratios (HRs) for 1-, 3-, and 5-year overall mortality (not cancer-specific mortality) in patients treated at non-capital- versus capital-region hospitals were calculated using Cox proportional hazards regression. Follow-up began on the date of lung cancer diagnosis and continued until death or December 31, 2022, whichever occurred first. Patients who remained alive on December 31, 2022, were administratively censored. To isolate the independent effect of hospital location (capital vs. non-capital region) on mortality risk, we applied stratification in the Cox proportional hazards model by stage at diagnosis, medical institution type, and cell type. Socioeconomic status (SES) (income level) and comorbidity burden (CCI) were included as covariates rather than stratification variables because they influence overall mortality risk and not the baseline hazard structure. This specification enabled control for potential confounding while maintaining model stability and adequate subgroup sample sizes. Analyses were performed using SAS statistical software (version 9.4; SAS Institute Inc., Cary, NC, USA). P<0.05 was considered statistically significant.

Results

Baseline characteristics of study participants

From 2012 to 2019, 184,511 patients with lung cancer were identified in Korea. The annual number of newly diagnosed patients steadily increased, from 19,277 in 2012 to 27,370 in 2019. Most were male (70.1%) aged ≥70 years (56.9%) (Table 1). The proportion of patients aged ≥70 years showed gradually increased from 55.0% in 2012 to 58.3% in 2019. The proportion of localized-stage diagnoses increased from 22.3% in 2012 to 26.8% in 2019, suggesting potential improvements in early detection. Socioeconomic disparities were evident: 31.9% of patients belonged to the highest income quintile, while 6.5% were below the poverty line. Most patients were covered by medical insurance (93.5%). Comorbidity burden was high: 67.2% of patients had a CCI score ≥4. Adenocarcinoma was the most common histologic subtype (43.2%), followed by squamous/transitional cell carcinoma (23.5%). One-third of patients (33.5%) did not receive any cancer-specific treatment. A total of 19.7% underwent surgery alone, 16.2% received chemotherapy alone, and 14.1% received both radiotherapy and chemotherapy.

Trends and regional disparities in capital hospital utilization by residence among patients with lung cancer

Throughout the study period (2012–2019), a modest upward trend in the proportion of patients with lung cancer was observed in non-capital regions who utilized capital-region hospitals (increasing from 49.7% to 51.6%). Correspondingly, the use of hospitals in non-capital regions declined from 50.3% to 48.4% (Figure 1). Moreover, most non-capital provinces showed increasing utilization of capital-region hospitals, with notable inter-regional variation. Among non-capital regions, Jeju showed the largest increase in capital-region hospital utilization, rising from 59.4% in 2012 to 75.0% in 2019, followed by Sejong (66.7% to 80.6%). Jeonbuk and Daejeon also showed notable increases, from 47.3% to 57.2% and from 51.3% to 60.8%, respectively. Meanwhile, capital-region residents (Seoul, Incheon, and Gyeonggi) consistently received care within their own region, with utilization rates exceeding 98.8% throughout the study period (Table 2).

Figure 1 Trends in hospital utilization by hospital region among non-capital residents with lung cancer, 2012–2019.

Utilization of capital hospitals by patient characteristics among non-capital residents

