Technically efficient surgeons provide shorter active clinical services in a university hospital setting, based on data collected from a leading university hospital in Japan. Learn what it means for other HCOs.
ABSTRACT: Technically efficient surgeons provide shorter active clinical services in a university hospital setting, based on data collected from a leading university hospital in Japan. Output-oriented data models calculated each surgeon’s technical efficiency, and six control variables were factored in. The outcome: Efficiency scores had a significantly negative association with length of active clinical services, while experience and surgical volume had a significantly positive association.
Operating room efficiency is an important concern for physician leaders in most hospitals.1 This measurement depends upon surgeons’ technical efficiency because they usually use the biggest portion of the operating room time. Recruiting and retaining technically efficient surgeons are keys for hospitals to survive in an increasingly fierce health care market competition. However, it might be difficult for physician leaders at university hospitals to recruit and retain technically efficient surgeons because their missions include not only clinical services but also teaching and research. But no study has evaluated retention of technically efficient surgeons in university hospitals. We hypothesized that technically efficient surgeons do not continue to provide active clinical services in a university hospital over the long term.
Data envelopment analysis (DEA) is a measure of technical efficiency that takes account of multiple inputs and outputs. It has been used to measure technical efficiency of individual surgeons as well as health care entities.2-6 A surgeon’s technical efficiency can be measured by efficiency scores calculated from DEA after defining decision-making units and their inputs and outputs.4-6 The purpose of this study is to determine the relationship between length of active clinical services and surgeons’ technical efficiency by using actual surgical data.
Data: Teikyo University Hospital, whose institutional review board approved our study, is located in metropolitan Tokyo, serving a population of approximately 1 million. It has 1,152 beds and 13 surgical specialty departments. We collected data from the electronic medical record system about all the surgical procedures performed in the hospital’s main operating rooms between April 1 and Sept. 30 for the years 2013-17. (Our budget constraints allowed us to collect data for only six months each year.)
Exclusion criteria for surgery were as follows: First, surgical procedures performed under local anesthesia by surgeons were excluded. Second, oral and dermatologic surgical procedures were excluded because most of these cases were minor surgeries that are clinically different from major surgeries. Third, the surgical procedures were excluded if the patients died within a month after surgery, to maintain a constant quality outcome of surgery. Fourth, the surgical procedures were excluded if records were incomplete for any reason.
Efficiency scores (independent variable): The method to calculate surgeons’ technical efficiency was similar to that described in our previous studies.4-6 We regarded technically efficient surgeons as those who maximize their output while minimizing their input use. We employed the output-oriented Charnes-Cooper-Rhodes model of DEA under the constant returns-to-scale assumptions.7 A decision-making unit (DMU) is defined as the entity considered responsible for converting inputs into outputs.8 We defined the DMU as the surgeon with the highest academic rank who scrubbed. All the inputs and outputs are under the control of a DMU.
Inputs were defined as the number of physicians who assisted surgery, and the time of surgical operation from incision to closure. The output was defined as the surgical fee for each surgery. Japan has a universal health insurance system, and most surgeons are reimbursed on a fee-for-service basis according to the fee schedule that sets prices uniformly at the national level.9 It is classified as K000-K915 in the Japanese surgical fee schedule and is called “K codes.” Each surgical procedure is assigned to a K code that corresponds with surgical fees.10-12 The fee is identical regardless of who performs surgery, how many assistants they use or how long it takes to complete surgery. Other fees — such as those for blood transfusion, medications, special insurance, medical materials and anesthesia — were excluded.
We considered all of the inputs and outputs of the surgical procedures for each DMU in each year and calculated his/her efficiency scores using DEA Solver Pro software (Saitech, Tokyo).13 The efficiency scores all lie between 0 and 1; the most efficient surgeons have a score of 1. All surgeons in the sample received an efficiency score for each year.2-6 We used as independent variables the mean efficiency scores of years when they performed surgery as senior surgeons.
Dependent variable: This was defined as a length of active clinical services of each surgeon. It was calculated as the number of years a surgeon was a senior surgeon. We analyzed the data from 2013 through 2017. Therefore, the dependent variable ranges between 1 and 5.
Control variables: We selected six control variables available to us that might influence surgeons’ length of clinical services. We previously demonstrated they were independent from surgeons’ efficiency scores.6 They are:
- Experience and medical school: Most surgeons publish the medical schools that they attended and their years of graduation in directories and/or websites.14-18 Their experience was defined as the number of years since medical school graduation on the date of their last surgical procedure. Seven former imperial universities are highly regarded among medical schools in Japan. They attract the brightest candidates, and their entrance examination is notoriously competitive every year. Surgeons who graduated from them are expected to be more academically oriented than those who did not. We classified medical schools by whether they are former imperial universities. If so, we assigned a dummy variable of 1. If not, we assigned a dummy variable of 0.
