Objective To quantify the extent to which time-based and operational factors drive emergency department (ED) overcrowding and evaluate the impact of targeted process improvements without constructing a new composite index.
Methods We conducted a single-center, retrospective before–after study of 6,881 adults with high triage acuity (Korean Triage and Acuity Scale levels 1–3) who were admitted from the ED across two consecutive 12-month periods: period 1 (2021-07-01 to 2022-06-30) and period 2 (2022-07-01 to 2023-06-30). Interventions were launched hospital-wide, targeting consultation responsiveness, earlier admission decision-making, faster ward bed assignment, and post-assignment transport and handover, which were implemented concurrently. The primary outcome was the change in ED length of stay (LOS; arrival-to-ED exit), and the secondary outcomes were changes in five key interval times along the care pathway.
Results The Mean ED LOS decreased significantly from 424.8±471.3 minutes to 283.2±306.9 minutes (–141.6 minutes, approximately 33% reduction; p<0.05). All interval means improved significantly (all p<0.05); notably, the 'admission decision-to-ED exit' interval shortened from 173.3±295.2 to 84.7±118.4 min. Improvements were consistent across all the high-acuity strata.
Conclusion Targeted pathway-level process improvements, especially those that shorten boarding, substantially reduced ED LOS and relieved overcrowding among admitted highacuity patients. A boarding-focused, hospital-wide approach is an operationally transferable framework for institutions that cannot implement composite crowding indices owing to data limitations although attribution to specific components requires more granular study designs.
OBJECTIVE This study integrates a machine learning (ML) based Score for Emergency Risk Prediction (SERP), developed using objective mortality endpoints with the Patient Acuity Category Scale (PACS) and evaluated its effectiveness in clinical use.
METHODS This single-centre, retrospective cohort study included all ED patients from a large tertiary hospital between 1 January 2018 and 31 December 2019. Using a reclassification framework, SERP was incorporated into PACS to derive two enhanced triage models. PACS+ model 1 downtriaged patients with low predicted 30-day mortality risk and up-triaged those with high risk. PACS+ model 2 up-triaged only high-risk patients, while low-risk patients retained their original category. Predictive performance in the test cohort was assessed using the area under the receiver operating characteristic curve (AUC) and decision curve analysis (DCA).
RESULTS The derivation cohort included 97,188 ED visits, and test cohort included 97,212 ED visits. In the derivation set, the mean (SD) age of patients was 58.97 (18.41) years old and 47,993 (49.4%) were females. Of all patients, 19.9%, 57.5%, 22.5%, and 0.2% were triaged to PACS categories 1–4 respectively. The 30-day mortality rate in the derivation set was 2.8% and 2.7% in the validation cohort. For 30-day mortality prediction, PACS+ model 1 (AUC 0.828 [95% CI 0.820-0.836]) and PACS+ model 2 (AUC 0.812 [95% CI 0.805-0.818]) outperformed PACS (AUC 0.722 [95% CI 0.714-0.729]). PACS+ model 1 consistently achieved greater net benefit across the range of clinical thresholds.
CONCLUSION Integrating ML-based SERP with PACS improved 30-day mortality prediction in ED triage.
365
View
13
Download
Emergency Medicine Practice and Administration | Education & Simulation
Donghyun Kim, Junsang Yoo, Ye Rim Lee, Ji Sim Yoon, Seung Jin Maeng, Minha Kim, Sejin Heo, Jong Eun Park, Gun Tak Lee, Se Uk Lee, Taerim Kim, Sung Yeon Hwang, Hee Yoon, Won Chul Cha, Hansol Chang
In Press, Received September 23, 2025 Accepted December 29, 2025 Available online February 27, 2026
Objective Emergency department (ED) physicians face substantial cognitive and physical demands, yet workload data applicable to real-world staffing and operational decisions remain limited. This study aimed to quantify perceived workload across diverse ED tasks using the NASA Task Load Index (NASA-TLX) and to determine how workload varies by physician experience, patient acuity, and clinical context. A secondary aim was to generate practical insights that may inform resource allocation and experience-based task distribution in the ED.
