Development of efficiency indicators of operating room management for multi-institutional comparisons

Development of efficiency indicators of operating roommanagement for multi-institutional comparisonsjep_1829 1..7

Masayuki Tanaka MPH,1 Jason Lee PhD,2 Hiroshi Ikai MD PhD3 and Yuichi Imanaka MD MPH PhD4

1Graduate Student, 2Post-Doctoral Fellow, 3Assitant Professor, 4Professor, Department of Healthcare Economics and Quality Management, KyotoUniversity Graduate School of Medicine, Yoshida Konoe-cho, Kyoto, Japan

Keywords

administrative data, assessment ofperformance, efficiency indicators,improvement, multi-institutional comparison,operating room management

Correspondence

Yuichi ImanakaDepartment of Healthcare Economics andQuality ManagementKyoto University Graduate School ofMedicineYoshida Konoe-cho, Sakyo-kuKyoto 606-8501JapanE-mail: [email protected]

Accepted for publication: 8 December 2011

doi:10.1111/j.1365-2753.2012.01829.x

AbstractObjectives The efficiency of a hospital’s operating room (OR) management can affect itsoverall profitability. However, existing indicators that assess OR management efficiency donot take into account differences in hospital size, manpower and functional characteristics,thereby rendering them unsuitable for multi-institutional comparisons. The aim of thisstudy was to develop indicators of OR management efficiency that would take into accountdifferences in hospital size and manpower, which may then be applied to multi-institutionalcomparisons.Methods Using administrative data from 224 hospitals in Japan from 2008 to 2010, weperformed four multiple linear regression analyses at the hospital level, in which thedependent variables were the number of operations per OR per month, procedural fees perOR per month, total utilization times per OR per month and total fees per OR per month foreach of the models.Results The expected values of these four indicators were produced using multipleregression analysis results, adjusting for differences in hospital size and manpower, whichare beyond the control of process owners’ management. However, more than half of thevariations in three of these four indicators were shown to be explained by differences inhospital size and manpower.Conclusion Using the ratio of observed to expected values (OE ratio), as well as thedifference between the two values (OE difference) allows hospitals to identify weak-nesses in efficiency with more validity when compared to unadjusted indicators. The newindicators may support the improvement and sustainment of a high-quality health caresystem.

IntroductionHospital managers, doctors, nurses, medical support staff, admin-istrative clerks and other support staff all play roles in carrying outimprovements in management efficiency [1]. The assessment ofindividual hospitals’ performance in management efficiency is animportant step towards these improvements. Tools such as TotalQuality Management and Balanced Score Cards have been intro-duced to evaluate and improve the efficiency of individual hospitaldepartments, as well as total hospital management [2]. In particu-lar, the management of hospitals’ operating rooms (ORs) has beenthe focus of recent attention [3–7]. Previous studies showing ahigh correlation between the total number of operations in anOR and overall hospitalization fees [7,8], while our analysis ofdata from 153 Japanese hospitals during the period of April toSeptember 2009 revealed that surgery and anaesthetizations feescomprised 28.1% of all hospitalization fees (data not shown). This

highlights the substantial contribution of surgeries to the overallincome of a hospital.

Several studies have emphasized the need for evaluations of ORmanagement efficiency [4–6,8–15]. The assessment of OR perfor-mance has been previously conducted using indicators such as thenumber of operations, the procedural fees per OR, the total utili-zation times per OR and the total fees per OR. However, suchassessments tend to be conducted at an individual hospital level,and such indicators are unable to take into account differencesin hospital size, manpower and functional characteristics, such asdifferences in the number of surgeons, anaesthesiologists andnurses per OR; or the average length of stay (LOS) for eachhospital [7,10–15]. There is therefore a necessity for assessmentindicators that can adjust for differences in structural factorssuch as hospital size and manpower, which are beyond control ofowners’ hospital management, thereby allowing their applicationsin multi-hospital comparisons.

Journal of Evaluation in Clinical Practice ISSN 1365-2753

© 2012 Blackwell Publishing Ltd, Journal of Evaluation in Clinical Practice 1

The aim of this study was to develop indicators of ORmanagement efficiency that would take into account differences inhospital size and manpower, which may then be applied to multi-institutional comparisons.

