Emergency department (ED) visits have increased significantly over the past decade [1]. ED overcrowding has been known to lead to patient dissatisfaction, compromise patient safety, and even increase mortality [2], [3], [4], [5], [6]. There have been many attempts to assess the causes of ED overcrowding [7], [8], and length of stay (LOS) in the ED is generally used to estimate the degree of overcrowding [9]. Factors associated with ED LOS are divided roughly into input and output factors. Input factors consist of the number of patients, disease severity, and the percentage of non-emergency complaints [10], [11], while patient boarding time is the main output factor [12], [13]. Previous studies have suggested effective solutions including changing the ED registration process, additional physicians in triage, admitting boarded patients to hospital inpatient hallways, and transferring boarded patients to local hospitals [8], [14], [15].
In previous studies, it has also been suggested that consultation is an important independent variable that influences ED LOS and that consultation varies from 20% to 40% for all patients [10], [16], [17]. In most training hospitals in Korea, on-call resident physicians from each department are primarily responsible for the care of consulted patients and for admission decisions. Therefore, the process of the admission decision varies according to admission policies of each department and the consulted residents’ discretion. The custom of a multi-step notification process from junior residents through senior residents to faculty is another factor of delay.
With these factors in mind, we developed a new program by which short messages were auto-transmitted to specialized consultants based on information stored in the Computerized Physician Order Entry (CPOE) database. Short messages were transmitted if the dispositions of patients were not established at the times of 2 and 4 h after consultation and if the patient was still waiting for a hospital bed in the ED at 8 h after the admission decision. This strategy, called the “2-4-8 SMS (short message service) project,” was designed to encourage decision-makers to attend to patient delay and to facilitate decision-making. We hypothesized that the 2-4-8 SMS project would reduce ED LOS of consulted patients.
This was a pre–post-study performed in a single ED of an urban, academic tertiary care center with an average of 45,000 annual patients. This hospital was located in a densely populated area of 40 km2 with an estimated population of 570,000 people (14,000/km2). During the study period, there were no hospital closures in the area. There were 30 beds (9 monitored beds), 1 triage room, 1 resuscitation room and 1 procedure room in the ED, which was staffed by 11 emergency physicians, including 7
A total of 7518 patients were registered to the ED during the pre- and post-periods (Table 1). The proportion of children was lower, and the number of patients who visited during the day or on weekdays was smaller in the post-period. Other demographic findings were similar for both periods. Factors affecting ED overcrowding such as the ED facility (number of beds, triage and resuscitation room), medical equipment (portable ultrasonography and ultrasound cardiac output monitor) and human
The 2-4-8 SMS project successfully called attention to delays with emergency patients, resulting in reduced ED LOS. The median LOS was decreased 14 min for discharged patients and 36 min for admitted patients. When we analyzed the subcategories of LOS in admitted patients, the effect of the notification system was observed for both disposition time and boarding time.
Prolonged LOS resulting in ED overcrowding causes stress for the ED medical team. The cooperation of physicians among clinical
This study has several limitations. First, this study was a before-and-after comparative study. Although several known factors were adjusted for in the regression analysis, unknown confounding factors could have affected the results. Second, due to the nature of the study design as well as the short study time frame, we cannot presume a true relationship between the SMS implementation and reduction in length of stay of admitted patients. We made a continuous effort to reduce ED length of stay
This study suggests that the CPOE-based SMS program used to inform decision-makers of patient delay could reduce the LOS of consulted patients in the ED. This effect was mainly derived from the reduction in disposition time in single consultations or when they were conducted at night.
The work presented here was carried out collaboratively by all authors. HSL designed and implemented the SMS project with HSC and YSP. JMP and JSY contributed to the data acquisition and analysis. SMJ helped interpret the experimental results. MJK and SPC drafted this manuscript and all other authors critically revised it.
None of the authors have any conflict of interest, financial or otherwise, relevant to the conduct or reporting of this study. Summary points What was already known on the topic: Specialty consultation is one of the factors that cause prolonged ED LOS. It is necessary for consulting physicians to decide upon patients’ disposition more promptly in order to reduce ED LOS of consulted patients. CPOE is useful in improving patient care by enhancing communication and providing immediate access to
The authors express their thanks to the CPOE team of the Department of Information Systems at Gangnam Severance Hospital for their technical support. We would especially like to thank Mr. Soo Wan Park as the main contributor to this project.
Another study found improved physician consult response times after implementation of an institutional guideline,8 which is 60 min at our institution, a time frame our residents are comfortable seeing patients in. Other studies have found that computerized consultation management systems with automated messages have decreased consultation time.9,10 For our residents, where delays are usually due to being in the OR, with sick floor patients, in a trauma code, or during shift change, an automated message system would increase the quantity of pages and contribute further to pager fatigue.
There are fewer reports of process improvements in ED throughput and output, which often involve external stakeholders and can be harder to implement. Reported successes in this area include reducing laboratory processing time (in both a central laboratory and with point-of-care testing), using a bed supervisor to assure timely discharge from critical inpatient units, using an “emergency journey coordinator,” and using a short text message reminder to encourage consultants to complete ED evaluations (20–24). Holding orders (also known as interim orders, transfer orders, transition orders, or bridging orders) are another potential process change to improve ED flow.
They applied neural network methodology and the findings showed that neural net methodology application resulted in the derivation and validation of a modestly stronger hospital admission prediction model after the addition of CCC data. Kim et al. (2012) aimed at LOS reduction. They implemented a computerized short messaging service to inform care providers of patient delay in order to reduce LOS.
