Measurement of Relative Efficiency of Health Service Organizations with Data Envelopment Analysis - a Simulation

Data Envelopment Analysis (DEA) , a new methodology based on linear programming concepts, provides an approach to evaluate the relative technical efficiency of nonprofit organizations which have multiple inputs and outputs. This approach potentially will identify inefficient units and the magnitude of the inefficiency to provide a basis to select inefficient units for management review or efficiency audits and to help locate areas where operations might be improved. This is believed to be an improvement over existing approaches to evaluate efficiency of such organizations and is directed toward health service organizations in this study because of the potential value of such an approach in this sector. This paper investigates an application of DEA to an artificial data set reflecting the operations of a hospital department. The underlying technology is specified from which a set of efficient and inefficient hospital units are developed. Without knowledge of this technology, DEA accurately identifies the inefficient units when the inputs and outputs are properly specified. In contrast, the widely used single -output measures applied to this data set are found to be less reliable in this multiple output environment. The strengths and limitations of DEA are further elaborated to anticipate issues that may arise in subsequent field applications of DEA to