Specialty hospital’s operational performance on length of stay and healthcare expenditure: a cross-sectional study analyzing national health insurance claims data

Introduction

Background

Medical facilities that provide specialized medical services to patients are called specialty hospitals (SHs). According to the Deficit Reduction Act of 2006 in the United States, an SH is defined as “hospital that is exclusively or primarily engaged in the care or treatment of one of the following categories: patients with a cardiac condition; patients with an orthopaedic condition; or patients receiving a surgical procedure” (1). General examples of SHs include physician-owned orthopedic, cardiac, and surgical SHs (2). The Center for Medicare and Medicaid Services applied for various payments for SHs (3). More than 600 SHs are present in the US (4).

In the UK, there are 149 specialized medical facilities that function similar to SHs in the US (5,6). Although their names vary, each country has its own hospitals similar to SHs. For example, Japan has “specified function hospitals” (7); China also has pediatric specialty hospitals and mental specialty hospitals (8,9). Since 2011, the Korean government has also implemented a program for the designation of SHs similar to those of the US. Any hospital can be designated as an SH if it applies for the designation of SH and satisfies governmental rules on SHs. As of January 1, 2024, there are 109 SHs in South Korea (10).

The appearance of SHs in medical markets is known to have a positive effect on existing hospitals from the standpoint of offering services, use of high-technology-related diagnostic services, and improvement of managerial efficiency (11,12). Owing to direct market competition with hospitals, certain benefits to patients, such as improving the quality of care and increasing patients’ choices, are anticipated. Practically, many medical consumers and patients tend to prefer SHs when expecting good medical care (13).

Study topics related to SHs have been diverse such as expenses, profits status, health care delivery, and production efficiency (14-17). Among these topics, efficiency may be the most important when it comes to running an organization, especially a hospital (18,19). This is because efficiency is linked to all sectors of hospitals related to survival and production process. According to the theory of division of labor advocated by Adam Smith (20,21), as an organization grows, it divides the production process into several parts and assigns workers or specific departments to take charge of the divided parts. The organization becomes more productive or efficient as the department or workers can easily handle the work. This process is called specialization. Based on this theory, this study predicts that SHs would be more efficient than non-SHs because they are organizations that operate through this specialization process. In this study, efficiency, one of theoretical contexts, can be evaluated through average length of stay (ALOS). ALOS is one of efficiency measures (22). From the hospital’s revenue perspective, it is also necessary to analyze the total length of stays (TLOS) and healthcare expenditure as input variables for efficiency.

Several empirical field studies suggest that SHs have greater efficiency including volume of healthcare and healthcare expenditure than non-SHs. Orthopedic SHs demonstrated a greater surgical volume and shorter length of stay (LOS) than those of general hospitals (23). The opening of specialized cardiac hospitals was also related to the increased use of coronary revascularization services (24). A study found that efficiency measured by using data envelopment methods and various input variables in hospitals was associated with different levels of specialization in patient services (25). According to studies on joint and spine SHs in Korea, SHs demonstrated greater inpatient charges, but shorter LOS than non-SHs (26,27). These findings were also observed by a different study comparing colorectal-anal SHs with hospitals providing the same care as those of SHs (28). Care provided by gastroenterologists resulted in significantly shorter length of hospital stay than generalists (29). These prior studies clearly suggest that hospital’s specialization may be deeply related to hospital output and efficiency in any environment.

Although there have been several pioneering studies, they were conducted more than five years ago and different study settings. Thus, it is necessary to revisit the current status of SHs and compare them with those previous studies. Two studies with different datasets, environments, and time intervals may or may not produce the identical research results. If two studies produce identical results, then we can easily generalize the results and policymakers can apply more solid, evidence-based research findings to practice in empirical healthcare fields.

Finally, this study is important from the following perspectives. In the healthcare market, patients have some preferences for healthcare institutions. Many patients prefer to go to tertiary or higher-level of hospitals to receive good medical services (30-32). As a result, tertiary general hospitals have many patients. SHs play an important role in preventing the concentration of patients to tertiary or general hospitals because they provide specialized medical services with good quality and treatment. Therefore, related research is needed for efficient use of healthcare resources and knowledge expansion for future healthcare policies on SHs.