Methods

Data sources

Data were obtained from the Quality Indicator/ImprovementProject (QIP), which is a programme administrated by the Depart-ment of Healthcare Economics and Quality Management in KyotoUniversity. In this programme, member hospitals from all regionsof Japan voluntarily provide administrative claims data for analy-sis. The objectives of the QIP are (1) to measure objective numericindicators (performance indicators) in which the process, outcomeand economy of diagnoses and treatments can be evaluated; (2) toprovide feedback of these findings to the participant hospitals ona regular basis; and (3) to contribute to the improvement of thequality of health care in hospitals.

The administrative data used were based on the Diagnosis Pro-cedure Combination system, which is a standardized per-diempatient classification system for the purpose of hospital reimburse-ment in Japan. Specifically, data were obtained from the E andF files from the database, which contain detailed information suchas general anaesthesia duration and dosages for all prescribedmedications on a daily basis. In this study, we utilized data frompatients that were admitted to 224 hospitals from April 2008 toMarch 2010.

Information regarding the number of beds, number of staff,average LOS and the number of ORs in each individual situationwere obtained from a yearly survey conducted on all QIP memberhospitals. Data on the operation durations were obtained fromthe Confederation of Social Insurance Committees of SurgicalSocieties (Gaihoren) [16].

Indicator development and analysis

We focused on existing indicators of OR management efficiencythat are currently utilized: number of operations per OR permonth, procedural fees per OR per month, total utilization timesper OR per month and total fees per OR per month [7,8,11,14].These unadjusted rates of each of the indicators were calculatedusing data from the QIP database. Operations performed in eachOR were identified on the basis of operation records of thepast 3 years in the OR and the judgment of specialists from eachdepartment. Anaesthesia utilization duration was used as a proxyfor OR utilization time, with data for epidural anaesthesia, intrath-ecal anaesthesia, intravenous anaesthesia or local anaesthesia cal-culated using Gaihoren data. The associations of indicators of ORmanagement efficiency with hospital size and characteristics wereanalyzed using Pearson’s correlation coefficient.

The 224 hospitals in the sample were randomly divided into twogroups: one group served as the sample for indicator development,and the other was used for verification purposes [17]. To takeinto account inter-hospital variations, we performed four multiplelinear regression analyses at the hospital level using a stepwiseprocedure, in which the dependent variables were (1) the numberof operations per OR per month, (2) procedural fees per OR per

month, (3) total utilization times per OR per month and (4) totalfees per OR per month for each of the models. Independent vari-ables in the regression models were factors related to hospital sizeand characteristics, namely the total number of beds, average LOS,number of nurses per OR, number of ORs, number of surgeonsper OR, number of anaesthesiologists per OR and number ofemergency patients. The validity of the models was verified usingthe verification sample group [17].

In order to develop indicators of OR management efficiency thattook into account variations in hospital size and characteristics,we utilized the abovementioned regression models to calculateexpected values of the number of operations per OR per month,procedural fees per OR per month, total utilization times of per ORper month and total fees per OR per month. We then calculatedthe ratio between the observed and expected values (OE ratio) andthe difference between the observed and expected values (OEdifference) to be used as new indicators [18,19]. An OE ratiogreater than one indicates that the observed value has exceeded theexpected value after taking into account inter-hospital variations,while the ratio less than one indicates that the expected value hasexceeded the observed value. An OE difference less than zeroindicates that the expected value has exceeded the observed valueafter taking into account inter-hospital variations, while the differ-ence greater than zero indicates that the observed value hasexceeded the expected value. Statistical analyses were conductedusing Dr. SPSS II, and statistical significance was set at P < 0.05.All fees were expressed as US dollars using the purchasing powerparity rate for Japanese yen to US dollars in 2010 (¥111 = $1), asstipulated by the Organization for Economic Cooperation andDevelopment National Account database.