Between 2012 and 2019, capital-region hospital utilization among non-capital residents with lung cancer showed heterogeneous trends across patient subgroups (Figure 2, Table S1). Females consistently demonstrated higher utilization than males (2012: 51.4% vs. 48.8%; 2019: 55.1% vs. 49.3%). Younger patients exhibited the largest increase, particularly those <50 years (48.6% to 61.4%), followed by those aged 50–59 years (54.8% to 61.9%). Utilization among patients ≥70 years remained unchanged (46.9% to 46.5%). Utilization progressively increased with income level, from 16.7% to 21.6% in the lowest quintile and from 49.5% to 55.9% in the fourth quintile. The highest quintile showed minimal change (60.3% to 60.5%). Patients with medical insurance consistently had higher utilization than Medicaid recipients (2019: 53.1% vs. 21.6%). Urban residents had the highest utilization (57.7% to 59.9%), while metropolitan residents had the lowest (42.4% to 45.5%). Utilization increased from 52.6% to 58.2% in patients with fewer comorbidities (CCI 0–3), while those with severe comorbidities (CCI ≥4) experienced a marginal decline (46.1% to 45.8%). Stage-specific trends revealed a marked increase among localized patients (44.4% to 52.0%), whereas regional (55.6% to 53.6%) and distant-stage patients (54.5% to 42.6%) showed declining utilization. Patients treated with surgery alone showed a steady increase (49.3% to 53.7%), whereas those receiving multimodal treatments such as surgery with chemotherapy or radiotherapy declined (e.g., surgery + chemotherapy, 47.4% to 43.9%). Adenocarcinoma cases consistently showed higher utilization than squamous cell carcinoma (2019: 52.1% vs. 44.0%).

Figure 2 Trends in capital-region hospital utilization by patient characteristics among non-capital residents with lung cancer, 2012–2019. SEER, Surveillance, Epidemiology, and End Results; Q, quintile.

Logistic regression analysis of factors influencing capital hospital utilization among non-capital patients with lung cancer

Multivariate logistic regression analysis identified several factors associated with receiving surgery in capital-region hospitals among non-capital residents between 2017 and 2019 (Table 3). Female patients had higher odds than males [adjusted odds ratio (aOR) 1.186, 95% confidence interval (CI): 1.090–1.290], and patients <70 were more likely to use capital hospitals than those ≥70. The strongest association was observed in the 50–59 age group (aOR 1.577, 95% CI: 1.404–1.772). Urban and rural residents showed higher odds than metropolitan residents (aORs 2.124, 95% CI: 1.948–2.316 and aORs 1.781, 95% CI: 1.611–1.970, respectively). A strong positive gradient by income level was observed, with patients in the highest income quintile being more than four times as likely to receive surgery in capital hospitals than those below the poverty line (aOR 4.357, 95% CI: 3.560–5.333). Similarly, medical insurance holders were more likely to use capital hospitals than Medicaid recipients (aOR 3.064, 95% CI: 2.522–3.772). Patients with lower comorbidity burden (CCI 0–3) had higher odds of capital hospital use (aOR 1.595, 95% CI: 1.472–1.727). Regional-stage patients had higher odds than distant-stage patients (aOR 1.441, 95% CI: 1.319–1.574), and patients with squamous cell carcinoma had lower odds than those with adenocarcinoma or other histological types. Those receiving surgery + chemotherapy were less likely to use capital-region hospitals (aOR 0.758, 95% CI: 0.644–0.893) than those who had surgery only.

Mortality risk by hospital region, stage, and cell type in non-capital patients with lung cancer

Cox proportional hazards models compared all-cause mortality among non-capital residents undergoing lung cancer surgery in capital vs. non-capital hospitals (Table 4). Notably, in localized stage, 1-, 3-, and 5-year mortality did not differ between capital- and non-capital-region hospitals. Alternatively, in the regional stage, mortality was higher in non-capital hospitals at 3 years (aHR, 1.219; 95% CI, 1.076–1.382) and 5 years (aHR, 1.188; 95% CI, 1.067–1.322). In the distant stage, non-capital hospitals showed higher mortality at 1, 3, and 5 years (5 years: aHR, 1.239; 95% CI, 1.011–1.519). By hospital type, general hospitals showed no regional mortality differences. In tertiary general hospitals, non-capital hospitals showed higher mortality at 1, 3, and 5 years (1 year: aHR 1.279; 95% CI, 1.075–1.522; 3 years: aHR 1.194; 95% CI, 1.079–1.321; 5 years: aHR 1.104; 95% CI, 1.013–1.202), consistent with stage-specific results showing no regional difference in the localized stage and higher mortality in regional and distant stages. For squamous cell carcinoma, no regional difference in localized disease was observed; however, non-capital hospitals had higher mortality for regional-stage disease at 3 and 5 years (aHR, 1.228; 95% CI, 1.003–1.504; and aHR, 1.243; 95% CI, 1.042–1.482, respectively) and for distant-stage disease at 1 and 3 years (aHR, 2.473; 95% CI, 1.092–5.604; and aHR, 1.925; 95% CI, 1.076–3.447, respectively), with no 5-year difference (aHR, 1.640; 95% CI, 0.957–2.808). For adenocarcinoma, no regional differences appeared in localized or regional disease, whereas distant-stage disease showed higher mortality in non-capital hospitals at 1 and 3 years (aHR, 2.320; 95% CI, 1.251–4.300; and aHR, 1.459; 95% CI, 1.040–2.048, respectively), with no 5-year difference (aHR, 1.147; 95% CI, 0.857–1.536).