Objective

This study aimed to investigate the organizational performance of SHs based on the patient’s LOS and healthcare expenditures, compared to those of non-SHs. We selected two types of SHs: joint and spine care. Health policymakers running the program and other nations considering its implementation could gain insights from this study regarding the efficiency of the program. We present this article in accordance with the STROBE reporting checklist (available at https://jhmhp.amegroups.com/article/view/10.21037/jhmhp-24-114/rc).

Methods

Study design

Individual hospitals were considered the units of analysis. The study employed a case-control study design. The case group comprised SHs providing joint (21 hospitals) and spine (17 hospitals) specialty care, which were designated as SHs by the Korean government (Figure 1). This study selected two types of SHs owing to their similarity to each other and the number of hospitals being above 30, which requires a minimal sample size for statistical power. The control group included other hospitals that were not designated as SHs but were providing joint and spine care to at least 70 patients annually. The Korean government mandated that any hospital could apply for SHs, but should have at least 10 patients annually. This study examined the distribution of hospitals providing joint and spine care among non-designated SHs, and found that most hospitals attended to at least 70 patients with joint and spine diseases. Thus, this study employed 70 patients as the cutoff point, excluding the case groups. A total of 1,148 hospitals had at least one patient receiving joint or spine care among the non-SHs. After applying the cutoff value, the control group was reduced to 758 hospitals (758/1,148×100=66.0% of the initial study population). We referred to several previous studies conducted in Korea for developing the research design (26-28,33).

Figure 1 Study hospital selection process. HIRA, Health Insurance Review and Assessment Service; DW, data warehouse.

The study was conducted in South Korea. Korea has adopted the national health insurance system. There are medical institutions such as tertiary general hospitals, general hospitals, hospitals, and clinics. Hospitals and clinics serve as primary medical institutions, and patients can receive medical services at general hospitals or higher-level general hospitals upon receiving a referral letter. For specialized care, in 2011, the Korean government implemented a program for designating SHs. Hospitals can earn this designation after successful application and subsequent evaluations. A requirement for this designation is the volume of care, which is determined by TLOS of patients receiving specialty care. This was measured using the Korean Diagnosis Related Group (KDRG) codes. Currently, the KDRG version (“v”) 3.5 is being used for the designation process. As of January 2023, there are 109 SHs and 1,907 non-SHs. Regarding SHs, Korean government designated hospitals as SHs based on the disease and medical specialty. The 10 diseases include joints diseases (n=21), cerebrovascular (n=4), colon and anus (n=5), finger replantation (n=5), heart (n=1), alcohol (n=9), breast (n=1), spine (n=17), burns (n=5), and perinatal care (n=1), totaling 69 SHs as of January 1, 2024. The six medical specialties, excluding oriental medicine (n=11), were obstetrics and gynecology (n=11), pediatrics (n=2), ophthalmology (n=11), surgery (n=3), and otolaryngology (n=2), totaling 29 SHs.

Data collection

This study used health insurance claims and administrative data from the Health Insurance Review and Assessment Service (HIRA) in Korea. The HIRA is a third-party administrator that runs the national health insurance program. The HIRA provides professional HIRA to the National Health Insurance Services and individual healthcare institutions. Thus, the HIRA contains both health insurance claims and administrative data. Regarding health insurance claims data, this study used claims in which healthcare was provided to patients from January 1, 2022, to December 31, 2022. These data were aggregated by healthcare institution as units of analysis.

Outcome and predictor variables

Three dependent variables were employed in this study. One was the TLOS for all patients. Another dependent variable was the ALOS with specialized care such as joint and spine care. The final dependent variable was patient healthcare expenditure. In the case of specialty care, this study used the KDRG version 3.5. KDRG codes were automatically issued upon submission of the When health insurance claims in the HIRA filing system. The Korean government uses KDRG codes for designating SHs.

