Skip to main content
PMC home page
Iranian Journal of Public Health logoLink to Iranian Journal of Public Health
. 2014 Nov;43(11):1576–1581.

Hospitals Productivity Measurement Using Data Envelopment Analysis Technique

Amin TORABIPOUR 1,*, Maryam NAJARZADEH 2, Mohammad ARAB 1, Freshteh FARZIANPOUR 1, Roya GHASEMZADEH 2
PMCID: PMC4449509  PMID: 26060727

Abstract

Background

This study aimed to measure the hospital productivity using data envelopment analysis (DEA) technique and Malmquist indices.

Methods

This is a cross sectional study in which the panel data were used in a 4 year period from 2007 to 2010. The research was implemented in 12 teaching and non-teaching hospitals of Ahvaz County. Data envelopment analysis technique and the Malmquist indices with an input-orientation approach, was used to analyze the data and estimation of productivity. Data were analyzed using the SPSS.18 and DEAP.2 software.

Results

Six hospitals (50%) had a value lower than 1, which represents an increase in total productivity and other hospitals were non-productive. the average of total productivity factor (TPF) was 1.024 for all hospitals, which represents a decrease in efficiency by 2.4% from 2007 to 2010. The average technical, technologic, scale and managerial efficiency change was 0.989, 1.008, 1.028, and 0.996 respectively. There was not a significant difference in mean productivity changes among teaching and non-teaching hospitals (P>0.05) (except in 2009 years).

Conclusion

Productivity rate of hospitals had an increasing trend generally. However, the total average of productivity was decreased in hospitals. Besides, between the several components of total productivity, variation of technological efficiency had the highest impact on reduce of total average of productivity.

Keywords: Hospital productivity, DEA, Malmquist indices, Technical efficiency

Introduction

Hospital is one of the main organizations in health service system. It has a special importance in health economics, and imposes higher costs on the health system compared to the other health system components (1, 2). Hospitals are the main consumer of resources in any health sector thus; improvement of their efficiency is the main way to decrease the hospital costs (3).

Thus, providing criteria for evaluation of hospital performance and productivity is important (2, 4). These features are doubled, especially in developing countries, considering their economic structure and extreme vulnerability to deal with fluctuations in currency and commodity markets. In a way, that full response to the needs of consumers in this sector, even in the most advanced countries, seems to be out of reach (5, 6). Hence, the Iranian hospitals are no exception from this issue. In this situation, the hospitals managers’ strategies for reducing the costs and increasing the productivity is necessary (7). Productivity is one of the important concepts in assessing the performance of organization over time. Productivity expresses a kind of relationship between the amount of the produced goods or services and the amount of resources, which is consumed in producing goods and services. Therefore, productivity is a measure that shows the relationship between outputs and inputs (8). Productivity indicators show improvement or reduction in performance (9). The information obtained from productivity sub-indicators are used to identify the problems of unites, define the strategies, and in case of need, make change in the staff behaviors (10).

Several techniques to assessment hospital efficiency and productivity are usually considered either parametric or non-Parametric. (11). Non-parametric methods such as data envelopment analysis are the most popular (12).

This study was performed to determine the productivity rate of hospitals by using Data Envelopment Analysis technique in Ahvaz, Iran.

Materials and Methods

This is a cross sectional study in which the panel data were used in a 4 year period from 2007 to 2010. The research was implemented in 17 teaching and non-teaching hospitals of Ahvaz County, southern Iran. As five hospitals did not agree to participate in this study; and finally 12 hospitals were assessed. Variables were selected based on the experts’ opinions and previous studies (1, 13, 14). The set of input and output variables that we have used in this study are similar to those applied in other studies (15–17). The input measures included are number of nurses, number of occupied beds and number of physicians. The output measures included are number of outpatients and inpatients, average of hospital stay, and number of major operations

Data collection

Data were collected by studying the medical records and documents of the hospitals retrospectively. Data were collected using standard checklists prepared by the Ministry of Health.

Data Analysis

Data envelopment analysis technique and the Malmquist indices with an input-orientation approach, was used to analyze the data and estimation of productivity. We selected input-orientation approach and variable returns to scale (VRS) model for data analysis. In input-orientation approach, decision-making units (DMUs) can change their inputs (18, 19). When the productivity and efficiency measurement are based on the input-orientation approach (minimizing the production factors), a value of more than one show productivity decline and a value of less than one show Productivity growth (1, 7). Data was analyzed by SPSS.18 and DEAP.2 software.