Methods We conducted an observational survey of interns, residents, and specialists working in the ED of a tertiary hospital between June and July 2022. NASA-TLX questionnaires were administered to assess workload across common procedures and patient-care tasks. Analyses were stratified by physician experience, Korean Triage and Acuity Scale (KTAS) level, and chief complaint. Nonparametric methods were used to evaluate differences in workload patterns.
Results Sixty physicians participated (30 interns, 30 residents/specialists). Procedures with high technical complexity, such as thoracentesis and lumbar puncture, showed the highest workload among interns. Among residents, workload decreased from postgraduate year 1 to 3 but rose again in year 4, reflecting increased supervisory responsibilities. Higher patient acuity (KTAS 1–2) and neurological chief complaints were consistently associated with elevated workload across all experience levels.
Conclusion Perceived workload in the ED varies significantly by task type, experience level, and patient acuity. These findings provide actionable data that may support evidence-based staffing decisions, workload redistribution, and training strategies to optimize physician performance and mitigate cognitive overload in resource-limited emergency departments.
Objective This study aims to systematically review the ethical and legal discussions regarding the utilization of artificial intelligence (AI) for patient triage and resource allocation in emergency medicine, and to identify the current state of discussions, their limitations, and future research directions.
Methods A comprehensive literature search was conducted following scoping review methodology. Relevant literature published after January 2020 was searched in the Web of Science, Scopus, CINAHL, PubMed, and Cochrane Library databases. Based on a PCC (population, concept, and context) framework (emergency patients/medical staff; triage, resource allocation; and emergency medicine with AI application), a final selection of 27 articles was analyzed.
Results The selected literature raised various ethical and legal issues related to the introduction of AI triage systems and AI utilization in emergency medicine, including data privacy, algorithmic bias, automation dependency, accountability, and explainability. In response to these issues, human-centered design, implementation of explainable AI, establishment of regulatory frameworks, continuous verification and evaluation, and ensuring human-in-the-loop were discussed as major solutions. However, discussions on the risks of “persuasive AI” that could mislead users, ethical issues of generative AI, and social validation and patient and public involvement were found to be insufficient.
Conclusion Ethical and legal discussions regarding AI in emergency medicine are evolving toward seeking concrete solutions at technical, institutional, and relational dimensions. However, in-depth research on ethical challenges, such as reflecting the specificity of rapidly developing AI and the values of emergency medicine, is urgently required.
Citations
Citations to this article as recorded by
The AI-IARA framework: How to cultivate human agency before artificial intelligence optimizes it a(ny)way Llewellyn E. van Zyl The Journal of Positive Psychology.2026; : 1. CrossRef
Objective Staffing significantly influences emergency department (ED) throughput; however, there is a shortage of registered nurses (RNs), impacting ED flow and crowding. Non-RN providers, like licensed practical nurses (LPNs), could potentially assist with tasks traditionally assigned to RNs. To improve the front-end ED process, we implemented an attending physician–LPN team (PNT) positioned next to triage and utilized existing ED hallway space. Methods This study took place at a tertiary care ED with over 110,000 annual visits. We compared postintervention (post-PNT) data (November 1, 2022–February 28, 2023) to preintervention (pre-PNT) data (July 31, 2022–October 31, 2022). The PNT, positioned adjacent to triage, expedited care for ED patients awaiting open rooms. The PNT selected patients from the waiting room to bypass the main ED, evaluated them in a private room, and then moved them to the hallway pending further care. Multivariable regression analysis was utilized to measure the impact of different factors on ED length of stay (LOS). Results We analyzed 23,516 patient visits, 10,288 in the pre-PNT period and 13,288 in the postPNT period. Post-PNT consisted of 2,454 PNT visits and 10,834 non-PNT visits. The intervention led to significant improvements, including a decrease in mean ED LOS from 492 to 425 minutes, a decrease in 72-hour revisits from 5.1% to 4.0%, a decrease in left-without-being-seen rate from 6.7% to 3.3%, and a decrease in the mean arrival-to-provider time from 74 to 60 minutes. Multivariable regression analysis showed that ED LOS was significantly lower for post-PNT patients than pre-PNT. Conclusion By leveraging the scope of LPNs and utilizing existing ED space, the PNT model successfully reduced front-end bottlenecks, leading to improved throughput and reduced revisitation and left-without-being-seen rates.
