Abstract
Japan’s healthcare expenditures, which are largely publicly funded, have been growing dramatically due to the rapid aging of the population as well as the innovation and diffusion of new medical technologies. Annual costs for surgical treatments are estimated to be approximately USD 20 billion. Using unique longitudinal clinical data at the individual surgeon level, this study aims to estimate the technical efficiency of surgical treatments across surgical specialties in a high-volume Japanese teaching hospital by employing stochastic frontier analysis (SFA) with production frontier models. We simultaneously examine the impacts of potential determinants that are likely to affect inefficiency in operating rooms. Our empirical results show a relatively high average technical efficiency of surgical production, with modest disparity across surgical specialties. We also demonstrate that an increase in the number of operations performed by a surgeon significantly reduces operating room inefficiency, whereas the revision of the fee-for-service schedule for surgical treatments does not have a significant impact on inefficiency. In addition, we find higher technical efficiency among surgeons who perform multiple daily surgeries than those who perform a single operation in a day. We suggest that it is important for hospital management to retain efficient surgeons and physicians and provide efficient healthcare services given the competitive Japanese healthcare market.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Out of total healthcare expenditures (about USD 350.0 billion) in FY 2015, public funding accounted for 38.9% (about USD 136.1 billion), insurance premiums were 48.8% (about USD 170.8 billion), and the rest, including patient funding, amounted to 12.3% (about USD 43.1 billion) [2]. All expenditures are calculated as 100 Japanese yen = 0.826173 USD, which is the period average exchange rate for 2015.
The Japanese Medical Service Law defines two types of medical facilities: hospitals and clinics. A “hospital” is a medical facility with 20 or more beds, while a “clinic” has less than 20 beds or no beds at all.
Hospitalization charges refer to admission and management fees for various types of inpatient care.
DPC/PDPS refers to a flat-rate payment system for each hospital stay based on diagnosis group. To control rising costs, MHLW introduced DPC/PDPS as of April 2003 [3]. The ratio of hospital beds under DPC/PDPS increased gradually from 7.7% in FY 2003 to more than 50% in FY 2010, and it has remained at approximately 50–55% since [4].
Since SMCAPHI is based on only one month’s medical records (every June), total spending for all surgical treatments for one year is unfortunately not available. Assuming the ratio of surgical fees per day (15.6%) is constant for all hospitalizations, we estimate the total amount of surgical spending by multiplying total inpatient care (about USD 128.7 billion) by 0.156.
Medical equipment and ancillary services (e.g., nursing practices and availability of support personnel) in operating rooms are considered to comprise a significant resource environment. These factors are set to be identical across operating rooms and thus held constant in a single hospital setting.
Surgeons in this study belong to one of the following 12 surgical specialties: thoracic surgery, cardiovascular surgery, neurosurgery, obstetrics and gynecology, ophthalmology, plastic surgery, orthopedics, general surgery, pediatric surgery, urology, emergency surgery, and otorhinolaryngology.
An operation’s start time occurs at skin incision and the end time occurs at skin closure. The operating rooms officially run from 8:30 to 17:00 on weekdays and 8:30 to 12:30 on Saturday.
Patient’s sex is defined as a dummy variable taking the value of 1 if sex is female, and the value of 0 if sex is male.
Patient’s prognostic information on in-hospital mortality refers to whether a patient dies within one month after surgery. It is considered a proxy for the severity of a patient’s condition.
In the fee schedule, K codes are attached to medical practices such as surgeries and treatments. They evaluate different types of surgical procedures and enable us to compute the total reimbursements for heterogeneous surgical procedures.
Certain percentages are added to the surgical fee when a patient is admitted to the hospital outside of regular business hours (e.g., in the middle of the night or on holidays) and goes directly into surgery.
The Japanese Joint Committee of Social Insurance by the Multidisciplinary Group of Surgical Associations, which is often called “Gaihoren” in Japanese, attempts to establish a fair scale of surgical fee reimbursement by revealing the original cost and markup methods. It reports the expected level of technical difficulty and the estimated costs for the number of attending staff and the duration of each surgical procedure. It then provides an approximate estimate for the final price [20].