Results

We studied twelve hospitals in four-year periods. As noted in Table 1, J hospital has the highest value and H hospital has the lowest value of total productivity changes. Therefore, the range of changes of productivity for hospitals was from 0.785 to 1.127, and J and H hospitals had the best and worst performance respectively. K hospital had the maximum value of changes in technical, technological, managerial and scale efficiency. H hospital had the lowest technical and scale efficiency changes, G hospital had the lowest technological efficiency changes and I hospital had the minimum management performance changes. Six hospitals (50%) had a value lower than 1, which represents an increase in total productivity between 2007 and 2010. Five hospitals (41.66 %) had a value higher than 1, which represents a decrease in total productivity and one hospital (8.33 %) had a value equal to 1 which reflects stagnation in total productivity changes.

Table 1.

Efficiency and total productivity change by hospital type (Teaching and non-teaching hospital)

Hospital type N=12 Technical efficiency change Technological efficiency change Managerial efficiency change Scale efficiency change Productivity change Productivity status
A(teaching) 0.873 0.985 0.889 0.982 0.860 Productive
B(teaching) 0.960 1.008 0.907 1.058 0.968 Productive
C(teaching) 0.955 0.973 1 0.955 0.929 Productive
D(teaching) 0.934 1.07 0.906 1.032 1 Neutral
E(teaching) 0.997 1.003 0.965 1.034 1.001 Non-productive
F(nonteaching) 1.077 1.007 1.042 1.033 1.085 Non-productive
G(nonteaching) 1 0.861 1 1 0.861 Productive
H(nonteaching) 0.882 0.889 1 0.882 0.785 Productive
I(nonteaching) 0.909 0.986 0.858 1.059 0.896 Productive
J(nonteaching) 1 1.127 1 1 1.127 Non-productive
K(nonteaching) 1.486 1.412 1.384 1.074 1.097 Non-productive
L(nonteaching) 1 1.105 1 1 1.105 Non-productive
Mean 0.996 1.028 0.989 1.008 1.024 Non-productive
Open in a new tab

According to the Table 2, the average of total productivity was 1.024, which reflects a decrease in productivity changes by 2.4% in study period. The average total productivity change of all factors in the year 2008 was more than 1, which shows that the level of productivity of hospitals has been decreased in this year. Changes in technological performance had a negative impact on the rate of productivity and had the main role in reducing productivity.

Table 2.

Efficiency and productivity change in three years-period (200-8-2010)

Productivity status Total productivity change Scale efficiency change Managerial efficiency change Technological efficiency change Technical efficiency change Year*
Non-productive 1.226 1.007 0.963 1.265 0.969 2008
productive 0.972 0.980 1.071 0.926 1.049 2009
productive 0.901 1.038 0.937 0.926 0.972 2010
Non-productive 1.024 1.008 0.989 1.028 0.996 Total Mean
Open in a new tab

*In this study, the 2007 year is basic year for assessment other years (so has not been showed in table)

As noted in Table 3, mean of the productivity change had a significant difference among teaching and non-teaching hospitals (P=0.001) in 2009 and had not significant differences in other years. In general, there was no significant difference in average of productivity changes between teaching and non-teaching hospitals.

Table 3.

Comparison of productivity change between teaching hospitals and non-teaching

Year Productivity Total productivity change (3 year) Productivity status
Hospital type 2008 2009 2010
Teaching 0.967 1 0.907 0.951 productive
Non-teaching 2.370 0.428 0.966 1.124 Non-productive
Mean 1.668 0.714 0.936 1.037 -
P value 0.381 0.001 0.695 0.446 -
Open in a new tab

Discussion

We measured total hospital productivity using malemquist index. Malemquist index is a useful technique to assess hospital productivity (20). The results showed that the productivity of hospitals over the years 2007-2010 has increased and had an ascending trend. However, the average of total productivity was 1.024, which shows 2.4% reduction; and changes in technological performance had the highest impact on decrease in value of total productivity. Different values for total factors productivity have been reported in different researches in Iran and other countries.

A study in Iran showed that just fewer than 60% of all hospitals at were technically efficient and therefore they were non-product because efficiency of units has direct association with productivity trend (2). Chang et al. (8) in a similar study reported increase in the productivity of 31 regional hospitals in Taiwan. The average of productivity changes in the Taiwan’s hospitals was (0.874) was lower than average of productivity changes in our study.