Objective Effective triage of febrile patients in the emergency department is crucial during times of overcrowding to prioritize care and allocate resources, especially during pandemics. However, available triage tools often require laboratory data and lack accuracy. We aimed to develop a simple and accurate triage tool for febrile patients by modifying the quick Sequential Organ Failure Assessment (qSOFA) score.
Methods We retrospectively analyzed data from 7,303 febrile patients and created modified versions of qSOFA using factors identified through multivariable analysis. The performance of these modified qSOFAs in predicting in-hospital mortality and intensive care unit (ICU) admission was compared using the area under the receiver operating characteristic curve (AUROC).
Results Through multivariable analysis, the identified factors were age (“A” factor), male sex (“M” factor), oxygen saturation measured by pulse oximetry (SpO2; “S” factor), and lactate level (“L” factor). The AUROCs of ASqSOFA (in-hospital mortality: 0.812 [95% confidence interval, 0.789–0.835]; ICU admission: 0.794 [95% confidence interval, 0.771–0.817]) were simple and not inferior to those of other more complex models (e.g., ASMqSOFA, ASLqSOFA, and ASMLqSOFA). ASqSOFA also displayed significantly higher AUROC than other triage scales, such as the Modified Early Warning Score and Korean Triage and Acuity Scale. The optimal cutoff score of ASqSOFA for the outcome was 2, and the score for redistribution to a lower level emergency department was 0.
Conclusion We demonstrated that ASqSOFA can be employed as a simple and efficient triage tool for emergency febrile patients to aid in resource distribution during overcrowding. It also may be applicable in prehospital settings for febrile patient triage.
Objective Emergency department (ED) triage systems are used to classify the severity and urgency of emergency patients, and Korean medical institutions use the Korean Triage and Acuity Scale (KTAS). During the COVID-19 pandemic, appropriate treatment for emergency patients was delayed due to various circumstances, such as overcrowding of EDs, lack of medical workforce resources, and increased workload on medical staff. The purpose of this study was to evaluate the accuracy of the KTAS in predicting the urgency of emergency patients during the COVID-19 pandemic. Methods This study retrospectively reviewed patients who were treated in the ED during the pandemic period from January 2020 to June 2021. Patients were divided into COVID-19–screening negative (SN) and COVID-19–screening positive (SP) groups. We compared the predictability of the KTAS for urgent patients between the two groups. Results From a total of 107,480 patients, 62,776 patients (58.4%) were included in the SN group and 44,704 (41.6%) were included in the SP group. The odds ratios for severity variables at each KTAS level revealed a more evident discriminatory power of the KTAS for severity variables in the SN group (P<0.001). The predictability of the KTAS for severity variables was higher in the SN group than in the SP group (area under the curve, P<0.001). Conclusion During the pandemic, the KTAS had low accuracy in predicting patients in critical condition in the ED. Therefore, in future pandemic periods, supplementation of the current ED triage system should be considered in order to accurately classify the severity of patients.
Citations
Citations to this article as recorded by
Prognostic Performance of the Korean Triage and Acuity Scale Combined with the National Early Warning Score for Predicting Mortality and ICU Admission at Emergency Department Triage: A Retrospective Observational Study Jungtaek Park, Sang Hoon Oh, Ae Kyung Gong, Jee Yong Lim, Sun Hee Woo, Won Jung Jeong, Ji Hoon Kim, In Soo Kim, Soo Hyun Kim Diagnostics.2026; 16(2): 345. CrossRef
Prognostic value of novel indices combining Shock Index, Reverse Shock Index, age, and oxygen saturation for predicting mortality in COVID-19 patients in Iran at emergency department triage: a cross-sectional study Mehdi Torabi, Atefe Noroozpour, Neda Naeemi Bafghi, Moghaddameh Mirzaee Acute and Critical Care.2025; 40(3): 425. CrossRef
Objective We evaluated the utility of the Korean Modified Early Warning Score (KMEWS), which combines the Modified Early Warning Score (MEWS) and the Korean Triage and Acuity Scale (KTAS), as a triage tool to screen for infection in patients who visit the emergency department.