Most cases that take less than five minutes of surgical time are minor surgeries performed on infants or young children under general anesthesia (e.g., removing a foreign body from the esophagus or removing a subcutaneous tumor).
Reinhardt [22, 23] justifies the use of patient billings as an output index by presupposing that the medical fees assigned to particular services are closely related to the relative factor costs of these services, which is supported by evidence that relative physician fees vary as a function of relative factor costs.
Patient’s age is categorized into 10 groups with 10 years per group (e.g., 0–9, 10–19, 20–29, etc.), then these age groups are transformed into dummy variables. The reference group is determined as 0–9.
If one assumes that the inefficiency terms follow a non-negative half-normal distribution (i.e., \( {u}_i^{CD,T}\sim i.i.d.{N}^{+}\left(0,{\sigma}_{u^{CD,T}}^2\right) \)), then the assumption of heteroskedastic inefficiency terms allows us to model linear variance functions of a set of exogenous inefficiency determinants instead as \( {\sigma}_{u_i^{CD,T}}^2=\exp \left({\psi}_0^{CD,T}+{\boldsymbol{Z}}_i^{\prime }{\boldsymbol{\psi}}_{CD,T}\right) \). As there is no a priori justification for the use of any particular distribution for the inefficiency terms [26], the hypothesis of a half-normal distribution should be statistically tested against the assumptions of a truncated-normal distribution.
Surgical volume is taken as the logarithm in the estimation. Other than that, we also considered the inclusion of surgeons’ accumulated clinical experience, defined as the number of years since medical school graduation on the date of surgical procedure or their current academic rank. However, this information is only partly available to us; thus, the use of this variable would lead to a serious issue of selection bias.
The revision of the surgical reimbursement system is simply a dummy variable taking the value of 1 during the period from April to September for the years 2016 and 2017, and the value of 0 during the period from April to September for the years 2014 and 2015.
Surgeons performing fewer than five surgical cases in a day are assigned from the first time slot of the day. For example, a surgeon who performs three surgeries in a day is assigned to the first three slots of the day. Note that each time slot does not correspond to the actual time of day, rather it represents the sequence of surgical operations in a day.
One of the benefits of using panel data is being able to control for unobservable individual heterogeneity. It is tempting to employ the “true” fixed-effects or “true” random-effects models proposed by Greene [29] instead, which disentangle time-varying inefficiency from unit-specific time-invariant unobserved heterogeneity. We discuss the application of these models later.
The Wald tests also reject the hypothesis of constant returns to scale, which assumes that the sum of the output elasticities of capital and labor turns out to be one.
We can reject the hypothesis that there is no significant difference in means among surgical specialties based on the one-way analysis-of-variance (ANOVA) and Kruskal-Wallis H tests.
There were no major changes in the surgical fee schedule when the revision of the fee schedule was implemented in April 2016; most changes were focused on division of functions among healthcare facilities and nursing care. There were some minor changes, for example, reimbursement for an emergency Caesarean section increased 10.2% in April 2016.
The Wald tests reject the hypothesis of constant returns to scale again.
We can also reject the hypothesis that there is no significant difference in means among surgical specialties based on the one-way ANOVA and Kruskal-Wallis H tests.
While our estimates are lower than the results in [16, 17], which show extremely high technical efficiency (over 0.97 on average), their results may not be appropriate for comparison; as indicated in Section 1, Introduction, the results of [16, 17] are questionable from practical and clinical perspectives.
The former model includes unit-specific dummies as an additional set of explanatory variables to capture unobserved individual heterogeneity, and the latter model treats the unit-specific intercepts as random variables [29].
The output elasticities of capital and labor were estimated to be over 1.0 in the “true” fixed-effects model with the assumption that the inefficiency term follows a non-negative truncated-normal or exponential distribution. Applying the “true” random-effects model to our sample did not result in successful convergence.
There are several requirements for approval of special-functioning hospitals: providing medical care to patients who are referred by other hospitals or clinics; having 400 or more beds; strict staff deployment (e.g., there must be twice as many doctors as in ordinary hospitals); having medical facilities such as intensive care units, sterile rooms, and drug information management rooms; improvement of a medical safety management system; recognizing 16 specified clinical areas in principle; and so forth [31]. There are 85 approved special functioning hospitals as of June 2017.
