The Korean Triage and Acuity Scale (KTAS)
The Korean Triage and Acuity Scale (KTAS) represents a cornerstone in South Korea's emergency medical services, serving as a pivotal 5-level emergency department (ED) triage tool. Developed in 2012 by the Korean Society of Emergency Medicine (KSEM)
The Korean Triage and Acuity Scale (KTAS) represents a cornerstone in South Korea's emergency medical services, serving as a pivotal 5-level emergency department (ED) triage tool. Developed in 2012 by the Korean Society of Emergency Medicine (KSEM) and based on the Canadian Triage and Acuity Scale (CTAS) with adaptations for the local medical environment, KTAS was nationally implemented in emergency medical centers starting January 2016. Its fundamental purpose is to mitigate ED congestion, enhance patient safety through prioritized care based on severity and acuity, and streamline emergency medical service delivery.
The implementation of KTAS has yielded significant positive outcomes, including a notable reduction in the average length of stay (LOS) for patients in the ED and a decrease in overall mortality rates. Furthermore, the system has demonstrably improved the patterns of patient admission and disposition, indicating more appropriate patient channeling within the healthcare system. The scale consistently exhibits high inter-rater reliability among emergency nurses and expert assessors, which is crucial for its consistent application across different medical facilities.
Despite its successes, KTAS faces several challenges and limitations. These include persistent issues with mistriage, particularly stemming from the subjective nature of pain assessment, which can lead to overclassification of patient severity. Public health crises, such as the COVID-19 pandemic, have also exposed limitations in its predictive accuracy for critical conditions, especially those related to viral diseases. Moreover, the scale's performance in specific patient populations, notably the elderly, has shown reduced predictive power, indicating a need for specialized considerations. A significant policy concern arises from administrative reconfigurations by the Ministry of Health and Welfare (MOHW) to reduce the 5-level system to 3 or 2 levels for payment purposes. This simplification contradicts global best practices for triage systems and has been shown to diminish the tool's effectiveness in predicting patient disposition.
Looking ahead, the evolution of KTAS is poised for continuous refinement. This includes the integration of advanced technologies such as artificial intelligence (AI) and natural language processing (NLP) to further enhance triage accuracy and efficiency. The development of specialized guidelines and adaptations for vulnerable populations, such as incorporating frailty scales for the elderly, is also a crucial area for improvement. Adapting the system for future pandemic scenarios to ensure robust performance during widespread infectious disease outbreaks remains a key imperative.
Introduction to Emergency Department Triage and the Korean Context
The Indispensable Role of Triage in Modern Emergency Medicine
Triage stands as an indispensable process within the emergency department, serving as the initial and critical step in prioritizing patient care. Its fundamental purpose is to ensure patient safety and facilitate the smooth operation of the ED, a necessity given the annually increasing number of patients seeking emergency care. At its core, triage involves the judicious distribution of limited medical resources based on the acuity and severity of a patient's condition.
An effective triage system goes beyond mere classification; it accurately prioritizes patients, aiming to prevent both undertriage (underestimation of severity) and overtriage (overestimation of severity). Such precision is vital for managing ED overcrowding, a pervasive issue that can lead to a decline in the quality of care, prolonged patient waiting times, increased medical errors, and, critically, higher mortality rates. The ability of a triage system to effectively mediate between finite medical resources and unpredictable patient demand positions it as a cornerstone of public health infrastructure, directly impacting morbidity and mortality by ensuring that the most critical patients receive timely attention while less urgent cases are managed efficiently.
Global Triage Systems Landscape: A Brief Overview
Globally, emergency department triage systems exhibit variations in their structural design, commonly employing 3-level, 4-level, or 5-level scales. Among these, 5-level systems have gained broader acceptance due to their demonstrated reliability and capacity for more nuanced patient assessment. Prominent international examples include the Canadian Triage and Acuity Scale (CTAS), the Emergency Severity Index (ESI) used in the United States, the Australasian Triage Scale (ATS), and the Manchester Triage System (MTS).