Methods We retrospectively reviewed data extracted from electronic medical records. Patients aged ≥18 years with an infection who were admitted to the hospital via the emergency department between January 2018 and December 2019 were eligible for inclusion. The KMEWS score was calculated as the sum of the KTAS level and the MEWS score. We generated receiver operating characteristic curves and determined the area under the receiver operating characteristic curve (AUC) for the KMEWS, KTAS, MEWS, and Mortality in Emergency Department Sepsis (MEDS) scales. The primary outcome was septic shock, and secondary outcomes were intensive care unit admission and in-hospital mortality.
Results The AUC values (95% confidence interval) for predicting septic shock were as follows: KMEWS, 0.910 (0.902–0.918); MEWS, 0.896 (0.887–0.904); KTAS score, 0.809 (0.798–0.819); and MEDS, 0.927 (0.919–0.934). The AUC values (95% confidence interval) for predicting in-hospital mortality were as follows: KMEWS, 0.752 (0.740–0.764); MEWS, 0.717 (0.704–0.729); KTAS score, 0.764 (0.752–0.776); and MEDS, 0.844 (0.834–0.854). The AUC values (95% confidence interval) for predicting intensive care unit admission were as follows: KMEWS, 0.826 (0.816–0.837); MEWS, 0.782 (0.770–0.793); KTAS score, 0.821 (0.810–0.831); and MEDS, 0.839 (0.829–0.849).
Conclusion The KMEWS, which is a combination of the MEWS and the KTAS scores, might be a useful triage tool in emergency department patients who present with infection, particularly for predicting septic shock.
Citations
Citations to this article as recorded by
Evaluating the National Early Warning Score (NEWS) in triage: A machine learning perspective Arian Zaboli, Francesco Brigo, Serena Sibilio, Magdalena Massar, Gabriele Magnarelli, Gloria Brigiari, Gianni Turcato International Emergency Nursing.2025; 80: 101602. CrossRef
Development and validation of a transformer model-based early warning score for real-time prediction of adverse outcomes in the emergency department Hansol Chang, Jong Eun Park, Daehwan Lee, Kiwon Lee, Se Yong Jekal, Ki Tae Moon, Sejin Heo, Doyeop Kim, Gun Tak Lee, Sung Yeon Hwang, Won Chul Cha, Wonhee Kim, Tae Ho Lim, Tae Gun Shin Scientific Reports.2025;[Epub] CrossRef
Association between initial patient acuity and the predictive performance of the MREMS: A nationwide retrospective cohort study Álvaro Astasio-Picado, José Luis Martín-Conty, Begoña Polonio-López, Cristina Rivera-Picón, Alberto López Ballesteros, Alberto José Aragón Granados, Diego Villalobos Buitrago, Paula Álvarez Buitrago, Samanta Diaz-Gonzalez, Juan Dueñas-Ruiz, Francisco Mart The American Journal of Emergency Medicine.2025; 97: 84. CrossRef
Predictive validity of resource-adjusted Korean Triage and Acuity Scale in pediatric gastrointestinal tract foreign body patients Jin Hee Lee, Jin Hee Jung, Hyun Noh, Mi Jin Kim Scientific Reports.2024;[Epub] CrossRef
A model study for the classification of high-risk groups for cardiac arrest in general ward patients using simulation techniques Seok Young Song, Won-Kee Choi, Sanggyu Kwak Medicine.2023; 102(37): e35057. CrossRef
Objective To analyze the clinical significance of a heart rate (HR) or respiratory rate (RR) higher or lower than the normal in pediatric triage.
