سیستم پشتیبانی تصمیم گیری برای مدیریت تریاژ: استدلال مبتنی بر منطق فازی Decision support system for triage management: A hybrid approach using rule-based reasoning and fuzzy logic

1. Introduction Hospital emergency departments (EDs) attempt to provide a timely service for clients who do not plan ahead. Compared with other health centers, the ED is a unique environment with limited resources and a wide range of patients in need of care [1]; it is considered one of the most important departments among all health care systems. However, as overcrowding in EDs threatens the health of patients, triage is performed as an effective solution to tackle this problem [2]. The triage process is the first critical step in giving care to the clients of EDs by prioritizing patients at different triage levels based on the severity of their clinical conditions. Triage servers prioritize patients for urgent care based on a short initial clinical assessment usually performed by emergency nurses. In some hospitals, in addition to treatment priority, triage determines the visiting location of patients; for example, interior room, trauma room, cardio-pulmonary resuscitation room, or an outpatient room [3,4]. In emergency situations, fast and accurate patient triage for the response process is critical in the coordination of medical services with hospital sources since there is a high mortality rate. In many hospitals, the triage process is often performed using a paper-based mode; however, this method intensifies workload and difficulty, wastes time, and is open to human errors [5,6]. Triage decision-making is an important task that should be conducted for each patient referring to the ED. However, the characteristics of the triage server, such as his or her evaluation and experience, the patient’s clinical history, and the availability of necessary resources all contribute to the complexity of the triage process. The most important task to accomplish in the ED is enabling the available physician to quickly and accurately recognize the patient’s medical needs to avoid costs of unnecessary surgeries and other medical treatments [7]. For more accurate recognition, some standardized classification systems were developed for the triage process. For example, the Emergency Severity Index (ESI) is a standard instrument for classifying patients based on the estimated acuity and resource consumption [2]. Research findings have shown that the ESI triage method is regarded as one of the best ways to prioritize patients in many countries, including Iran, and that this method is considered a valid and accurate system for improving the access to medical care that is used in the ED of Iranian hospitals [2]. Levels 1 and 2 of the ESI have been respectively assigned for emergency and urgent situations requiring the immediate assessment and intervention by service providers to prevent death. Clients at triage levels 3, 4, and 5 are less urgent, and they are classified based on the prediction of required resources. Version 4 of the ESI triage system, which has been approved by the Agency for Healthcare Research and Quality (AHRQ), is derived from evidence-based research [8,9]. Nurses routinely perform the triage process since they have received extensive practical and academic training on the employment of the ESI rules and the assessment of patient acuity. In such situations, the occurrence of miss-triage arises in the form of under-triage or over-triage, which might result in negative outcomes for patients waiting to receive care [10].