This study targets a main independent variable, indicating whether hospitals were designated as SHs based on joint and spine diseases. As previously mentioned, the Korean government has been implementing SHs programs since 2011. This study included seven control variables: location (mega-metro city versus others), years of operation, types of hospitals (general hospitals versus small hospitals), sex (proportion of male patients/hospital), average age of the patients per hospital, severity of the patients, and type of foundation (public versus private). Several previous studies have used these variables as control variables; thus, this study adopted similar methods (27,28) to reduce potential bias. Regarding the location of hospitals, mega-metropolitan cities included Seoul (the capital of South Korea) and other districts. For types of hospitals, Korean medical law defines general hospitals as medical facilities with seven or more medical specialties with 100–300 beds, or nine or more medical specialties with ≥300 beds compared to small hospitals with 30–100 beds (34). For sex and age, the proportion of male patients and the average age of patients were calculated based on each hospital’s health insurance claim records. For the severity of patients, this study used patient clinical complexity level (PCCL) scores (35,36) from the KDRGs code of health insurance claims (KDRG version 3.5). Specifically, the sixth digit of KDRG-v3.5 presents the effect of complications and comorbidities in a patient ranging from 0 to 4: “0” [no complication and comorbidity (CC)], “1” (light or minor CC), and “2” (moderate CC), “3” (severe CC), and “4” (very severe or catastrophic CC). This study calculated the proportion of patients per hospital not having “0” and used it for controlling the severity of patients. Thus, the variable severity of patients per hospital was the percentage of patients who were relatively unhealthy. This study presented the number of beds in the table of general characteristics but excluded it from the main analysis because it was highly correlated with the types of hospitals.

Data analysis

All variables were cross-tabulated by the SH status, and descriptive statistics were calculated using the chi-square test for dichotomous variables and mean differences using the t-test for numeric variables. After examining the descriptive statistics, this study examined the correlation between independent variables and covariates to prevent multicollinearity issues for the main data analysis in the final model. Correlation matrix was used for the correlation analysis. This study found that the number of beds had a high correlation (−0.745) with the type of hospital and, thus, was excluded from the main analysis. For the final main analysis, this study had numeric outcome variables; thus, a generalized linear model (GLM) analysis was conducted to investigate the relationship between the outcome variables and the targeted independent variables after controlling for all covariates. Before the main analysis, this study conducted the Modified Park Test to diagnose the variance structure of the model (37,38). Regarding the Modified Park Test, first, this study ran a GLM with our proposed model and a link function using a Gamma family. No rule exists about initial family, but Gauss or Gamma families are recommended and this study used the latter one. Second, this study predicted yhat and residuals and took log of yhat (lnyhat) and squared residuals (res2). Third, this study estimated the regression coefficients of the model using lnyhat, res2, and a link function using the same Gamma family. Finally, this study looked at the regression and this study selected Gamma family because the coefficient was close to 2.0. The test results suggest gamma distribution to all models except a model with ALOS of specialty care (joint and spine diseases) that had a Poisson distribution. These results were used in the final model. Finally, this study reported the adjusted relative risk (aRR) with exponentiation of the regression coefficient of the GLM regression after controlling for all covariates in the model. SAS/STAT, version 9.4 (SAS Institute, Cary, NC, USA) was used for all the data analyses.

Results

General characteristics of the study hospitals

The general characteristics of the study hospitals are presented in Table 1. Specialty hospitals were more likely to be located in mega-metropolitan cities (65.8% vs. 46.3%; P=0.02) and small hospitals (84.2% vs. 68.7%; P=0.043) than non-SHs. They had a lower proportion of male patients (44.3% vs. 46.9%; P=0.03) and had less severe disease (29.3% vs. 40.5%; P<0.001) than those without SH. Among the outcome variables, the ALOS of patients receiving joint and spine care for SHs was lower than that of patients receiving the same types of service as non-SHs (10.9 vs. 12.6 days; P=0.001). However, the inpatient healthcare expenditure of SHs was higher than that of non-SHs for all claims ($1,953 vs. $1,565; P<0.001).

TLOS of all claims per hospital based on SH status

Table 2 presents the results of comparing the TLOS between SHs and non-SHs after controlling for hospital covariates in all the patients. SHs were more likely to have longer TLOS than those in the control group (aRR: 1.900; 95% CI: 1.553–2.326; P<0.001) for all health insurance claims.