Methods A retrospective observational study was conducted with data from the Korean National Emergency Department Information System. The subjects were children <15 years of age in 2016. Reported HRs and RRs were divided into seven groups: grade -3 (3 or more standard deviations [SDs]normal), grade 2 (2 SDs>normal), and grade 3 (3 or more SDs>normal). The main outcomes were hospitalization and intensive care unit (ICU) admission rates. Logistic regression analysis was used to analyze the relationship of the outcomes according to grade in each group.
Results Data for 981,297 patients were analyzed. Hospitalization and ICU admission rates increased significantly in the higher HR group (grades 1 to 3; odds ratio [OR], 1.353; P<0.001; OR, 1.747; P<0.001; respectively) and in the higher RR group (OR, 1.144; P<0.001; OR, 1.396; P<0.001; respectively), compared with grade 0 group. In the lower HR group (grades -1 to -3), the hospitalization rate decreased (OR, 0.928; P<0.001), whereas the ICU admission rate increased (OR, 1.207; P=0.001). Although the hospitalization rate increased. In the lower RR group (OR, 1.016; P=0.008), the ICU admission rate did not increase (OR, 0.973; P=0.338).
Conclusion Deviations in HR and RR above normal are related to increased risks of hospitalization and ICU admission. However, this association may not apply to deviations below normal.
Objective We conducted a study to validate the effectiveness of the Korean criteria for trauma team activation (TTA) and compared its results with a two-tiered system.
Methods This observational study was based on data from the Korean Trauma Data Bank. Within the study period, 1,628 trauma patients visited our emergency department, and 739 satisfied the criteria for TTA. The rates of overtriage and undertriage in the Korean one-tiered system were compared with the two-tiered system recommended by the American College of Surgery-Committee on Trauma.
Results Most of the patient’s physiologic factors reflected trauma severity levels, but anatomical factors and mechanism of injury did not show consistent results. In addition, while the rate of overtriage (64.4%) was above the recommended range according to the Korean criteria, the rate of undertriage (4.0%) was within the recommended range. In the simulated two-tiered system, the rate of overtriage was reduced by 5.5%, while undertriage was increased by 1.8% compared to the Korean activation system.
Conclusion The Korean criteria for TTA showed higher rates of overtriage and similar undertriage rates compared to the simulated two-tier system. Modification of the current criteria to a two-tier system with special considerations would be more effective for providing optimum patient care and medical resource utilization.
Objective Vital sign trends are used in clinical practice to assess treatment response and aid in disposition, yet quantitative data to support this practice are lacking. This study aimed to determine the prognostic value of vital sign normalization.
Methods Secondary analysis of a prospective cohort of adult emergency department (ED) patients admitted a single urban tertiary care hospital. A random sample of 182 days was chosen, and a manual review of all admissions was undertaken. Persistent tachycardia or tachypnea was defined as failure to decrease to a normal value in the ED. Elevated upon admission was defined as an abnormal value at the last set of vital signs documented. The primary outcome was in-hospital mortality.
Results 4,878 patients were enrolled and 4.5 (±3.8) sets of vital signs were checked per patient. 1,770 patients were tachycardic and 1,499 were tachypneic. Among tachycardic patients, 941 (53%) were persistently tachycardic and 1,074 (61%) were tachycardic upon admission. Among tachypneic patients 639 (42%) were persistently tachypneic and 768 (51%) were tachypneic upon admission. Mortality was higher in patients persistently tachycardic (5.7% vs. 3.1%, P=0.008) or tachycardic upon admission (5.5% vs. 3.0%, P=0.014). Similar results were found in tachypneic patients (8.3% vs. 4.5%, P=0.003; 7.8% vs. 4.4%, P=0.006).
Conclusion Persistent tachycardia and tachypnea are associated with an increased risk of mortality in ED patients admitted to the hospital. Further study is necessary to determine if improved recognition or earlier interventions can affect outcomes.