ResultsA preliminary analysis revealed a large degree of inter-hospitalvariations in the general indicators of OR management efficiency,hospital size and characteristics (Table 1). The mean of proceduralfees per OR per month was $67 492 [standard deviation (SD):$33 712], and mean of number of operations per OR per monthwas 42 (SD: 16 operations).

Table 2 shows the associations of indicators of OR managementefficiency with hospital size and characteristics using Pearson’scorrelation. The four performance indicators were found to besignificantly associated with hospital size and characteristics(r = 0.277~0.863; P < 0.01). The number of operations showedpositive correlations with the number of surgeons per OR, numberof nurses per OR, total number of beds, number of ORs, numberof anaesthesiologist per OR and number of emergency patients(r = 0.332~0.628; P < 0.01); and negative correlations with theaverage LOS(r = -0.417; P < 0.01). Procedural fees per OR permonth were significantly associated with hospital size and charac-teristics, showing positive correlations with the number of sur-geons per OR, number of nurses per OR, total number of beds,number of ORs, number of anaesthesiologist per OR and numberof emergency patients (r = 0.451~0.863; P < 0.01) negative corre-lations with the average LOS (r = -0.329; P < 0.01).

The results of the multiple linear regression analyses are pre-sented in Table 3. The number of number of surgeons per OR, totalnumber of beds, number of anaesthesiologists per OR and numberof nurses per OR showed significant association with the number

Efficiency indicators of operating room M. Tanaka et al.

© 2012 Blackwell Publishing Ltd2

of operations per OR per month. The regression model showedthat these four variables were able to account for 63% of theobserved inter-hospital variations in the number of operations permonth. When the dependent variable was the procedural feesper OR per month, significant associations were found withthe number of surgeons per OR, total number of beds, number ofanaesthesiologists per OR and number of nurses per OR. Thesevariables were able to account for 56% of the observed inter-hospital variations. When the dependent variable was the totalutilization time per OR per month, significant associations werefound with the number of anaesthesiologists per OR, number ofORs and number of nurses per OR. These variables explained 64%of the observed inter-hospital variations. When the dependent vari-able was the total fees per OR per month, significant associationswere found with the number of nurses per OR, total number ofbeds and number of anaesthesiologists per OR. These variablesexplained 33.8% of the observed inter-hospital variations.

To verify the stability of the four multiple linear regressionmodels, we generated R2 values for similar models using the veri-fication group. In this sample, the R2 values showed that the modelswere able to account for 50%, 48%, 47% and 34.1% of variationsfor the number of operations per OR per month, procedural fees perOR per month, total utilization times per OR per month and totalfees per OR per month, respectively.

As novel indicators of OR management efficiency, the OE ratioand OE difference were calculated for each dependent variable(number of operations per OR per month, procedural fees peroperation per month, total utilization time per OR per month, totalfees per OR per month) using the expected values from the regres-sion models, and the results for number of operations per OR permonth, procedural fees per OR per month, total utilization time perOR per month and total fees per OR per month are shown inFigures 1 and 2. Figure 1 showed that the OE ratio of: the numberof operation per OR per month ranged from 0.29 to 1.93; proce-dural fees per OR per month ranged from 0.25 to 3.1; total utili-zation time per OR per month ranged from 0.18 to 2.08; the totalfees per OR per month ranged from 0.28 to 4.5. Figure 2 showedthat the OE difference of the number of operation per OR permonth ranged from -39.4 to 38.8; procedural fees per OR permonth ranged from $-68 818 to $132 530; total utilization timeper OR per month ranged from -81.62 hours to 62.8 hours; totalfees per OR per month ranged from $-133 677 to $304 851.

DiscussionIn this study, we have developed new indicators to assess ORmanagement efficiency that takes into account variations in hos-pital size and characteristics suitable for multi-institutional com-parisons, and used these indicators to conduct such a comparisonusing administrative data from multiple hospitals in Japan.