Discussion

This nationwide study examined how hospital location affects lung-cancer surgical outcomes in Korea, where care remains highly centralized. Contrary to previous studies that consistently reported superior outcomes for patients treated in capital hospitals, our findings suggest that this advantage is not observed across all stages. For localized disease, survival was comparable in capital and non-capital hospitals, indicating that early-stage lung cancer can be effectively treated in regional hospitals with adequate surgical capacity. In contrast, patients with advanced-stage disease treated in non-capital hospitals showed poorer survival, suggesting that centralization benefits are stage-specific rather than universal. These results highlight the need for a coordinated national strategy that optimizes the roles of capital and regional hospitals according to disease stage.

Our stage-specific findings both extend and qualify international evidence on cancer care centralization. Prior studies from high-income countries have reported that consolidating thoracic surgery in high-volume centers is associated with lower postoperative mortality, and a systematic review similarly concluded that centralization reduces short-term surgical mortality (7,13). However, much of this evidence emphasizes volume–outcome relationships and perioperative endpoints rather than long-term, stage-specific survival, and travel burden associated with centralized care has itself been linked to adverse outcomes and access barriers (9). Our results align with the international literature in showing a survival benefit for advanced-stage disease treated in higher-resource capital-region centers, but diverge in demonstrating no such benefit for localized disease—indicating that a uniform centralization policy may over-concentrate early-stage patients who could be effectively treated locally, while under-serving advanced-stage patients who most require multidisciplinary resources.

Comparable outcomes for patients with early-stage disease in regional hospitals stem from standardized surgical care and adherence to national treatment guidelines. Operative quality is generally consistent across institutions when qualified thoracic surgeons perform procedures. Early-stage lung cancer relies primarily on timely surgery and standard postoperative care rather than advanced systemic therapies.

In contrast, patients with advanced-stage disease treated in non-capital hospitals demonstrated poorer survival, likely reflecting limited access to systemic therapy, clinical trials, and multidisciplinary care. Notably, these disparities may also be influenced by patient preferences—as those who travel to capital hospitals tend to pursue more aggressive or novel therapies, whereas those remaining in regional hospitals may be more inclined to decline them despite availability. The scarcity of clinical trials in regional hospitals further contributes to these outcomes. The 2023 Korean Cancer Research Trends Report indicates that more than half of lung cancer trials occur in major capital-region institutions, restricting trial access for patients in non-capital areas. This centralization limits access to innovative treatments, increases travel and financial burdens on patients, and exacerbates inequities for those with limited resources. Such disparities highlight that access to clinical research and emerging therapies remains concentrated in metropolitan centers, reinforcing the clinical advantage of capital hospitals. Addressing this imbalance requires sustained investment and system-level reform. Establishing satellite clinical trial units in regional cancer centers, expanding funding for decentralized trial models, and developing unified accreditation systems for quality assurance could enhance accessibility and patient safety.