ALOS of patients based on SH status

Table 3 shows the results of analyzing the ALOS of patients between SHs and non-SHs for all claims and only specialty care (joint and spine), respectively, after controlling for hospital covariates. For claims with only two specialties of care (joint and spine), the ALOS of patients with SHs was 10.0% lower than that of non-SHs (aRR: 0.900; 95% CI: 0.814–0.994, P=0.04).

Healthcare expenditure of all patients and patients with two specialized care (joint and spine) based on SH status

Table 4 shows the results of comparing patients’ healthcare expenditures between SHs and non-SHs after controlling for the hospital covariates. For all claims, the healthcare expenditure of SHs was 24.8% higher than that of non-SHs (aRR: 1.248; 95% CI: 1.125–1.384; P<0.001). For claims with only joint and spinal diseases, the healthcare expenditure of SHs was still 13.0% higher than that of non-SHs (aRR: 1.130; 95% CI: 1.036–1.232; P=0.006).

Discussion

Key findings

This study aimed to investigate the performance of SHs compared to non-SHs providing the same services. This study selected SHs specializing in joint and spine care and selected three performance indicators: TLOS, ALOS, and healthcare expenditure. All values were calculated on a hospital basis in terms of the SH status. The results showed that SHs performed better than the control group, indicating a longer TLOS but shorter ALOS, except for healthcare expenditure. These results are consistent with those of several previous studies.

Comparison with similar researches

Our study found that SHs had a longer TLOS than non-SHs although they had fewer beds than non-SHs. These facts indirectly implies that SHs may create more healthcare volume or utilization. These findings have also been observed in other studies. According to a systematic literature review targeting orthopedic specialty hospitals, SHs generated a greater surgical volume (23). Another study on SHs targeting cardiac care clearly presents that the opening of SHs was clearly related to an increase in the quantity of specialized medical services (24). Although this study focused on one variable, TLOS, it is an interesting finding to see that SHs had a larger volume of specialized care than general hospitals.

Regarding ALOS, SHs had lower ALOS than non-SHs in all patients and in two specialty care areas (joint and spine); however, statistical differences were only observed in these two care areas. This fact indirectly implies that SHs may have a higher bed turnover rate than non-SHs, implying efficient bed resource operations. This study result is partially aligned with the results of a previous study that found that SHs were more efficient (25). The previous study used completely different dataset and methodology from this study. The results of this study also provide an opportunity to compare the results with those of previous studies to see to what extent changes have occurred in outcome variables. A study conducted in 2013 in Korea found that SHs specializing in joint diseases and spine care had 31.7% and 23.5% shorter LOS, respectively, compared to non-SHs (26,27). Our study found that SHs had 10.0% shorter ALOS than non-SHs after controlling for hospital covariates. These previous studies clearly show that the current values are much lower than those of the previous studies on ALOS. The ALOS gap between the two time points was significantly reduced. There are two possible explanations for this observation. One is that non-SHs have provided intensive health care and, thus, their ALOS might have approached those of SHs, and the other is that the ALOS of SHs might be increasing for some reason and, thus, the gap between the two groups might be shortened. As the ALOS of the previous study with 17 SHs specializing in spine care was 10.9 days (27) and this study was 11.0 days, our study findings are almost the same as those of the previous study. According to a study conducted in the US, general hospitals have also achieved specialization and efficiency in providing medical services through their market positioning and experience (16).

Regarding healthcare expenditure for patients with all and specialized care, our findings indicate that SHs had higher healthcare expenditures than non-SHs, both for all and specialized care. Similar results have also been reported in several previous studies (14,26,27). Two studies conducted in Korea targeting joint diseases and spine care, demonstrated that SHs had a higher healthcare expenditure (34.6% and 27.4%, respectively) than non-SHs (26,27), Another previous study on SHs targeting colorectal-anal specialty hospitals found that the inpatient charges of SHs were 27% higher than those of the control group, non-SHs (28). Our study found that SHs had 24.8% and 13.0% higher healthcare expenditures in all claims and two specialized (joint and spine) care sections, respectively. These results suggest that specialty care is associated with high medical prices. In the national health insurance program, the Korean government provides additional reimbursement payments to SHs, such as higher relative value scales for physician’s consultation fees. An empirical study has shown that SHs use healthcare differentiation strategies to maintain high healthcare prices (17). Thus, our study findings correlated with those of previous studies. However, further studies are warranted on the volume of diagnostic tests or the use of advanced medical devices for SHs compared to non-SHs regarding healthcare expenditure.

