A dynamic benchmarking system for assessing the recovery of inpatients: Evidence from the neurorehabilitation process

The shortage of medical resources (mainly beds) is a critical and increasingly prevalent problem affecting hospitals. Of the factors that contribute to these shortages, the ambiguity and insufficiency of the criteria used to identify whether an inpatient should be discharged are among the most detrimental. To address this issue, this study applies data envelopment analysis (DEA) on existing inpatient data from the Neurorehabilitation Center at Toronto’s Bridgepoint Hospital to create a dynamic benchmarking system to evaluate the health stage of an inpatient ready to be discharged. Unlike the more traditional parametric techniques, DEA provides non-subjective benchmarking that does not require any prior specification of the production function making it a more desirable choice for this application. The dynamic model categorizes the inpatient’s discharge status as rejected, under observation, or approved. This new approach not only allows managers to gain insight into the potential causes of medical resource shortages, but also allows clinicians to treat inpatients more effectively based on their discharge categories. For validation, the results of the dynamic model were compared with actual inpatient discharge assessments provided by the Bridgepoint Hospital.