As shown in Table 2, we found that existing unadjusted ORperformance indicators (number of operations per OR per month,procedural fees per OR per month, total utilization times per ORper month and total fees per OR per month) showed significantcorrelations to hospital size and characteristics (the numbersof surgeons, anaesthesiologists and nurses per OR, the totalnumber of beds and the average LOS of each hospital) usingmulti-institutional data. Furthermore, the R2 values were 0.63,0.56, 0.64 and 0.338 for the number of operations per OR permonth, procedural fees per OR per month, total utilization timesper OR per month and total fees per OR per month, respectively,providing satisfactory explanatory power for all four dependentvariables. We then calculated the OE ratio and OE differenceusing these models, and compared the results between multiplehospitals. These results were a more meaningful comparison ofefficiency, as they allow for the adjustment of hospital size andmanpower.

Figures 1 and 2 showed the application of the new indicatorsof OE ratio and OE difference for the number of operations perOR per month, procedural fees per OR per month, total utilizationtime per OR per month and total fees per OR per month. Using theOE ratio and OE difference, each individual hospital can assessthe efficiency of their OR when compared to the results of otherhospitals. A hospital with an OE ratio above one or an OE differ-ence above zero indicates a higher efficiency than expected aftertaking into account variations in hospital characteristics, while ahospital with an OE ratio below one or an OE difference belowzero indicates a lower efficiency than expected. As these indica-tors take into account the variations in hospital size and charac-teristics, inter-hospital comparisons become more meaningful.Such benchmarking among multiple facilities and working basedon the best practice for achieving objectives are thought to beimportant in organizational management, with benefits includingthe clarification of strengths and weaknesses, knowledge ofmultiple positions and reference to the best practice [20]. The

Table 1 Descriptive statistics of indicators for OR management efficiency and hospital size and characteristics

Mean Median SD Min Max n

Number of operations per OR per month 41.6 41.1 15.7 10.6 84.6 224Procedural fee per OR per month ($) 67 492 62 366 33 712 13 949 195 389 224Total utilization time per OR per month(hours) 60.2 57.5 29.6 3.8 160.8 209Total fees per OR per month ($) 111 740 96 986 72 034 15 584 461 148 224Total number of beds 343.7 314 188.9 43 1 118 224Average length of stay 15.1 15 2.6 9.2 34.1 219Number of ORs 5.9 5 3.1 1 20 224Number of surgeons per OR 4.7 4.5 1.6 1 11 214Number of anaesthesiologist per OR 0.6 0.6 0.3 0 1.5 178Number of nurses per OR 3 2.9 1 0.6 8 184Number of emergency patients per year 2 088 1 719 1 767 2 11 211 219

OR, operating room.

M. Tanaka et al. Efficiency indicators of operating room

© 2012 Blackwell Publishing Ltd 3

methodology proposed here can be applied to the analysis ofacute hospitals in general, and the new indicators presented in thisstudy may contribute to benchmarking of OR efficiency in hospi-tals. Benchmarking using multiple facilities is an important toolin organizational management, with benefits including the clari-fication of strengths and weaknesses, knowledge of each hospi-tal’s performance in relation to the performance of numerousother institutions and greater awareness of best practices [20].Furthermore, these indicators can be applied to evaluate theefficiency of other hospital departments, adjusting for structuralfactors such as hospital size and manpower, which are beyond thecontrol of owners’ hospital management.

Further analysis revealed that hospitals with a high OE ratio orOE difference tended to have more single-discipline hospitals suchas cardiology, cardiovascular surgery or neurosurgery departments(data not shown).

At present, there are very few multi-institutional study thathas analyzed OR management efficiency in Japan, and that studyhad utilized only unadjusted total utilization time and number ofoperations per OR using data from an individual hospital [7,21].Therefore, this study is the first to conduct a multi-institutionalcomparison of OR data in Japan. Previous studies analyzingOR management efficiency have generally been based on ques-tionnaire surveys [7,21], and these findings may therefore not bebased on objective data. The hospitals involved in this analysiswere also provided feedback regarding the results of this study,thereby evaluating their OR management efficiency in the contextof the performance of other hospitals.