- Average surgical volume per six months: This was defined as the mean of surgical cases that a surgeon performed during the six-month period in each year. This information was extracted from the electronic medical record system.
- Gender and academic ranks: Teikyo University Hospital publishes surgeons’ gender and academic rank on its website for patients’ convenience.18 We assigned a dummy variable of gender; female equaled 1 and male equaled 0. Academic ranks during the last year a surgeon performed surgery were recorded from the Teikyo University Hospital website. We assigned two dummy variables of academic ranks; full professor equaled 1 and otherwise equaled 0.
We used Stata software (Stata 14, StataCorp, Texas) for our statistical analysis. We performed multiple regression analysis using ordinary least squares and ordered logit models.19,20 A p-value of less than 0.05 was considered statistically significant.
We analyzed 13,911 surgical cases in a 30-month study period, 2013-17. The number of surgeons analyzed was 288. Efficiency scores were calculated for all of them in each year, and means for each surgeon were calculated.
The characteristics of the dependent and independent variables are shown in Table 1. We could obtain information on medical schools and experience from only 229 and 215 surgeons, respectively.
We performed multiple regression analysis for 211 surgeons who published information about both their medical schools and their work experience. The results of ordinary least squares model multiple regression analysis are shown in Table 2. Efficiency scores had significantly negative association with length of clinical services (p-value equals 0.011), while experience and surgical volume had positive association (p-value of 0.000 for both). The coefficients of medical schools, gender, and academic ranks were statistically insignificant (p-values greater than 0.05).
The results of ordered logit model multiple regression analysis are shown in Table 3. Efficiency scores had significantly negative association with length of clinical services (p-value equals 0.020), while experience and surgical volume had positive association (p-value of 0.000 for both). The coefficients of medical schools, gender and academic ranks were statistically insignificant (p-values greater than 0.05).
From our ordinary least squares and ordered logit models multiple regression analysis, we demonstrated that technically efficient surgeons have shorter length of active clinical services at a university hospital. The longer their experiences were and the larger their surgical volumes were, the longer they provide clinical services in a university hospital. Their medical school, gender and academic ranks provided insignificant predictive values for their length of clinical services. Both statistical models reached the same results, which indicate that they were robust.
There are several possible reasons for our findings. First, university hospitals aim to achieve not only clinical but also academic and educational excellence. Technically efficient surgeons who focus only on clinical excellence might not fit the missions of university hospitals and choose to leave them. Second, university hospitals serve as educational institutes. It is possible that surgeons work in university hospitals until they become efficient, and then seek employment in private practice, which often provides better monetary rewards. Third, proficiency in teaching and research does not necessarily correspond with clinical excellence. Surgeons who meet the academic and educational criteria of university hospitals may not be as technically efficient as surgeons outside this environment.
Surgeons with long experience provide longer active clinical services in our university hospital. This might be because surgeons with long experience usually are older and are likely to have an established academic career. In addition, they might have few employment opportunities outside a university hospital, which might have led to their longer tenures. Small surgical volume led to the shorter length of active clinical services. Less-active surgeons have few surgical opportunities at a university hospital, and they were likely to seek their patients outside a university hospital.
There are some limitations in our study. First, this is a study conducted in a single large teaching hospital in Tokyo. Our hospital and surgeons may not represent all surgeons everywhere. However, there is an advantage to studying surgeons’ technical efficiency in a single hospital: Because one of the significant resource inputs is ancillary services, such as operating room nursing practices and availability of support personnel, all these factors are held constant in a single hospital. Comparing surgeons in different hospitals can be misleading if some ancillary services are more efficient than others. By comparing surgeons in the same institution, they all face the same systemic advantages and disadvantages of ancillary services.21
Second, we evaluated only surgeons’ technical efficiency in the operating rooms, not their academic and educational achievements. Academic ranks might serve as a proxy variable for these achievements although they did not have significant effects on the length of employment. It would be difficult to identify the exact causes of our findings without this information. However, the exact data on their academic and educational achievements were unavailable to us.
Third, we regarded being a senior surgeon as active clinical services. This is not necessarily the case. Some surgeons stop doing surgery to spend more time in research, education, administration or outpatient clinics. However, from the viewpoint of physician leaders in charge of a university hospital, this change is similar to a resignation because the surgeon no longer generates significant revenue for the hospital in the operating rooms.
In conclusion, we demonstrated that technically efficient surgeons have shorter lengths of active clinical services at a university hospital.
Yoshinori Nakata, MD, MBA, is a professor at Teikyo University Graduate School of Public Health and director of Teikyo University Medical Information and System Research Center in Tokyo, Japan. He is a member of the American Association for Physician Leadership editorial board.