The continued preference for capital hospitals among advanced-stage patients reflects structural and perceptual barriers within regional care. The centralization of lung cancer care in Korea arises from a combination of perceptual and structural factors. Patient trust is crucial—those confident in local hospitals are less inclined to seek care in the capital—yet many continue to believe that metropolitan centers offer superior expertise and resources (4). Our findings also reflect this pattern, with patients with early-stage disease and younger disproportionately seeking care in capital hospitals despite sufficient surgical capacity in regional centers. Improved transportation and weak primary-care gatekeeping further facilitate direct access to tertiary hospitals, while limited clinical trials and advanced therapies in non-capital regions sustains this imbalance (11,17-21). Therefore, strengthening trust in regional hospitals and expanding access to innovative treatments and clinical research opportunities are essential to mitigate unnecessary centralization.

Demographic and socioeconomic characteristics also influenced hospital selection. Younger age, female sex, higher income, and private insurance were associated with greater use of capital hospitals, aligning with evidence that patients with high SES prioritize perceived quality and physician reputation, whereas those with lower SES face constraints related to distance and cost (22-27). Although high-income patients remain the primary users of capital-region hospital users, our analysis newly shows a substantial rise in Medicaid beneficiaries receiving surgery in these institutions. This indicates that centralization has intensified to the point that even low-income patients feel compelled to seek care in the capital region. These findings highlight that socioeconomic disparities now coexist with growing dependence on metropolitan care, highlighting the need to strengthen both the capacity and credibility of regional hospitals to reduce this over-reliance on the capital region.

Despite the establishment of Korea’s Regional Cancer Center Program in 2004—modeled after the U.S. comprehensive cancer center system to strengthen regional capabilities in cancer care, research, and prevention—its impact on improving local treatment capacity has been limited. Although the program initially provided substantial funding for infrastructure, advanced equipment, and cancer research, these resources were markedly reduced within a few years. Over the past two decades, limited institutional and financial support has hindered the development of regional cancer centers, sustaining dependence on capital-region hospitals for lung cancer treatment. To address this imbalance requires nationally coordinated efforts to develop cancer care capacity more evenly across regions. Localized cases should be managed within regional hospitals with adequate capability, while structured collaboration with high-capacity capital hospitals should be promoted for patient with advanced-stage disease. Strengthening and actively promoting the treatment capacity of regional cancer centers is essential to build public trust and ensure timely, equitable, and continuous cancer care within local communities.

Taken together, these findings support a differentiated, stage-specific allocation of national lung cancer care resources rather than uniform centralization. First, regional hospitals with demonstrated surgical capacity and quality accreditation should serve as the primary setting for early-stage lung cancer surgery, supported by transparent public reporting of stage-specific outcomes to build patient trust. Second, capital-region and regional centers should be linked through formal referral and shared-care networks for advanced-stage disease, so that patients can access systemic therapy, multidisciplinary care, and clinical trials without permanent relocation. Third, decentralized and satellite clinical-trial models should be expanded to reduce the concentration of trial access in metropolitan institutions (11). Fourth, sustained and protected funding for regional cancer centers is needed to reverse the erosion of local capacity that has perpetuated dependence on the capital region. Such measures would reduce unnecessary centralization while ensuring that patients who genuinely benefit from concentrated resources retain timely access.

This study had several limitations. First, administrative claims data may be subject to coding inaccuracies and do not capture detailed information on patient treatment preferences and satisfaction, clinical indicators that may influence treatment decisions—such as pulmonary function, performance status, and treatment intent— or provider-level factors related to referral and treatment practices. These unmeasured factors may have influenced hospital selection and introduced residual confounding. Second, use of all-cause rather than cancer-specific mortality may obscure the direct impact of treatment location. Third, unmeasured factors—including patient preferences, institutional characteristics, and hospital resources—could not be fully adjusted for. Because this was an observational study, the associations between hospital region and survival should not be interpreted as causal effects; residual confounding from unmeasured patient preferences, disease severity, and referral patterns cannot be excluded. Fourth, formal diagnostic assessments of model assumptions, including goodness-of-fit for the logistic regression models and the proportional hazards assumption for the Cox models, were not performed; therefore, potential model misspecification cannot be excluded. Fifth, sensitivity analyses were not performed to assess the robustness of the findings under alternative analytic specifications. In addition, restricting the analysis to surgically treated patients may underestimate disparities among those receiving non-surgical care. Despite these limitations, this nationwide, population-based study with long-term follow-up provides valuable evidence on persistent inequities in lung cancer management and offers important insights to guide policies that strengthen regional capacity and promote equitable access to advanced treatment.