Our study has several advantages when compared to thosein the existing literature. First, we used objective quantitativedata, which reduces the inherent biases based on survey analyses.Second, we conducted a multi-institutional analysis on a largerscale than previous studies [7–10,22]. Third, we conductedinter-hospital comparisons after taking into account variationsin hospital size and characteristic suitable [7–11]. Fourth, weshowed that the new indicators of OE ratio and OE difference forthe number of operations per OR per month, procedural fees perOR per month, total utilization time per OR per month and totalfees per OR per month have the potential to evaluate OR man-agement efficiency. The results of the number of operations andthe total utilization time per OR are easy to interpret for bothmanagement staff and OR staff, whereas the indicators of pro-cedural fees and total fees per OR may be of more interest tohospital management staff.

Our study has several limitations. First, surgery time was notonly approximated through the time of general anaesthesia, butalso inclusive of the time of epidural anaesthesia, intrathecalanaesthesia, intravenous anaesthesia and local anaesthesia in ourstudy. However, these anaesthesia times are utilized as standardoperative times by the Gaihoren, and are assumed to have satis-factory validity [16]. Second, there is a possibility that therewere other surgeries performed outside of the OR (e.g. thoseconducted in endoscope rooms) that are not included in thisanalysis. Since they do not affect the efficiency of OR manage-ment, their absence from this analysis was not thought to be amajor limitation. Furthermore, the surgeries analyzed in thisstudy were identified from operations records over a 3-year timespan and the judgment of specialists, and therefore should havea high degree of validity.T

ab

le2

Cor

rela

tion

betw

een

the

indi

cato

rsof

OR

man

agem

ent

and

hosp

italr

esou

rces

Num

ber

ofop

erat

ions

per

OR

per

mon

th

Pro

cedu

ral

fee

per

OR

per

mon

th($

)

Tota

lutil

izat

ion

time

per

OR

per

mon

th(h

our)

Tota

lfee

spe

rO

Rpe

rm

onth

($)

Tota

lnu

mbe

rof

beds

Ave

rage

leng

thof

stay

Num

ber

ofO

Rs

Num

ber

ofsu

rgeo

nspe

rO

R

Num

ber

ofan

aest

hesi

olog

ist

per

OR

Num

ber

ofnu

rses

per

OR

Pro

cedu

ralf

eepe

rO

Rpe

rm

onth

($)

0.86

3**

Tota

lutil

izat

ion

time

per

OR

per

mon

th(h

our)

0.81

2**

0.83

4**

Tota

lfee

spe

rO

Rpe

rm

onth

($)

0.70

5**

0.88

5**

0.67

0**

Tota

lnum

ber

ofbe

ds0.

512*

*0.

863*

*0.

531*

*0.

386*

*A

vera

gele

ngth

ofst

ay-0

.417

**-0

.329

**-0

.355

**-0

.279

**-0

.218

**N

umbe

rof

OR

s0.

332*

*0.

509*

*0.

436*

*0.

277*

*0.

884*

*-0

.280

**N

umbe

rof

surg

eons

per

OR

0.62

8**

0.52

3**

0.52

1**

0.34

6**

0.36

1**

-0.2

46**

0.14

8*N

umbe

rof

anae

sthe

siol

ogis

tpe

rO

R0.

466*

*0.

459*

*0.

504*

*0.

420*

*0.

249*

*-0

.177

*0.

132

0.36

9**

Num

ber

ofnu

rses

per

OR

0.44

7**

0.45

1**

0.49

7**

0.40

5**

0.05

7-0

.137

-0.1

100.

463*

*0.

428*

*N

umbe

rof

emer

genc

ypa

tient

s0.

499*

*0.

439*

*0.

494*

*0.

319*

*0.

659*

*-0

.338

**0.

620*

*0.

401*

*0.

243*

*0.

107

*P<

0.05

;**

P<

0.01

.O

R,

oper

atin

gro

om.



ConclusionsNew indicators were produced using multiple regression analysisresults, adjusting for differences in hospital size and manpower,which are beyond the control of process owners’ management. Weproposed that the use of OE ratios and OE differences allowshospitals to identify weaknesses in efficiency with more validity

when compared to unadjusted indicators. Approximately 60% ofthe variations in the majority of unadjusted indicators were shownto be explained by factors such as hospital size and manpower,which the new indicators take into account. Therefore, the varia-tions shown in the new indicators would be better representatives ofdifferential efficiency among the hospitals, and may support theimprovement and sustainment of a high-quality health care system.

