Reasons for measuring efficiency

http://pages.stern.nyu.edu/~wgreene/FrontierModeling/SurveyPapers/Lovell-Fried-Schmidt.pdf pages 10-11

Why the interest in measuring efficiency and productivity? We can think of three reasons. First, only by measuring efficiency and productivity, and by separating their effects from those of the operating environment so as to level the playing field, can we explore hypotheses concerning the sources of efficiency or productivity differentials. Identification and separation of controllable and uncontrollable sources of performance variation is essential to the institution of private practices and public policies designed to improve performance. Zeitsch et al. (1994) provide an empirical application showing how important it is to disentangle variation in the operating environment (in this case customer density) from variation in controllable sources of productivity growth in Australian electricity distribution.

Three problems of measuring efficiency

After determining three problems of measuring efficiency’ they explained the reason of usage of frontiers. (MSY)

http://pages.stern.nyu.edu/~wgreene/FrontierModeling/SurveyPapers/Lovell-Fried-Schmidt.pdf (some passages from page 1-10)

By the efficiency of a producer we have in mind a comparison between observed and optimal values of its output and input. The exercise can involve comparing observed output to maximum potential output obtainable from the input, or comparing observed input to minimum potential input required to produce the output, or some combination of the two.

Even at this early stage three problems arise, and much of this Section is devoted to exploring ways each has been addressed. First, which outputs and inputs are to be included in the comparison? Second, how are multiple outputs and multiple inputs to be weighted in the comparison? And third, how is the technical or economic potential of the producer to be determined?

Understanding differences in efficiency levels

It is a application presentation from https://www.healtheconomics.org/congress/2013/  Our presentation about medical tourism will be there too… Analysing Unexplained Variations in Performance to Increase the Efficiency of Health Care: Case Studies from the ECHO Project Ceu Mateus (National School of Public Health, Nova University of Lisbon)

Background
Measuring efficiency in health care presents analytical challenges including defining measurable outputs and the availability of information about inputs. International comparisons are particularly difficult given the lack of reliable and comparable data.

Aims
i. To identify differences in the production functions of hospitals in each country;
ii. To analyse which resources and hospital characteristics affect efficiency levels;
iii. To assess levels of technical efficiency in each country;
iv. To characterise and understand differences in efficiency levels between countries.

Data
Patient level data is available for all public hospitals in the participating countries, providing detailed information on patient characteristics and medical conditions, accounting for differences in case-mix. Data on hospital resources includes the number of employees, number of physicians, number of nurses, number of beds, teaching status of the hospital and other structural characteristics.

Methods
For efficiency measurement the number of discharges weighted for the complexity of patients treated in the hospital is used as a proxy for output. Variables relating to hospital resources are used to estimate the quantity of inputs. Stochastic frontier analysis with bootstrapping is used to estimate each country’s production function and to analyse efficiency levels. Technical efficiency of each hospital in each country is then estimated enabling the characterisation of the reality during the period under analysis. Differences in efficiency levels between countries can therefore be assessed, by the estimation of the optimal level of resources in order to identify any excess resource use.

Results
Using data for Portugal, preliminary results are available which indicate that the number of beds is the most significant estimator for the hospital production function. The number of doctors and the number of employees are also significant in some years. The average efficiency level has been stable over the period analysed but the tails of the distribution are becoming more heavy over time, meaning less hospitals are near the median. The percentage of hospitals achieving levels of efficiency above 90% increased over time, particularly in 2008 and 2009, but the percentage of those achieving less than 20% of efficiency also rose. The presentation will include results from other countries in the ECHO consortium, permitting hospital level comparisons across countries.

Conclusions
The estimation and comparison of hospital technical efficiency over time provides valuable information on how to improve hospital management and performance. This can be done by identifying first how differently hospitals are producing and second by understanding how and which resources affect efficiency.

Individual Hospital Evaluation with Stochastic

How do we asses and evaluate hospitals with stochastic frontier when there are ‘significant concerns’ in literature ? Then It can only be used for policy making. (MSY)

Analysis to U.S. Hospitals? What Have We Learned From the Application of Stochastic Frontier Michael D. Rosko and Ryan L. Mutter
 http://mcr.sagepub.com/content/68/1_suppl/75S.full.pdf+html 

The literature has not focused on the use of SFA to assess individual hospital performance, and there are significant concerns about using SFA for that type of analysis (Newhouse, 1994; Street, 2003). While some of these concerns have been addressed (Rosko & Mutter, 2008), others remain, so SFA has primarily been used to examine the relative efficiency of groups of hospitals (Folland & Hofler, 2001).

 Newhouse, J. (1994). Frontier estimation: How useful a tool for health economics? Journal of Health Economics, 13, 317-322.
Street, A. (2003). How Much Confidence Should We Place in Efficiency Estimates? Health Economics, 12, 895-907.

Mutter, R., & Rosko, M. (2008). The impact of ownership on the cost-efficiency of U.S. hospitals.
In J. Blank & V. Valdmanis (Eds.), Evaluating hospital policy and performance: Contributions from hospital policy and productivity research (pp. 113-138). New York, NY: Elsevier.

 Folland, S., & Hofler, R. (2001). How reliable are hospital efficiency estimates? Exploiting the dual to homothetic production. Health Economics, 10, 683-698.