While CTAS and ATS primarily focus on presenting symptoms and diagnoses to determine how long a patient can safely wait for care, the ESI, also a 5-level system, incorporates anticipated resource utilization as a key determinant for classifying patients into levels 3, 4, and 5, in addition to acuity. The global inclination towards 5-level systems underscores a recognition of the need for finer granularity in patient assessment. This detailed stratification is essential for optimizing resource allocation and enhancing patient safety, as it allows for a more precise differentiation of urgency and severity, which directly influences wait times, treatment pathways, and the efficient deployment of resources. This global trend highlights a critical contrast with certain internal Korean policy decisions regarding KTAS, which have moved towards simplifying triage levels for administrative purposes, potentially undermining the very functionality that 5-level systems are designed to achieve.
The Unique Challenges and Evolution of Emergency Medical Services in South Korea
Before the advent of KTAS, South Korea's emergency departments operated with a fragmented array of triage approaches. Some institutions utilized internationally recognized tools such as ATS, CTAS, or ESI. Others relied on a rudimentary two-level system that merely distinguished between "emergency" and "non-emergency" patients based on definitions stipulated in Korean emergency medical law. A third category involved practitioners making triage decisions based solely on subjective judgment, lacking any validated or reliable tool. A concerning statistic reveals that approximately 10% of emergency medical institutions in the country did not employ any formal triage tool whatsoever.
This pervasive variability and the widespread absence of validated triage tools led to significant complexity and ambiguity in patient classification across the nation's emergency medical landscape. The lack of standardization meant inconsistent patient care, inefficient resource utilization across different institutions, and considerable difficulty in collecting uniform data for national quality improvement initiatives. It also implied potential disparities in patient outcomes depending on which ED a patient visited. This historical fragmentation underscored a profound systemic vulnerability in emergency care prior to KTAS, highlighting the urgent need for a unified, evidence-based framework to elevate the entire emergency medical system and ensure equitable and consistent care delivery nationwide.
The Korean Triage and Acuity Scale (KTAS): Foundations and Operational Framework
Historical Development and Genesis
Previous Triage Systems in Korea: A Landscape of Variability
Before 2012, the emergency medical centers in Korea presented a diverse and often inconsistent approach to patient triage. The landscape was characterized by a mix of practices: some institutions had adopted internationally recognized tools like the Australasian Triage Scale (ATS), the Canadian Triage and Acuity Scale (CTAS), and the Emergency Severity Index (ESI). However, a significant portion of centers either employed a simplistic two-level distinction between "emergency" and "non-emergency" based on Korean legal definitions, or relied entirely on the subjective judgment of triage practitioners, without any verified validity or reliability in their methods. Alarmingly, about 10% of emergency medical institutions operated without any formal triage tool at all. This patchwork of approaches created substantial complexity and ambiguity in patient classification, representing a critical gap in patient safety and efficient resource management within the Korean emergency medical system. The inconsistencies arising from unvalidated or absent triage tools directly impacted patient outcomes, contributed to ED overcrowding, and hindered the overall efficiency and reputation of emergency care.
The Imperative for Standardization: Role of the Korean Society of Emergency Medicine (KSEM)
Recognizing the urgent need for a standardized, reliable, and valid triage tool to address the fragmented and ambiguous systems, the Korean Society of Emergency Medicine (KSEM) took a proactive leadership role. In 2012, KSEM spearheaded the development of the Korean Triage and Acuity Scale (KTAS). KSEM, a professional organization deeply involved in academics, examinations, training, and policy related to emergency medicine, developed KTAS through extensive expert consultation, including surveys to gather diverse perspectives.
Following the development, KSEM commissioned rigorous research to confirm the validity and reliability of the KTAS, demonstrating a commitment to evidence-based practice. The multi-faceted involvement of KSEM, encompassing development, validation, and training, was crucial for the successful national implementation of KTAS. This approach highlights how professional medical societies can effectively drive national healthcare quality improvements, particularly when existing governmental frameworks are insufficient. KSEM's role ensured that the new triage tool was not only clinically relevant but also continuously supported through education and validation efforts, establishing a unified and equitable standard of care nationwide.
KTAS: Adaptation from the Canadian Triage and Acuity Scale (CTAS) – Key Similarities and Differences
The Korean Triage and Acuity Scale (KTAS) was developed based on the Canadian Triage and Acuity Scale (CTAS), incorporating "slight modifications" to align with the specific medical environment in Korea. This strategic choice to leverage an internationally validated framework accelerated the development process and built upon established best practices in emergency triage.