Yuichi Watanabe, MS, MPH, is a doctoral candidate at Waseda University Graduate School of Economics in Tokyo.
Sayaka Horiuchi, MD, MPH, is an assistant professor at Teikyo University Graduate School of Public Health in Tokyo.
Hiroshi Otake, MD, MBA, is a professor and chair of the Department of Anesthesia at Showa University School of Medicine in Tokyo.
Hiroto Narimatsu, MD, PhD, is director of the Cancer Prevention and Information division at Kanagawa Cancer Center Research Institute in Yokohama, Japan.
Tomohiro Sawa, MD, PhD, is a professor at Teikyo University Medical Information and System Research Center in Tokyo.
- Eappen S, Flanagan H, Bhattacharyya N. Introduction of anesthesia resident trainees to the operating room does not lead to changes in anesthesia-controlled times for efficiency measures. Anesthesiology. 101(5):1210-4, November 2004.
- Hollingsworth B, Peacock SJ. Efficiency measurement in health and health care. Oxon: Routledge; 2008.
- Ozcan YA. Health care benchmarking and performance evaluation: an assessment using data envelopment analysis. 2nd ed. New York: Springer; 2014.
- Nakata Y, Watanabe Y, Otake H, Nakamura T, Oiso G, Sawa T. The Japanese surgical reimbursement system fails to reflect resource utilization. Int J Health Serv. 45(4):801-9, May 2015.
- Nakata Y, Yoshimura T, Watanabe Y, Otake H, Oiso G, Sawa T. Resource utilization in surgery after the revision of surgical fee schedule in Japan. Int J Health Care Qual Assur. 28(6): 635-43, 2015.
- Nakata Y, Watanabe Y, Narimatsu H, Yoshimura T, Otake H, Sawa T. Predictors of surgeons’ efficiency in the operating rooms. Health Serv Manag Res. 30(1):16-21, February 2017
- Farrell MJ. The measurement of productive efficiency. J Royal Stat Soc, Series A (General). 120(3):253-281, 1957.
- Thanassoulis E, Portela MCS, Despic O. Data envelopment analysis: the mathematical programming approach to efficiency analysis. In: Fried HO, Lovell CAK, Schmidt SS, eds. The measurement of productive efficiency and productivity growth. New York: Oxford University Press; 2008:251-420.
- Ikegami N, Yoo BK, Hashimoto H, et al. Japanese universal health coverage: evolution, achievements, and challenges. Lancet. 78(9796): 1106-14, September 2011.
- Social Insurance Institute (Shakaihoken Kenkyuusho). Interpretation of medical reimbursement scales, April 2012 (Ikatensuuhyou no Kaishaku). 2012 [in Japanese].
- Social Insurance Institute (Shakaihoken Kenkyuusho). Interpretation of medical reimbursement scales, April 2014 (Ikatensuuhyou no Kaishaku). 2014 [in Japanese].
- Social Insurance Institute (Shakaihoken Kenkyuusho). Interpretation of medical reimbursement scales, April 2016 (Ikatensuuhyou no Kaishaku). 2016 [in Japanese].
- Cooper WW, Seiford LM, Tone K. Data envelopment analysis: a comprehensive text with models, applications, references and DEA-Solver software, 2nd ed. New York: Springer; 2007.
- Faculty Members’ Directory of Japanese Teaching Medical Institutions 2014-15. (Iikukikanmeibo 2014-15). Tokyo: Yodosha, 2014 [in Japanese].
- Faculty Members’ Directory of Japanese Teaching Medical Institutions 2015-16. (Iikukikanmeibo 2015-16). Tokyo: Yodosha, 2015 [in Japanese].
- Faculty Members’ Directory of Japanese Teaching Medical Institutions 2016-17. (Iikukikanmeibo 2016-17). Tokyo: Yodosha, 2016 [in Japanese].
- Faculty Members’ Directory of Japanese Teaching Medical Institutions 2017-18. (Iikukikanmeibo 2017-18). Tokyo: Yodosha, 2017 [in Japanese].
- Teikyo University, http://www.teikyo-hospital.jp/medical/surgery/index.html. Accessed 3 December 2017 [in Japanese].
- Stock JH, Watson MW. Introduction to econometrics, updated 3rd ed. Essex: Pearson; 2015.
- Wooldridge JM. Introductory econometrics, 4th ed. Boston: Cengage Learning; 2009.
- Chilingerian JA. Exploring why some physicians’ hospital practices are more efficient: taking DEA inside the hospital. In: Charnes A, Cooper W, Lewin AY, Seiford LM, eds. Data envelopment analysis: theory, methodology, and application. New York: Springer, 1994, pp.167-194.
This work was supported by the Japan Society for the Promotion of Science, KAKENHI Grant Number 17K09247 to Yoshinori Nakata.