References
Ikegami N, Yoo BK, Hashimoto H, Matsumoto M, Ogata H, Babazono A, Watanabe R, Shibuya K, Yang BM, Reich MR, Kobayashi Y (2011) Japanese universal health coverage: evolution, achievements, and challenges. Lancet 378(9796):1106–1115
Ministry of Health, Labour and Welfare (2017a) Estimates of National Medical Care Expenditure, FY 2015. https://www.mhlw.go.jp/english/database/db-hss/dl/digest2015.pdf.
Ishii M (2012) DRG/PPS and DPC/PDPS as prospective payment systems. Japan Medical Association Journal 55(4):279–291
Ministry of Health, Labour and Welfare (2015) Summary of static/dynamic survey of medical institutions and hospital report, 2014. https://www.mhlw.go.jp/english/database/db-hss/mih_report_2014.html.
Ministry of Health, Labour and Welfare (2016) Statistics of medical care activities in public health insurance, 2015. https://www.mhlw.go.jp/toukei/saikin/hw/sinryo/tyosa15/dl/ika.pdf.
Hollingsworth B (2003) Non-parametric and parametric applications measuring efficiency in health care. Health Care Management Science 6(4):203–218
Hollingsworth B (2008) The measurement of efficiency and productivity of health care delivery. Health Econ 17(10):1107–1128
Worthington AC (2004) Frontier efficiency measurement in health care: a review of empirical techniques and selected applications. Med Care Res Rev 61(2):135–170
Chilingerian JA, Sherman HD (2011) Health-care applications: from hospitals to physicians, from productive efficiency to quality Frontiers. In: Cooper WW, Seiford LM, Zhu J (eds) Handbook on data envelopment analysis, 2nd edn. Springer, New York, pp 445–493
Kawaguchi H, Hashimoto H, Matsuda S (2010) Study about evaluation of hospital efficiency and function used by DPC data set. Iryo to Shakai 20(1):23–34 [in Japanese]
Besstremyannaya G (2011) Managerial performance and cost efficiency of Japanese local public hospitals: a latent class stochastic frontier model. Health Econ 20(S1):19–34
Nakata Y, Watanabe Y, Otake H, Nakamura T, Oiso G, Sawa T (2015a) The Japanese surgical reimbursement system fails to reflect resource utilization. Int J Health Serv 45(4):801–809
Nakata Y, Watanabe Y, Otake H, Nakamura T, Oiso G, Sawa T (2015b) Resource utilization in surgery after the revision of surgical fee schedule in Japan. International Journal of Health Care Quality Assurance 28(6):635–643
Nakata Y, Watanabe Y, Otake H, Nakamura T, Oiso G, Sawa T (2017) Predictors of surgeons’ efficiency in the operating rooms. Health Serv Manag Res 30(1):16–21
Koch SF, Slabbert JD (2012) Stochastic frontier analysis of specialist surgeon clinics. South African Journal of Economics 80(1):1–15
Heimeshoff M, Schreyögg J, Kwietniewski L (2014) Cost and technical efficiency of physician practices: a stochastic frontier approach using panel data. Health Care Management Science 17(2):150–161
Kwietniewski L, Schreyögg J (2018) Efficiency of physician specialist groups. Health Care Management Science 21(3):409–425
Japanese Joint Committee of Social Insurance by the Multidisciplinary Group of Surgical Associations (2015) Proposal about surgical payment system by GAIHOREN 2016 (Gaihoren Shian). Igakutushinsya, Tokyo [in Japanese]
Japanese Joint Committee of Social Insurance by the Multidisciplinary Group of Surgical Associations (2017) Proposal about surgical payment system by GAIHOREN 2018 (Gaihoren Shian). Igakutushinsya, Tokyo [in Japanese]
Hayashida K, Imanaka Y (2005) Inequity in the Price of physician activity across surgical procedures. Health Policy 74(1):24–38
Social Insurance Institute (2014) Interpretation of medical reimbursement scales, April 2014 (Ikatensuuhyou no Kaishaku). Shakaihoken Kenkyuusho, Tokyo [in Japanese]
Reinhardt U (1972) A production function for physician services. Rev Econ Stat 54(1):55–66
Reinhardt UE (1975) Physician productivity and the demand for health manpower: an economic analysis. Ballinger Publishing Company, Cambridge
Thurston NK, Libby AM (2002) A production function for physician services revisited. Rev Econ Stat 84(1):184–191
Jacobs R, Smith PC, Street A (2006) Measuring efficiency in health care: analytic techniques and health policy. Cambridge University Press, Cambridge