Similarities: Both KTAS and CTAS are structured as 5-level triage systems, a design widely accepted globally for its enhanced reliability and granularity in patient assessment. They share a very similar process for assessing patient acuity. The five levels and their core definitions are consistent across both scales:
Level 1: Resuscitation
Level 2: Emergent
Level 3: Urgent
Level 4: Less Urgent
Level 5: Non-Urgent
Differences: While the precise details of the "slight modifications" are not extensively enumerated in the available information, the intent was to tailor the system to the unique "medical environment in Korea". This implies adjustments to account for local disease prevalence, specific healthcare infrastructure, cultural factors influencing patient presentation, or resource availability within the Korean context. A notable difference lies in the specific complaint lists: KTAS includes 155 main complaints for adults and 165 for children , a level of specificity that may represent a tailored approach for the Korean population.
A significant divergence from the CTAS model, and indeed from global best practices, emerged when the Korean Ministry of Health and Welfare (MOHW) administratively reconfigured the 5-level KTAS. For purposes of administrative simplicity and payment, the MOHW consolidated the system into a 3-level tool, reassigning Levels 1 and 2 as "severely emergent patients," Level 3 as "severely emergent suspected patients," and Levels 4 and 5 as "mild emergent or non-emergent patients". The National Health Insurance Service (NHIS) further simplified this for payment, only providing weighted fees for "severely emergent patients" and "severely emergent suspected patients," effectively creating a two-level payment classification. This policy decision, contrary to recommendations against reducing the number of levels, risks undermining the nuanced clinical utility inherited from CTAS. Studies have indicated that the 5-level KTAS maintains its largest relative importance for ED disposition only at its original level, and attempts to reduce these levels diminish its predictive power. This creates a tension between administrative convenience and optimal clinical outcomes, where non-clinical objectives can inadvertently dilute the clinical efficacy of a well-designed triage system.
Core Principles and Structured Triage Process
The Five-Level Acuity System: Definitions and Clinical Criteria
The KTAS classifies patients into five distinct levels, meticulously designed to reflect both the severity and acuity of their condition. This detailed 5-level classification, coupled with specific clinical criteria and associated wait time goals, aims to standardize not only the assessment but also the expected care delivery timelines, which is crucial for managing patient expectations and ensuring timely interventions.
The levels, along with their definitions and representative symptoms, are as follows:
Level 1: Resuscitation (최우선수위 - Highest Priority)
Definition: Requires immediate intervention for conditions that are an immediate threat to life or limb, or have a high potential for rapid deterioration.
Representative Symptoms: Cardiac arrest, apnea, unconsciousness unrelated to alcohol, severe trauma. This also includes patients unable to protect their airway, those responding only to pain or loud noise without purpose, continuous seizures, progressive deterioration in consciousness (Glasgow Coma Scale 3-8), severe hypoperfusion (marked pallor, cool skin, weak pulse, hypotension), or severe respiratory distress (cyanosis, single-word speech, SpO2 <90%).
Wait Time Goal: Immediately.
Level 2: Emergent (2순위 - Second Priority)
Definition: Involves a potential threat to life, limb, or organ function, necessitating rapid treatment to prevent further deterioration.
Representative Symptoms: Myocardial infarction, cerebral hemorrhage, stroke, significant dyspnea, hematemesis. Patients may exhibit an inappropriate response to verbal stimuli, loss of orientation to person, place, or time, new impairment of recent memory, altered behavior, increased work of breathing (SpO2 <92%), or signs of borderline perfusion (pallor, history of diaphoresis, unexplained tachycardia, postural hypotension).
Wait Time Goal: Within 15 minutes.
Level 3: Urgent (3순위 - Third Priority)
Definition: A condition that requires treatment and has the potential to progress to a more serious state if not addressed.
Representative Symptoms: Mild respiratory distress (SpO2 >90%), diarrhea with bleeding, moderate abdominal pain, headache. Patients might present with dyspnea on exertion (SpO2 <94%), vital signs at the upper or lower ends of normal that differ from their usual values, or appear ill but with fewer than three Systemic Inflammatory Response Syndrome (SIRS) criteria.
Wait Time Goal: Within 30 minutes.
Level 4: Less Urgent (4순위 - Fourth Priority)
Definition: A condition where treatment or re-evaluation within 1-2 hours is acceptable, taking into account the patient's age, pain level, or potential for worsening or complications.
Representative Symptoms: Enteritis with fever above 38∘C, urinary tract infection with abdominal pain, confusion. Patients may have fever as their only positive SIRS criterion and appear comfortable and not in distress.