One way to increase technical efficiency is decrease of length of stay and increase of hospital size. Masayuki (21) showed that the larger hospitals have the higher productivity. As doubling hospital size lead to increase of productivity more than 10%. Of course, length of stay effects should be controlled simultaneously. It should be noted Productivity change is affected scale efficiency changes, technological performance and managerial performance (13, 15). Najafi et al. (22) reported that the average productivity of the hospitals has declined by 3.3% in Ardebil and the technological efficiency changes had the most negative effects on total productivity. Evaluation of the causes of these effects can be helpful in improvement of the total productivity. The lack of technological innovation in hospitals led to a decrease in TFP change. Technological improvements play an important role in increasing the productivity of larger hospitals (23).

Findings of our study showed that despite improvement in productivity trend of hospitals, the average of productivity of hospitals not improved; this result can be used as a warning in order to make better use of existing resources. Some hospitals (A, B, C, G, H and I) used resources the more appropriate which can be considered as a model for other hospitals (D, E, F, I, J, K and L). In hospitals that the main cause of low productivity is changes in technology efficiency, it seems that lack of enough familiarity of managers with the advanced hospital technologies, lack of equipment and inappropriate use of technology in diagnosis, care and treatment, are the main causes (22). Limitations of current study were first, databases are not well developed in hospitals. Second, medical record system is manual and third, DEA software has technical limitations.

Conclusions

In the F and K hospitals the main cause of the low productivity is changes in technical efficiency therefore mangers must use inputs (doctors, nurses and beds) efficiently and increase the outputs. In hospitals E that the main cause of low productivity is negative changes of scale efficiency therefore managers must improve the scale efficiency. Finally, according to results of this study, decrease of technological efficiency had the highest negative impact on average of total productivity of hospitals. Therefore, hospital Managers and policy makers can use economic of scale methods including: optimizing size of hospitals, increase of production volume and decrease of the length of stay (by means of discharge process optimizing) for productivity improvement of hospitals.

Ethical considerations

Ethical issues (Including plagiarism, informed consent, misconduct, data fabrication and/or falsification, double publication and/or submission, redundancy, etc.) have been completely observed by the authors.

Acknowledgments

The authors would like to appreciate the Research and Technology Deputy of Ahvaz Jundishapur University of Medical Sciences and School of Rehabilitation Sciences for their financial and administrative support to undertake this project. The authors declare that there is no conflict of interests.

