A Novel Deep Learning–Based System for Triage in the Emergency Department Using Electronic Medical Records: Retrospective Cohort Study

Background Emergency department (ED) crowding has resulted in delayed patient treatment and has become a universal health care problem. Although a triage system, such as the 5-level emergency severity index, somewhat improves the process of ED treatment, it still heavily relies on the nurse’s subjective judgment and triages too many patients to emergency severity index level 3 in current practice. Hence, a system that can help clinicians accurately triage a patient’s condition is imperative. Objective This study aims to develop a deep learning–based triage system using patients’ ED electronic medical records to predict clinical outcomes after ED treatments. Methods We conducted a retrospective study using data from an open data set from the National Hospital Ambulatory Medical Care Survey from 2012 to 2016 and data from a local data set from the National Taiwan University Hospital from 2009 to 2015. In this study, we transformed structured data into text form and used convolutional neural networks combined with recurrent neural networks and attention mechanisms to accomplish the classification task. We evaluated our performance using area under the receiver operating characteristic curve (AUROC). Results A total of 118,602 patients from the National Hospital Ambulatory Medical Care Survey were included in this study for predicting hospitalization, and the accuracy and AUROC were 0.83 and 0.87, respectively. On the other hand, an external experiment was to use our own data set from the National Taiwan University Hospital that included 745,441 patients, where the accuracy and AUROC were similar, that is, 0.83 and 0.88, respectively. Moreover, to effectively evaluate the prediction quality of our proposed system, we also applied the model to other clinical outcomes, including mortality and admission to the intensive care unit, and the results showed that our proposed method was approximately 3% to 5% higher in accuracy than other conventional methods. Conclusions Our proposed method achieved better performance than the traditional method, and its implementation is relatively easy, it includes commonly used variables, and it is better suited for real-world clinical settings. It is our future work to validate our novel deep learning–based triage algorithm with prospective clinical trials, and we hope to use it to guide resource allocation in a busy ED once the validation succeeds.

Comparison Between Emergency Severity Index Plus Cardiac Troponin I Rapid Test and Emergency Severity Index in Patients Presenting with Low-Risk Chest Pain: A Randomized Clinical Trial

Background: There is difficulty in identifying low-risk patients with acute coronary syndrome in the emergency department (ED). Objectives: The aim of this study was to compare mistriage between the Emergency Severity Index (ESI) plus the cardiac troponin I rapid test (cTnI) and ESI among patients with chest pain. Methods: A randomized clinical trial was conducted from January to April 2019. One hundred patients with low-risk chest pain were randomly allocated to the ESI + cTnI and ESI groups. Triage levels, used resources, and mistriage rate were compared between both groups among patients discharged from the ED and admitted to the cardiac unit (CU) or coronary care unit (CCU). Results: Our samples included 100 patients (age: 52.9 ± 13.92 years; 51% female) who were equally assigned to the ESI + cTnI and ESI groups. Overtriage rate was 6% and 88% for the ESI + cTnI and ESI groups, respectively. The triage level between the ESI + cTnI and ESI groups was significantly different among patients who were discharged from the ED (3.92 vs. 3.00). Conclusions: The ESI + cTnI score seems to be more valid than the ESI scale to triage patients with low-risk chest pain. It is recommended to add cTnI to the ESI for the triage of patients with low-risk chest pain in the ED.

Comparative Analysis Of Triage Systems At Emergency Departments Of Different Countries: Implementation In Kazakhstan

Medical sorting is aimed at assessment of disease severity and has to be carried out within a short time to determine the priorities for patient care and transportation to the most appropriate place for future treatment. The goal of this study was to provide an integrative review by analyzing the publications on the most common triage systems worldwide in order to select and implement the most reliable system at emergency departments. We searched for publications relevant to our comparative analysis in evidence-based medicine databases. A total of 1,740 literary sources were identified, of which 42 were selected for analysis. Comparative analysis of different triage systems may help implementing the most efficient system in Kazakhstan. The Emergency Severity Index is considered the most reliable and accurate tool used in international practice, and it could provide a basis for introduction of triage system at emergency departments in Kazakhstan.

Prevalence and Factors Associated with Attendance of Non-Urgent Cases of Pediatric Emergency Room in Riyadh, Saudi Arabia

The care for emergency conditions is an important feature of developed the systems of healthcare. Emergency medicine is the medical specialty dedicated to diagnosis and treatment of unforeseen illness or injuries. Overcrowding of Emergency Room (ER) by non-urgent cases is a problem that should be raised and reasons for that need to be sought. The aim is to identify the prevalence and factors behind non-urgent cases attending to acute care in Riyadh. This is a cross-sectional study which was conducted over a one-year period. Questionnaires were distributed to parents of children ≤14 years attending pediatric emergency room (ER) in multiple hospitals in Riyadh, Saudi Arabia. The sample size was 383 participants recruited using convenience sampling technique. The inclusion criteria comprised of pediatrics ≤14 years attending to ER with urgent cases (level 1, 2 and 3 according to Emergency Severity Index) and minor trauma (contusion, abrasion), sore throat, upper respiratory tract infection (URTI), mild abdominal pain, vomiting alone or diarrhea alone with no sign of dehydration of (level 4 and 5 according to Emergency Severity Index). 383 questionnaires were collected. The sample consisted mostly of females 56%. Upper respiratory tract infection (URTI) was the highest reason for pediatric emergency room visits (55.1%). Only 12.2% have visited primary health care (PHC) prior to emergency room. Most of participants have sources for medical advice other than emergency physicians (80.3%). Results also show that many parents who have brought their kids to emergency lacked the knowledge that their cases could be managed in primary health care (76.2%). In Conclusion, results have shown that many parents choose ER whenever their children have any symptoms. Reasons for that varied between each participant. Many participants are ignorant of the capabilities of PHC centers, and their services. Raising awareness regarding primary health should be the objective to reduce number of non-urgent cases visiting emergency room.

APLIKASI METODE TRIAGE ”EMERGENCY SEVERITY INDEX( ESI)” TERHADAP KETEPATAN PENENTUAN TINGKAT KEGAWATDARURATAN OLEH MAHASISWA PROGRAM S1 KEPERAWATAN

Triage merupakan salah satu skill yang wajib dikuasai seorang perawat yang telah dipelajari semenjak pada masa pendidikan keperawatan. Terdapat beberapa jenis triage, salah satunya adalah metode Emergency Severity Index (ESI). Metode ESI merupakan metode triage dengan prinsip memprioritaskan kegawatdaruratan pasien berdasarkan banyaknya jumlah kebutuhan medis yang di butuhkan. Tujuan pelitian ini adalah mengetahui efektifitas metode triage ESI terhadap ketepatan penentuan tingkat kegawatdaruratan pada mahasiswa program S1 keperawatan. Penelitian ini menggunakan metode one grup pretest postes dan menggunakan metode purposive sampling sejumlah 85 mahasiswa program S1 keperawatan dengan menetapkan kriteria inklusi untuk menghomogenkan sample. Intrumen dalam penelitian ini menggunakan kuesioner kasus klinik dan edukasi bagan ESI. Uji statistik menggunakan Wilcoxon Signed Ranks Test didapatkan p value sebesar 0.018, dapat disimpulkan terdapat perbedaan pemahaman prioritas kegawatdaruratan sebelum dan sesudah pengenalan metode pengenalan metode. Peneletian selanjutnya diharapkan mengaplikasikan metode pembelajaran triage yang menarik dan inovatif serta menganalisa efektifitasnya sebagai sarana pemahaman prioritas kasus-kasus kegawatdaruratan.