Strengths and limitations

This study had several limitations. First, this study selected only two types of SHs specializing in joint and spine care among various SHs. Although the two service areas (spine and joint care) are similar, this study has limited generalizability due to the small sample size. In addition, the generalizability of the study results may be limited to these two specialty areas. Second, this study did not set a unit of analysis for individual patients but for hospitals. All patient features are aggregated to the hospital level, which eliminates certain variations arising from patients. Although this approach loses some information from patient factors, it has strengths in the interpretation of results by focusing on SHs versus non-SHs. Finally, this study was conducted in Korea using national health insurance programs. Thus, the interpretation of the results may be limited to hospitals in Korea. However, if there are any nations running similar programs, this result would provide valuable comparative figures to those countries and valuable insights when they plan to introduce similar programs.

Despite these limitations, this study has several strengths. First, one of the key strengths is that the study results were based on quantitative empirical health insurance claims data. This study included national data on joint and spine care, thereby increasing the statistical power. Second, the study results are compatible with those of previous studies and clearly demonstrate the reduction in the ALOS over the past decade in Korea. This was possible because a study conducted 10 years ago investigated the ALOS and healthcare expenditure of SHs compared to non-SHs using the same research participants and topics. Thus, this study is valuable as it allows the comparison of the target outcome variables with previous ones. Third, it has been more than 10 years since the SHs program was introduced in Korea. During this time, several studies on SHs have been published in journals. However, no studies on SHs have been conducted or published in the past five years. Although this study focused on joint and spine diseases, it is meaningful because this study provides an opportunity to examine the overall performance of SHs using the latest data. The results of this study provide various comparative and valuable statistics on SHs that will contribute to expanding the efficiency and activities of SHs in healthcare markets.

This study suggests three implications. First, this study supports previous research findings that SHs are efficient. When studies conducted for the same purpose in different environments produce the same research results, we can generalize the results. This study contributes to the generalization of the fact that SHs are superior to general hospitals in terms of efficiency. Second, it is necessary as a health policy to maintain an appropriate level of supply for SHs as they demonstrate managerial efficiency in resource utilization such as ALOS. Third, given that healthcare expenditure at SHs is higher than those at non-SHs, in-depth research is further needed to know whether patients at SHs have higher severity of illness than those at non-SHs and how SHs achieve managerial efficiency.

Conclusions

This study confirmed that SHs performed better than non-SHs in terms of the ALOS. They had a longer TLOS but shorter ALOS than non-SHs. Although this study selected two types of SHs specializing in joint and spine care, the study results were based on a robust amount of health insurance claims data and quantitative statistical analysis. This conclusion is consistent with those of several previous studies. SHs may provide healthcare with intensive resources; thus, their managerial efficiency may be immeasurable. However, it is also necessary to verify whether SHs provide better-quality care. Although limited generalizability exists due to the targeting of two types of specialty care, this study provides valuable information on SHs and political grounds on why our medical delivery systems should run these kinds of SH programs.

Acknowledgments

We would like to extend our sincere gratitude to the HIRA Research Institute for providing their research data and invaluable support to this study.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jhmhp.amegroups.com/article/view/10.21037/jhmhp-24-114/rc

Data Sharing Statement: Available at https://jhmhp.amegroups.com/article/view/10.21037/jhmhp-24-114/dss

Peer Review File: Available at https://jhmhp.amegroups.com/article/view/10.21037/jhmhp-24-114/prf

Funding: This research was supported by the Health Insurance Review and Assessment Service (HIRA) in Korea through their internal research funds (fund number is not available).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jhmhp.amegroups.com/article/view/10.21037/jhmhp-24-114/coif). All authors currently are employees of HIRA and the manuscript was prepared through their voluntary daily activities. The authors have no other 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.

