Track: Poster Competition

Abstract

In the United States, federal law requires that hospital emergency departments (EDs) are to provide care to all patients, regardless of patient’s ability to pay. This has caused the emergency rooms to be stretched beyond their capacity which adversely affects the quality of care rendered to patients. To improve the quality of health care service provided in the EDs, hospital management has adopted benchmarking (BM) which is a quality improvement tool for performance measurement and efficiency analysis of the EDs. This study utilizes a structured BM model which consists of planning, analysis, integration, and action phases. Data envelopment analysis (DEA), which is a non-parametric technique for estimating the efficiency of a given set of decision-making units (DMUs) and Back-propagation neural network (BPNN), a supervised learning algorithm for training neural networks is incorporated into the analysis phase of benchmarking as a performance prediction tool. Results from the BP-DEA model shows that fifty percent of the DMUs are efficient while the remaining fifty percent is relatively inefficient. Areas of potential improvement in the less-efficient departments are investigated. Implementing recommendations from the analysis can lead to increased quality of healthcare services provided to patients and increased efficiency in hospital operations.