Conclusions

In this nationwide study of non-capital residents undergoing lung cancer surgery in Korea, we found that the impact of hospital location on survival outcomes varies by disease stage. While survival outcomes for localized-stage lung cancer were comparable between capital and non-capital hospitals, patients with regional and distant-stage disease experienced significantly higher mortality when treated in non-capital hospitals.

These findings suggest that the benefits of centralized cancer care are not universal but are concentrated in patients with more advanced disease who require multidisciplinary treatment and access to systemic therapies and clinical trials. In contrast, early-stage lung cancer can be effectively managed in regional hospitals with adequate surgical capacity.

Our results highlight the need for a differentiated, stage-specific healthcare delivery strategy that strengthens the role of regional hospitals in early-stage care while improving access to advanced treatments and clinical research in non-capital regions. Such efforts are essential to reduce unnecessary centralization and ensure equitable, timely, and high-quality cancer care across regions.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-0533/rc

Peer Review File: Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-0533/prf

Funding: This work was supported by the National R&D Program for Cancer Control, Ministry of Health & Welfare, Republic of Korea (No. HA21C0135000021), with additional funding from the Regional Innovation System & Education (RISE) Program through the Chungbuk Regional Innovation System & Education Center, funded by the Ministry of Education (MOE) and the Chungcheongbuk-do, Republic of Korea (No. 2025-RISE-11-014-03).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-0533/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Institutional Review Board of Chungbuk National University (CBNU-2024-A-0048). The requirement for informed consent was waived because this study used de-identified, secondary administrative data.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Bray F, Laversanne M, Sung H, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2024;74:229-63. [Crossref] [PubMed]
2. Lee KS, Chang HS, Lee SM, et al. Economic Burden of Cancer in Korea during 2000-2010. Cancer Res Treat 2015;47:387-98. [Crossref] [PubMed]
3. Park EH, Jung KW, Park NJ, et al. Cancer Statistics in Korea: Incidence, Mortality, Survival, and Prevalence in 2021. Cancer Res Treat 2024;56:357-71. [Crossref] [PubMed]
4. Lee MH, Lee JS, Do YK. Inter-regional patient outmigration to Seoul in South Korea: the role of regional healthcare quality perceptions. BMC Health Serv Res 2025;25:407. [Crossref] [PubMed]
5. Cho S, Chang Y, Kim Y. Cancer Patients' Utilization of Tertiary Hospitals in Seoul Before and After the Benefit Expansion Policy. J Prev Med Public Health 2019;52:41-50. [Crossref] [PubMed]
6. Choi DW, Kim SJ, Kim DJ, et al. Does fragmented cancer care affect survival? Analysis of gastric cancer patients using national insurance claim data. BMC Health Serv Res 2022;22:1566. [Crossref] [PubMed]
7. Walter J, Tufman A, Holle R, et al. Differences in therapy and survival between lung cancer patients treated in hospitals with high and low patient case volume. Health Policy 2020;124:1217-25. [Crossref] [PubMed]
8. Harrison S, Judd J, Chin S, et al. Disparities in Lung Cancer Treatment. Curr Oncol Rep 2022;24:241-8. [Crossref] [PubMed]