Table 3 Results of multiple linear regression analyses

Dependent variable Independent variable b P VIF R2

Number of operations per OR per month Number of surgeons per OR 0.346 0.000 1.515 0.634Total number of beds 0.361 0.000 1.163Number of anaesthesiologists per OR 0.271 0.001 1.194Number of nurses per OR 0.171 0.044 1.351

Procedural fees per OR per month ($) Number of surgeons per OR 0.195 0.048 1.515 0.561Total number of beds 0.387 0.000 1.163Number of anaesthesiologists per OR 0.298 0.001 1.194Number of nurses per OR 0.228 0.015 1.351

Total utilization time per OR per month (hours) Number of anaesthesiologists per OR 0.367 0.000 1.137 0.644Number of nurses per OR 0.427 0.000 1.140Number of ORs 0.481 0.000 1.018

Total fees per OR per month ($) Number of nurses per OR 0.297 0.005 1.125 0.338Total number of beds 0.312 0.002 1.030Number of anaesthesiologists per OR 0.251 0.017 1.152

OR, operating room; VIF, variance inflation factor.

Figure 1 OE ratio (observed value/expected value) for the number of operations per operating room per month for comparing multiple hospitals. (a)OE ratio for the number of operations per operating room per month at the hospital level. (b) OE ratio for the procedural fees per operating room permonth at the hospital level. (c) OE ratio for the total utilization time per operating room per month at the hospital level. (d) OE ratio for the total feesper operating room per month at the hospital level.



Ethical standardsThis study was approved by the Ethics Committee of KyotoUniversity Graduate School of Medicine.

FundingsThis work was supported in part by a Health Sciences ResearchGrant (Grant Number: 22249015) from the Ministry of Health,Labour and Welfare of Japan; and a Grant-in-Aid for ScientificResearch (Grant Number: H22-Iryo-Ippan-017) from the JapanSociety for the Promotion of Science. The sponsors had no role inthe study design in the collection, analysis or interpretation of data,in the writing of the manuscript or in the decision to submit themanuscript for publication.

References1. Wang, B. B., Wan, T. T., Falk, J. A. & Goodwin, D. (2001) Manage-

ment strategies and financial performance in rural and urban hospitals.Journal of Medical Systems, 25 (4), 241–255.

2. Imanaka, Y. (1998) From customer satisfaction to total quality man-agement. In Measurement of Healthcare, What is the Quality Controlof the Healthcare Service (ed. S. Iwasaki), pp. 82–90. Tokyo: ResearchInstitute of Health Science, Inc. (in Japanese).

3. Japanese Association for Operative Medicine (2009) Practiceguideline of operative medicine. Journal of Japanese Association forOperative Medicine, 29, 1–9. (in Japanese).

4. Kanich, D. G. & Byrd, J. R. (1996) How to increase efficiency in theoperating room. Surgical Clinics of North America, 76 (1), 161–173.

5. Weimer, D. M. (1993) A method of evaluating efficiency in theoperating room. American Association of Nurse Anesthetists, 61 (3),287–292.

6. Viapiano, J. & Ward, D. S. (2000) Operating room utilization: the needfor data. International Anesthesiology Clinics, 38 (4), 127–140.

7. Hotta, T., Sato, N., Arakawa, S., Takahashi, H., Usuki, H. & Toyabe,S. (2009) Proper number of nurses in operating room which was seenfrom current state in national university hospitals. Journal of JapaneseAssociation for Operative Medicine, 30, 9–19 (in Japanese).

8. Kamerlink, J. R., Quirno, M., Auerbach, J. D., Milby, A. H., Windsor,L., Dean, L., Dryer, J. W., Errico, T. J. & Lonner, B. S. (2010) Hospitalcost analysis of adolescent idiopathic scoliosis correction surgery in125 consecutive cases. The Journal of Bone and Joint Surgery Essen-tial Surgical Techniques, 92 (5), 1097–1104.

