Title:	A FRAMEWORK FOR MINING BEHAVIORS OF PATIENTS BASED ON LENGTH OF STAY IN THE EMERGENCY DEPARTMENT
Author(s):	I-Chin Wu, Tzu-Li Chen, Chung-Lun Yang, Ren-Shian Yang, Yen-Yi Feng
ISBN:	978-972-8939-98-4
Editors:	Piet Kommers, Tomayess Issa, Nurfadhlina Mohd Sharef and Pedro Isaías
Year:	2013
Edition:	Single
Keywords:	Data Mining, ED Overcrowding, Length of Stay, Multi-tiered Framework, Patient Behavior
Type:	Poster/Demonstration
First Page:	128
Last Page:	130
Language:	English
Cover:
Full Contents:	click to dowload
Paper Abstract:	The demand for emergency medical services has increased in recent years; therefore, the emergency department (ED) has become the most important and the busiest unit within a hospital, providing emergency care and treatment to patients in need of immediate medical attention. However, EDs in Taiwan have not been able to accommodate the rapidly increasing patient demand in the past 14 years. Such increases cause an imbalance or a mismatch between supply and demand, ultimately creating long-term overcrowding in hospital EDs, also known as ED crowding. This phenomenon is caused primarily by the sharp increase in the number of patients (demand side) and the insufficiency or lack of a corresponding increase in medical resources (supply side), including doctors, ED beds, nurses, and equipment. This condition compels patients to wait long hours to complete medical procedures, thereby contributing to long-term overcrowding in EDs. The objective of this study was to identify possible solutions for emergency department (ED) overcrowding in the Mackay Memorial Hospital, which has the second-largest ED in Taiwan and is a representative institute. In this research, we proposed a novel multi-tiered ED crowding data mining and analytical framework for mining patients’ behaviors in the ED based on different length of stay (LOS) to investigate the causes of ED crowding from the perspectives of patients’ behaviors. Basically, two types of patient behavior (PB), i.e., regular and exceptional behaviors of patients, were identified in this research. Consequently, we aim at decreasing the overcrowding conditions and improving the quality of health care in EDs.
	Go Back