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. United States Department of Health and Human Services(Centers for Medicare & Medicaid Services). Final Report to Congress Implementing the DRA Provision Affecting Specialty Hospitals. Aug 08, 2006. Available online: https://www.cms.gov/newsroom/fact-sheets/final-report-congress-implementing-dra-provision-affecting-specialty-hospitals (Accessed on July 7, 2024).
2. Carey K, Burgess JF Jr, Young GJ. Specialty and full-service hospitals: a comparative cost analysis. Health Serv Res 2008;43:1869-87. [Crossref] [PubMed]
3. MedPAC (USA). Report to the Congress: Physician-Owned Specialty Hospitals, March 2005. Available online: https://www.cms.gov/outreach-and-education/medicare-learning-network-mln/mlnproducts/downloads/recommendations_physownedspechosp.pdf (Accessed on July 15, 2024)
4. Cause IQ. Specialty hospitals. Available online: https://www.causeiq.com/directory/specialty-hospitals-list/ (Accessed on June 10, 2024)
5. NHS England. Specialised services. Available online: https://www.england.nhs.uk/commissioning/spec-services/ (Accessed on July 20, 2024)
6. Hurd R. Specialist hospitals have a vital role in quality healthcare delivery. March 28, 2024. (Health Service Journal (HSJ). Available online: https://www.hsj.co.uk/leadership/specialist-hospitals-have-a-vital-role-in-quality-healthcare-delivery/5068935.article (Accessed on May 5, 2024)
7. Miura M. Navigating the Differences Between Hospitals and Clinics in Japan. December 20, 2023. Available online: https://ubiehealth.com/company/blog/navigating-the-differences-between-hospitals-and-clinics-in-japan (Accessed on March 11, 2024)
8. Song N, Wang CL, Zhang LQ, et al. A study on the correlation between work stressors and the coping styles of outpatients and emergency nurses in 29 pediatric specialty hospitals across China. Front Psychol 2022;13:951671. [Crossref] [PubMed]
9. Shi J, Tang L, Jing L, et al. Disparities in mental health care utilization among inpatients in various types of health institutions: a cross-sectional study based on EHR data in Shanghai, China. BMC Public Health 2019;19:1023. [Crossref] [PubMed]
10. Ministry of Health and Welfare. Press release (Republic of Korea) (Korean). A total of 109 specialized hospitals to be operated by 2024. Available online: https://www.mohw.go.kr/board.es?mid=a10503000000&bid=0027&act=view&list_no=1479596&tag=&nPage=1 (Access on November 1, 2024).
11. Carey K, Burgess JF, Young GJ. Single specialty hospitals and service competition. Inquiry 2009;46:162-71. [Crossref] [PubMed]
12. Barro JR, Huckman RS, Kessler DP. The effects of cardiac specialty hospitals on the cost and quality of medical care. J Health Econ 2006;25:702-21. [Crossref] [PubMed]
13. Kim JE, Hahm MI. Healthcare consumers’ perceptions of specialty hospitals through a focus group interview. HIRA Research 2023;3:37-51. [Crossref]
14. Satiani B. Specialty hospitals: who do they help? Surgery 2008;143:589-98. [Crossref] [PubMed]
15. Samiedaluie S, Verter V. The impact of specialization of hospitals on patient access to care; a queuing analysis with an application to a neurological hospital. Health Care Manag Sci 2019;22:709-26. [Crossref] [PubMed]
16. Zepeda ED, Nyaga GN, Young GJ. On the relations between focus, experience, and hospital performance. Health Care Manage Rev 2021;46:289-98. [Crossref] [PubMed]
17. McCarthy IM, Raval MV. Price spillovers and specialization in health care: The case of children's hospitals. Health Econ 2023;32:2408-23. [Crossref] [PubMed]
18. Lindlbauer I, Schreyögg J. The relationship between hospital specialization and hospital efficiency: do different measures of specialization lead to different results?. Health Care Manag Sci 2014;17:365-78. [Crossref] [PubMed]