Citations
Citations to this article as recorded by
Integrating probabilistic trees and causal networks for clinical and epidemiological data Sheresh Zahoor, Pietro Liò, Gaël Dias, Mohammed Hasanuzzaman Artificial Intelligence in Medicine.2026; 173: 103350. CrossRef
Guidelines for the Initial Assessment of Respiratory Distress in the Emergency Department P. Le Borgne, A.W. Thille, J. Guenezan, N. Aissaoui, A.-S. Boureau, C. Bally, F. Balen, A. Basset, P. Bilbault, F. Boissier, Y.-E. Claessens, M. Decavèle, J.-L. Diehl, D. Douillet, A. Guillon, P. Hausfater, F. Javaudin, M. Jezequel, K. Kuteifan, E. L’Her, Annals of Intensive Care.2026; 16: 100005. CrossRef
Predicting 28-Day Mortality in Critically Ill Patients Receiving Continuous Renal Replacement Therapy: A Novel Interpretable Machine Learning Approach Tao Zhang, Zi-Han Nan, Xiao-Xuan Fan, Jing-Xiao Pang, Cong-Cong Zhao, Yan Xin, Zhen-Jie Hu, Shao-Han Guo Journal of Multidisciplinary Healthcare.2025; Volume 18: 5535. CrossRef
Multiple organ scoring systems for predicting in-hospital mortality of sepsis patients in the intensive care unit Xuan Zhou, Zhenen Zhang, Huimin Wang, Pengfei Chen Open Medicine.2025;[Epub] CrossRef
Frequency of and associations with alterations of medical emergency team calling criteria in a teaching hospital emergency department Simon R. Baylis, Luke R. Fletcher, Alastair J.W. Brown, Tamishta Hensman, Ary Serpa Neto, Daryl A. Jones Australian Critical Care.2024; 37(2): 301. CrossRef
Heart/breathing rate ratio (HBR) as a predictor of mortality in critically ill patients Tong Yan Zhang, Ya Jun Du, Ya Zhu Hou, Qian Du, Hai Rong Dou, Xiu Mei Gao Heliyon.2024; 10(10): e31187. CrossRef
The significance of APACHE II as a predictor of mortality in paraquat poisoning: A systematic review and meta-analysis Harsimran Kaur, Viji Pulikkel Chandran, Muhammed Rashid, Vijayanarayana Kunhikatta, Pooja Gopal Poojari, Shankar M. Bakkannavar, Jayaraj Mymbilly Balakrishnan, Girish Thunga Journal of Forensic and Legal Medicine.2023; 97: 102548. CrossRef
How to facilitate respiratory rate measurement in the emergency room Takahiko Nagamine Japan Journal of Nursing Science.2022;[Epub] CrossRef
Development of New Equations Predicting the Mortality Risk of Patients on Continuous RRT Min Woo Kang, Navdeep Tangri, Soie Kwon, Lilin Li, Hyeseung Lee, Seung Seok Han, Jung Nam An, Jeonghwan Lee, Dong Ki Kim, Chun Soo Lim, Yon Su Kim, Sejoong Kim, Jung Pyo Lee Kidney360.2022; 3(9): 1494. CrossRef
The Association Between Abnormal Vital Signs and Mortality in the Emergency Department Jood H Simbawa, Abdulkarim A Jawhari, Fay Almutairi, Ahlam Almahmoudi, Bashair Alshammrani, Raneem Qashqari, Ibtihal Alattas Cureus.2021;[Epub] CrossRef
Development of a Simple Sequential Organ Failure Assessment Score for Risk Assessment of Emergency Department Patients With Sepsis Faheem W. Guirgis, Michael A. Puskarich, Carmen Smotherman, Sarah A. Sterling, Shiva Gautam, Frederick A. Moore, Alan E. Jones Journal of Intensive Care Medicine.2020; 35(3): 270. CrossRef
The value of vital sign trends in predicting and monitoring clinical deterioration: A systematic review Idar Johan Brekke, Lars Håland Puntervoll, Peter Bank Pedersen, John Kellett, Mikkel Brabrand, Shane Patman PLOS ONE.2019; 14(1): e0210875. CrossRef
First
Prev