9. Stoyanov DS, Conev NV, Donev IS, et al. Impact of travel burden on clinical outcomes in lung cancer. Support Care Cancer 2022;30:5381-7. [Crossref] [PubMed]
10. Kim D, Kang GW, Jang H, et al. Trend of lung cancer surgery, hospital selection, and survival between 2005 and 2016 in South Korea. Thorac Cancer 2022;13:210-8. [Crossref] [PubMed]
11. Jang J, Choi W, Sim SH, et al. Regional disparities in the availability of cancer clinical trials in Korea. Epidemiol Health 2024;46:e2024006. [Crossref] [PubMed]
12. Kim W. An explanatory study on the impact of regional characteristics on various aspects of cancer mortality. In: National Cancer Center; Ministry of Health and Welfare; 2022.
13. Grilli R, Violi F, Bassi MC, et al. The effects of centralizing cancer surgery on postoperative mortality: A systematic review and meta-analysis. J Health Serv Res Policy 2021;26:289-301. [Crossref] [PubMed]
14. Shin DW, Cho B, Guallar E. Korean National Health Insurance Database. JAMA Intern Med 2016;176:138. [Crossref] [PubMed]
15. Yang MS, Park M, Back JH, et al. Validation of Cancer Diagnosis Based on the National Health Insurance Service Database versus the National Cancer Registry Database in Korea. Cancer Res Treat 2022;54:352-61. [Crossref] [PubMed]
16. Park SH, Kang MJ, Yun EH, et al. Epidemiology of Gastric Cancer in Korea: Trends in Incidence and Survival Based on Korea Central Cancer Registry Data (1999-2019). J Gastric Cancer 2022;22:160-8. [Crossref] [PubMed]
17. Lim Y, Lim JM, Jeong WJ, et al. Korean Cancer Patients' Awareness of Clinical Trials, Perceptions on the Benefit and Willingness to Participate. Cancer Res Treat 2017;49:1033-43. [Crossref] [PubMed]
18. Choi JK, Kim SH, Park MB. Effects of high-speed rail on health-care service utilization. Journal of Transport & Health 2019;12:13-20.
19. Kim AM, Cho S, Kim HJ, et al. Primary Care Patients' Preference for Hospitals over Clinics in Korea. Int J Environ Res Public Health 2018;15:1119. [Crossref] [PubMed]
20. Ock M, Kim JE, Jo MW, et al. Perceptions of primary care in Korea: a comparison of patient and physician focus group discussions. BMC Fam Pract 2014;15:178. [Crossref] [PubMed]
21. Sim B, Shin J, Kim HW, et al. Evidence of Overlapping Roles Between Clinics and Hospitals in Primary Care. J Korean Med Sci 2025;40:e219. [Crossref] [PubMed]
22. Lyratzopoulos G, Barbiere JM, Greenberg DC, et al. Population based time trends and socioeconomic variation in use of radiotherapy and radical surgery for prostate cancer in a UK region: continuous survey. BMJ 2010;340:c1928. [Crossref] [PubMed]
23. Wetterholm M, Turkiewicz A, Stigmar K, et al. The rate of joint replacement in osteoarthritis depends on the patient's socioeconomic status. Acta Orthop 2016;87:245-51. [Crossref] [PubMed]
24. Lin C, Tu P, Parker T, et al. The Influences of SES on Patient Choice of Doctor: A Systematic Review. Am J Prev Med 2024;67:759-69. [Crossref] [PubMed]
25. Gil MR, Choi CG. Factors Affecting the Choice of National and Public Hospitals Among Outpatient Service Users in South Korea. Inquiry 2019;56:46958019833256. [Crossref] [PubMed]
26. Jang SY, Seon JY, Oh IH. Influencing Factors of Transportation Costs Regarding Healthcare Service Utilization in Korea. J Korean Med Sci 2020;35:e290. [Crossref] [PubMed]
27. Lee S. Spatial and Socioeconomic Inequalities in Accessibility to Healthcare Services in South Korea. Healthcare (Basel) 2022;10:2049. [Crossref] [PubMed]