References

  1. Farzianpour F, Hosseini SH, Amali T, Hosseini S, Hosseini SS (2012). The Evaluation of Relative Efficiency of Teaching Hospitals. Am J Applied Sci, 9 (3): 392–398. doi: 10.3844/ajassp.2012.392.398 [DOI] [Google Scholar]
  2. Shahhoseini R, Tofighi S, Jaafaripooyan E and Safiaryan R (2011). Efficiency measurement in developing countries: Application of data envelopment analysis for Iranian hospitals. Health Services Manage Res, 24: 75–80. doi: 10.1258/hsmr.2010.010017 [DOI] [PubMed] [Google Scholar]
  3. Kiadaliri AA, Jafari M, Gerdtham U G (2013). Frontier-based techniques in measuring hospital efficiency in Iran: a systematic review and meta-regression analysis. BMC Health Serv Res, 13: 312–314. http://www.biomedcentral.com/1472-6963/13/312 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Afzali HHA, Moss JR, Mahmood MA (2009). A conceptual framework for selecting the most appropriate variables for measuring hospital efficiency with a focus on Iranian public hospitals. Health Serv Manage Res, 22: 81–91. doi: 10.1258/hsmr.2008.008020 [DOI] [PubMed] [Google Scholar]
  5. Barnum DT, Walton SM, Shields KL, Schumock GT (2009). Measuring hospital efficiency with data envelopment analysis: No substitutable Vs. Substitutable Inputs and Outputs. J Med Sys, 35: 1393–1401. [DOI] [PubMed] [Google Scholar]
  6. Karimi S, Sajadi HS, Karami M, Torkzad L, Bidram R (2009). Efficiency Estimation in General Hospitals of Isfahan University of Medical Science during 2005-2006 by Data Envelopment Analysis. Journal of Health Administration, 12(36): 39–46. [In Persian] [Google Scholar]
  7. Lobo DC, Ozcan YA, Silva AC, Lins MP, Fiszman R (2010). Financing reform and productivity change in Brazilian teaching hospitals: Malmquist approach. Central Eur J Oper Res, 18: 141–152. doi: 10.1007/s10100-009-0097-z [DOI] [Google Scholar]
  8. Chang SJ, Hsiao HC, Huang LH, Chang H (2011). Taiwan quality indicator project and hospital productivity growth. Omega, 39: 14–22. doi: 10.1016/j.omega.2010.01.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Cook WD, Zhu J (2008). Data Envelopment Analysis: Modeling Operational Processes and Measuring Productivity. USA, 1–262. [Google Scholar]
  10. Emrouznejad A, Parker BR, Tavares G (2008). Evaluation of research in efficiency and productivity: A survey and analysis of the first 30years of scholarly literature in DEA. Socio-Econ Plan Sci, 42: 151–157. doi: 10.1016/j.seps.2007.07.002 [DOI] [Google Scholar]
  11. Jacobs R (2001). Alternative methods to examine hospital efficiency: data envelopment analysis and stochastic frontier analysis. Health Care Manage Sci, 4(2):10315. [DOI] [PubMed] [Google Scholar]
  12. Pestana BC, Gomes MA, Peypoch N, Solonandrasana B, Vieira JC (2008). An analysis of hospital efficiency and productivity growth using the Luenberger indicator. Health Care Manage Sci, 11: 373–381. DOI 10.1007/s10729-007-9043-6. [DOI] [PubMed] [Google Scholar]
  13. Farzianpour F, Aghababa S, Delgoshaei B, Haghgoo M (2011). Performance evaluation a teaching hospital affiliated to Tehran University of medical sciences based on balding excellence model. Am J Econ Bus Admin, 3: 277–281. [Google Scholar]
  14. Tlotlego N, Nonvignon J, Sambo LG, Asbu EZ, Kirigia JM (2010). Assessment of productivity of hospitals in Botswana: A DEA application. Int Arch Med, 3: 27–27. doi: 10.1186/1755-7682-3-27 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Ghaderi H, Goudarzi G, Gohari M (2007). Determining the technical efficiency of hospitals affiliated to IUMS using DEA through 2000-2004. J Heal Manage, 26: 31–38. [In Persian] [Google Scholar]
  16. Goudarzi G (2006). Determination and Survey of Factors effecting in technical efficiency of hospitals affiliated with IUMS by SFA and DEA methods [MS Thesis]. Iran University of Medical Sciences. Iran: [In Persian]. [Google Scholar]
  17. Sabermahani A, Goudarzi G, Barooni M, Khakian M (2009). Evaluating the technical efficiency of public hospitals affiliated with KUMS in 2007. J Kerman University of Med Sci, 17(1): 59–67. [In Persian]. [Google Scholar]
  18. Martic MM, Novakovic MS, Baggia A (2009). Data Envelopment Analysis Basic Models and their Utilization. Organizacija, 42: 37–44. [Google Scholar]
  19. Tànfani E, Testi A (2012). Advanced Decision Making Methods Applied to Health Car. Springer: verlag-Italia, 1–236. [Google Scholar]
  20. Roos p. Measurement of productivity in hospital services using Malmquist index approaches: a discussion of methods and illustration to eye surgery. CSLS Conference on service Sector productivity and the productivity paradox. Chateau Laurier Hotel Ottawa, Canada: April 11-12, 1997. [Google Scholar]
  21. Masayuki, M. (2010). Economies of Scale and Hospital Productivity: An empirical analysis of medical area level panel data. RIETI Discussion Paper Series 10-E-050. The Research Institute of Economy, Trade and Industry. Available from: http://www.rieti.go.jp/en/
  22. Najafi B, Beheshti Dehkordi A, Emami Meibodi A (2011). The productivity of general hospitals of Ardebil Province (1999–2006). JQUMS, 14(4): 64–70. [In Persian] [Google Scholar]
  23. Gannon B (2008). Total factor productivity growth of hospitals in Ireland: a nonparametric approach. Applied Econ Letters, 15 (2): 131–135. [Google Scholar]

Articles from Iranian Journal of Public Health are provided here courtesy of Tehran University of Medical Sciences

RESOURCES