Road map for progress and attractiveness of Iranian hospitals by integrating self-organizing map and context-dependent DEA.

Hospitals play an important role in healthcare systems and usually stay on the end node of the healthcare chain. Thus, determining their road map to get close to the desired efficiency frontier and developing short-term and long-term plans could help to manage costs and resources, efficiently. As the efficiency frontier depends on the size of the hospital and the complexity of its structure, the homogeneity in benchmarking must be considered. For tackling this problem, the self-organizing map (SOM) is used to create homogeneous groups. On the other hand, data envelopment analysis (DEA) is a well-known methodology for evaluating decision-making units. Each unit obtains the efficiency score based on the ratio of weighted outputs to weighted inputs, where each unit can take the desirable weights for inputs and outputs to provide the maximum value. One of the problems of DEA is the selection of the reference set and distinguishing between the efficient hospitals. To overcome these problems, the context-depended DEA has been applied and the progress and attractiveness of hospitals are obtained. To evaluate the capability of the proposed approach, data of 288 Iranian hospitals are utilized. By applying SOM the hospitals are clustered into appropriate homogeneous groups and by applying context-dependent DEA, the road map for progress and attractiveness of each hospital is determined. In other words, using the proposed approach the hospitals are able to determine the short and long-term goals according to their strategic plans.

Do Geographical Locations Affect in Hospitals Performance? A Multi-group Data Envelopment Analysis.

Hospitals are the main and the last component of the health care systems. Therefore, evaluating the performance of hospitals is vital. Several studies are done in hospitals evaluation, but almost none of them consider the geographical features. This paper proposes a new approach for evaluating hospitals. In the proposed approach, hospitals are classified into various groups and each group is equivalent to a province. It causes hospitals in each category (province) must be evaluated in homogenous environment. For this purpose, the conventional data envelopment analysis (DEA) model has been developed to this structure. The main feature of the proposed model is that it takes into consideration the geographical location. In other words, we propose a multi-group DEA model. The data on 288 Iranian hospitals grouped under 31 provinces are used to demonstrate the model. The results show that the efficiency scores are greatly changed when hospitals are evaluated in own groups.