9. Stepaniak, P. S., Heij, C., Mannaerts, G. H., de Quelerij, M. & deVries, G. (2009) Modeling procedure and surgical times for currentprocedural terminology-anesthesia-surgeon combinations and evalua-tion in terms of case-duration prediction and operating room effi-ciency: a multicenter study. Anesthesia & Analgesia, 109, 1232–1245.

10. Friedman, D. M., Sokal, S. M., Chang, Y. & Berger, D. L. (2006)Increasing OR efficiency through parallel processing. Annals ofSurgery, 243 (1), 10–14.

11. Marjamaa, R., Vakkuri, A. & Kirvela, O. (2008) OR management:why, how and by whom? Acta Anaesthesiologica Scandinavica, 52(5), 596–600.

12. Wullink, G., Van, H. M., Hans, E. W., van Oostrum, J. M., van derLans, M. & Kazemier, G. (2007) Closing emergency operating roomsimproves efficiency. Journal of Medical Systems, 31 (6), 543–546.

Figure 2 OE difference (observed value–expected value) for the number of operations per operating room per month for comparing multiplehospitals. (a) OE difference for the number of operations per operating room per month at the hospital level. (b) OE difference for the procedural feesper operating room per month ($) at the hospital level. (c) OE difference for the total utilization time per operating room per month (hours) at thehospital level. (d) OE difference for the total fees per operating room per month ($) at the hospital level.



13. Stahl, J. E., Egan, M. T., Goldman, J. M., Tenney, D., Wiklund, R. A.,Sandberg, W. S., Gazelle, S. & Rattner, D. W. (2005) Introducing newtechnology into the operating room: measuring the impact on jobperformance and satisfaction. Surgery, 137 (5), 518–526.

14. Berry, M., Berry-Stölzle, T. & Schleppers, A. (2008) Operating roommanagement and operating room productivity: the case of Germany.Health Care Management Science, 11 (3), 228–239.

15. Weinbroum, A. A., Ekstein, P. & Ezri, T. (2003) Efficiency of theoperating room suite. The American Journal of Surgery, 185 (3),244–250.

16. Gaihoren (2007) The Seventh Edition of the Tentative Plan Concern-ing Operating Reward. pp. 32–209. Tokyo: Gaihoren (the Confedera-tion of Social Insurance Committees of Surgical Societies) (inJapanese).

17. Lokker, C., McKibbon, K. A., McKinlay, R. J., Wilczynski, N. L. &Haynes, R. B. (2008) Prediction of citation counts for clinical articlesat two years using data available within three weeks of publication:retrospective cohort study. British Medical Journal, 336 (7645), 655–657.

18. A United Healthcare Company (2007) Pacificare inpatient hospital

readmission index methodology. Available at: http://www.uhc.com/live/uhc_com/Assets/Documents/UnitedHealth%20Group%2030-day%20All-Cause%20Readmission.pdf#search=’Risk Adjustment Meth-odology Update 2007 (last accessed 6 December 2011).

19. CathPCI Registry (2009) Understanding risk adjusted mortality in theCathPCI Registry. National Cardivascular Data Registry. Available at:http://www.ncdr.com/WebNCDR/NCDRDOCUMENTS/CathPCI_RiskAdjustmentTechNotes_100109.pdf#search=’CathPCI RegistryUnderstanding Risk Adjusted Mortality(RAM)in the CATHPCI Regis-try (last accessed 6 December 2011).

20. Robert, C. C. (1994) Benchmarking. pp. 23–68. Munich: HanserPublishers.

21. Shiraishi, Y., Nakagawa, T., Morita, K. & Furuhashi, R. (2009) Oper-ating room management efficiency based on survey of the amount ofduties for nurse. Journal of Japanese Association for Operative Medi-cine, 30 (2), 155–158. (in Japanese).

22. Nocco, U. & del Torchio, S. (2011) The integrated OR efficiency andeffectiveness evaluation after two years use, a pilot study. Interna-tional Journal of Computer Assisted Radiology and Surgery, 6 (2),175–186.

