Measuring efficiency of governmental hospitals in Palestine using stochastic frontier analysis

Abstract

BackgroundThe Palestinian government has been under increasing pressure to improve provision of health services while seeking to effectively employ its scare resources. Governmental hospitals remain the leading costly units as they consume about 60 % of governmental health budget. A clearer understanding of the technical efficiency of hospitals is crucial to shape future health policy reforms. In this paper, we used stochastic frontier analysis to measure technical efficiency of governmental hospitals, the first of its kind nationally.MethodsWe estimated maximum likelihood random-effects and time-invariant efficiency model developed by Battese and Coelli, 1988. Number of beds, number of doctors, number of nurses, and number of non-medical staff, were used as the input variables, and sum of number of treated inpatients and outpatients was used as output variable. Our dataset includes balanced panel data of 22 governmental hospitals over a period of 6 years. Cobb–Douglas function, translog function, and multi-output distance function were estimated using STATA 12.ResultsThe average technical efficiency of hospitals was approximately 55 %, and ranged from 28 to 91 %. Doctors and nurses appear to be the most important factors in hospital production, as 1 % increase in number of doctors, results in an increase in the production of the hospital of 0.33 and 0.51 %, respectively. If hospitals increase all inputs by 1 %, their production would increase by 0.74 %. Hospitals production process has a decrease return to scale.ConclusionDespite continued investment in governmental hospitals, they remained relatively inefficient. Using the existing amount of resources, the amount of delivered outputs can be improved 45 % which provides insight into mismanagement of available resources. To address hospital inefficiency, it is important to increase the numbers of doctors and nurses. The number of non-medical staff should be reduced. Offering the option of early retirement, limit hiring, and transfer to primary health care centers are possible options. It is crucial to maintain a rich clinical skill-mix when implementing such measures. Adopting interventions to improve the quality of management in hospitals will improve efficiency. International benchmarking provides more insights on sources of hospital inefficiency.

Analysis of the technical efficiency of Nile perch (Lates niloticus) fishers in the Tanzanian portion of Lake Victoria: A stochastic frontier analysis

“…Thus, if RTS > 1, RTS = 1 or RTS < 1, one has increasing, constant or decreasing returns to scale, respectively (Ehiakpor et al, ; Sesabo & Tol, ). Following the approach of Sesabo and Tol, (), Hamidi () and Ehiakpor et al (), the specification of the input elasticity is presented as follows:…”

“…As the coefficient values of translog production do not have a straightforward interpretable meaning, the marginal effects for each input were estimated through the marginal product equal to the elasticity of scale for each input in order to assess the individual input effect on the output (Hamidi, 2016;Sesabo & Tol, 2007). The elasticity measures the responsiveness of output to a (4) (Ehiakpor et al, 2015;Sesabo & Tol, 2007).…”

Methodological Approach and Theoretical Concept for Measuring Hospital Technical Efficiency

“…A study conducted by (Hamidi, 2016) which applied SFA model as a parametric approach. This study aimed to examine technical efficiency of 22 government hospitals in Palestine over a period of 6 years, 2006, 2007 and 2009 to 2012.…”

Measuring Technical Efficiency of Hospitals Using Data Envelopment Analysis: A Review

“…The DEA has its capability to cope with complicated production environments with multiple input and output (Jacobs, 2001b;Ruggiero, 2007). In contrast, the parametric approach such as SFA only able to manage multiple inputs with one input (Hamidi, 2016;Ramírez-Valdivia et al, 2015). DEA does not require the input prices to estimate technical efficiency, while the parametric approaches are required the input prices (Grosskopf & Valdmanis, 1993).…”