19. Choi JH, Park I, Jung I, Dey A. Complementary effect of patient volume and quality of care on hospital cost efficiency. Health Care Manag Sci 2017;20:221-31. [Crossref] [PubMed]
20. Smith A. An Inquiry into the Nature and Causes of the Wealth of Nations. The Glasgow edition. Indianapolis: LibertyClassics; 1981.
21. Littek W. Division of Labor. In: Smelser NJ, Baltes PB, editors. International Encyclopedia of the Social & Behavioral Sciences. Pergamon; 2001:8220-6.
22. OECD. Length of Hospital Stay. 2024. doi: 10.1787/8dda6b7a-en.10.1787/8dda6b7a-en
23. Leider M, Campbell R, Boyce J, et al. Orthopaedic Specialty Hospitals Compared with General Hospitals: A Systematic Review of Demographic, Clinical, and Financial Parameters. JBJS Rev 2021; [Crossref] [PubMed]
24. Nallamothu BK, Rogers MA, Chernew ME, et al. Opening of specialty cardiac hospitals and use of coronary revascularization in medicare beneficiaries. JAMA 2007;297:962-8. [Crossref] [PubMed]
25. Lee KS, Chun KH, Lee JS. Reforming the hospital service structure to improve efficiency: urban hospital specialization. Health Policy 2008;87:41-9. [Crossref] [PubMed]
26. Kim SJ, Park EC, Jang SI, et al. An analysis of the inpatient charge and length of stay for patients with joint diseases in Korea: specialty versus small general hospitals. Health Policy 2013;113:93-9. [Crossref] [PubMed]
27. Kim SJ, Yoo JW, Lee SG, et al. Governmental designation of spine specialty hospitals, their characteristics, performance and designation effects: a longitudinal study in Korea. BMJ Open 2014;4:e006525. [Crossref] [PubMed]
28. Kim SJ, Lee SG, Kim TH, et al. Healthcare Spending and Performance of Specialty Hospitals: Nationwide Evidence from Colorectal-Anal Specialty Hospitals in South Korea. Yonsei Med J 2015;56:1721-30. [Crossref] [PubMed]
29. Provenzale D, Ofman J, Gralnek I, et al. Gastroenterologist specialist care and care provided by generalists--an evaluation of effectiveness and efficiency. Am J Gastroenterol 2003;98:21-8. [PubMed]
30. Liu Y, Zhong L, Yuan S, et al. Why patients prefer high-level healthcare facilities: a qualitative study using focus groups in rural and urban China. BMJ Glob Health 2018;3:e000854. [Crossref] [PubMed]
31. 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]
32. Jiang S, Gu Y, Yang F, et al. Tertiary hospitals or community clinics? An enquiry into the factors affecting patients' choice for healthcare facilities in urban China, China Economic Review 2020;63:101538.
33. Kim SJ, Kim SJ, Han KT, et al. Medical costs, Cesarean delivery rates, and length of stay in specialty hospitals vs. non-specialty hospitals in South Korea. PLoS One 2017;12:e0188612. [Crossref] [PubMed]
34. Choi BK, Park YT, Park HA, et al. Factors of quality of care and their association with smartphone based PHR adoption in South Korean hospitals. BMC Med Inform Decis Mak 2021;21:296. [Crossref] [PubMed]
35. Mehra T, Müller CT, Volbracht J, et al. Predictors of High Profit and High Deficit Outliers under SwissDRG of a Tertiary Care Center. PLoS One 2015;10:e0140874. [Crossref] [PubMed]
36. Liechti FD, Beck T, Ruetsche A, et al. Development and validation of a score to assess complexity of general internal medicine patients at hospital discharge: a prospective cohort study. BMJ Open 2021;11:e041205. [Crossref] [PubMed]
37. Manning WG, Mullahy J. Estimating log models: to transform or not to transform? J Health Econ 2001;20:461-94. [Crossref] [PubMed]
38. Glick H. Identifying an Appropriate Link and Family for Generalized Linear Models. ISPOR 20th Annual International Meeting. May 19 2015.