Rosko MD (2004) Performance of U.S. teaching hospitals: a panel analysis of cost inefficiency. Health Care Management Science 7(1):7–16
Kumbhakar SC, Lovell CAK (2000) Stochastic frontier analysis. Cambridge University Press, Cambridge
Battese GE, Coelli TJ (1995) A model for technical inefficiency effects in a stochastic frontier production function for panel data. Empir Econ 20(2):325–332
Greene W (2005) Reconsidering heterogeneity in panel data estimators of the stochastic frontier model. J Econ 126(2):269–303
Kumbhakar SC, Wang HJ, Horncastle AP (2015) A Practitioner’s guide to stochastic frontier analysis using Stata. Cambridge University Press, Cambridge
Ministry of Health, Labour and Welfare (2017b) Annual health, Labour and Welfare Report 2017. https://www.mhlw.go.jp/english/wp/wp-hw11/dl/02e.pdf.
Acknowledgments
We would like to sincerely thank to Teikyo University Hospital, which allowed us to utilize the valuable data. We also gratefully acknowledge the financial support from Waseda Institute of Social & Human Capital Studies (WISH) for travel expenses to present this paper at an international academic conference, and appreciate the Waseda Institute of Political Economy (WINPEC) for its financial support. Our special thanks go to Dr. Akira Kawamura for providing valuable comments, and Dr. Hiroyuki Kawaguchi for discussing our paper at the 2018 Annual Conference of the Japan Health Economics Association (JHEA). Further, we would like to extend appreciation to those who participated in the 5th European Health Economics Association (EuHEA) PhD Student-Supervisor and Early Career Researcher Conference for their helpful comments and suggestions. We take full responsibility for any errors.
Funding
This work was supported by the Japan Society for the Promotion of Science (JSPS) KAKENHI Grant Number 17 K09247 to Dr. Yoshinori Nakata.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Ethical approval
All procedures performed in this study involving human participants were in accordance with the ethical standards of the institutional and/or national research committee, and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Formal consent from patients was not required. The Teikyo University Institutional Review Board has approved our study. Anonymity of the data has been strictly maintained by de-identification.
Conflict of interest
The authors declare no potential conflicts of interest with respect to the research, authorship, and/or publication of this study. The authors also declare that they have no relevant or material financial interests that relate to the research described in this paper.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Highlights
• We estimate the technical efficiency of surgical operations across surgical specialties in a high-volume Japanese teaching hospital by employing stochastic frontier analysis (SFA) with production frontier models.
• We find a relatively high average technical efficiency of surgical production, with modest disparity across surgical specialties.
• Surgeons with higher surgical volumes exhibit reduced operating room inefficiency, while the revision of the fee-for-service schedule does not have a significant effect.
• Hospital management should retain efficient surgeons and physicians and provide efficient healthcare services given the competitive Japanese healthcare market.
• Efficiently operating teaching hospitals are not only expected to produce skilled and efficient surgeons, but are also expected to help improve fee schedules to reflect surgical efficiency and motivate other medical facilities in reorganizing their clinical management and resource utilization.
Rights and permissions
About this article
Cite this article
Watanabe, Y., Noguchi, H. & Nakata, Y. How efficient are surgical treatments in Japan? The case of a high-volume Japanese hospital. Health Care Manag Sci 23, 401–413 (2020). https://doi.org/10.1007/s10729-020-09507-3
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10729-020-09507-3
Keywords
- Technical efficiency
- Stochastic frontier analysis
- Surgical production
- Operating rooms
- Japanese fee-for-service system