Wait Time Goal: Within 1 hour.
Level 5: Non-Urgent (5순위 - Fifth Priority)
Definition: A non-emergent condition, often a chronic issue, or a condition with a low potential for worsening.
Representative Symptoms: Common cold, simple enteritis, diarrhea, wound dressing, medication prescriptions.
Wait Time Goal: Within 2 hours.
This granular classification with explicit clinical criteria and target wait times transforms KTAS from a mere classification tool into a comprehensive operational framework. This framework aims to optimize patient flow, ensure timely care based on urgency, and provide a measurable standard for ED performance, thereby directly addressing issues of overcrowding and patient safety.
The KTAS Algorithm: Critical First Look, Infection Screening, Primary Symptom Selection, and Primary/Secondary Considerations
The KTAS evaluation process is meticulously structured and symptom-oriented, designed to ensure comprehensive and consistent patient assessment. This multi-step, algorithmic approach moves from immediate life threats to more specific clinical details, aiming to reduce reliance on subjective judgment, although certain subjective elements continue to pose challenges.
Critical First Look: Upon a patient's arrival at the ED, a "critical first look" is immediately performed. This initial impression evaluation serves as a rapid filter to quickly identify immediately life-threatening conditions, reflecting the time-critical nature of emergency medicine.
Screening for Infectious Disease: Following the critical first look, a screening test for infectious diseases is conducted. This step is crucial for public health, as it influences patient allocation to either isolation or non-isolation areas, thereby preventing nosocomial transmission within the ED.
Primary Symptom Selection (Main Complaint): The triage process then proceeds to the selection of the patient's primary symptom or main complaint. KTAS provides a detailed list, including 155 main complaints for adults and 165 for children, ensuring a comprehensive initial categorization.
Primary Considerations: After the main complaint is identified, common and general characteristics applicable to a broad range of conditions are assessed. These "primary considerations" include:
Respiration status and respiratory rate.
Vital signs, encompassing systolic and diastolic blood pressure, pulse rate, body temperature, and oxygen saturation.
Mental status and level of consciousness.
Pain assessment.
Presence of hemorrhagic disease.
Mechanism of injury.
Secondary Considerations: These are more specific characteristics, relevant to the particular main complaint. Examples include:
Blood sugar levels measured at the time of visit.
Symptoms of dehydration.
High blood pressure.
Pregnancy status.
Mental health status, assessed based on medical history and patient responses.
The final KTAS level is determined by selecting the applicable items from both primary and secondary considerations. In instances where multiple considerations apply, the algorithm mandates that the more acute KTAS level be selected, demonstrating a built-in bias towards patient safety by prioritizing potential severity. The overarching goal of this entire process is to triage a patient within 10 to 15 minutes of their arrival at the ED. This comprehensive algorithm represents a sophisticated attempt to codify expert emergency medical decision-making into a reproducible process, balancing the need for rapid assessment of life-threatening conditions with a thorough evaluation of patient complaints and physiological parameters.
Categorization of Main Complaints: Adult and Pediatric Specificities
KTAS incorporates a meticulously detailed list of main complaints, distinguishing between adults and children with 155 and 165 complaints, respectively. This explicit differentiation in main complaints and assessment criteria for adult and pediatric patients signifies a recognition of the unique physiological and pathological characteristics inherent in these distinct patient populations.
Initial assessment for pediatric patients requires careful consideration of the broad presentation of illness and injuries, as their symptoms and signs often differ significantly from those in adults due to age-dependent physiological responses. To ensure appropriate classification, pediatric KTAS is specifically designed for individuals under 15 years old, while adult KTAS applies to those aged 15 and above. This tailored approach is crucial for accurate triage because applying adult criteria to children can lead to significant mistriage, either by missing subtle signs of severe illness (undertriage) or by misinterpreting normal pediatric variations as pathology (overtriage). The age-specific categorization within KTAS reflects an advanced understanding of emergency medicine, acknowledging that effective triage necessitates a developmentally appropriate approach. This design choice enhances the scale's precision and reinforces its commitment to patient safety across the lifespan, particularly for vulnerable pediatric populations.
IV. Implementation, Adoption, and Operational Impact of KTAS
Nationwide Rollout and Integration
The Korean Triage and Acuity Scale underwent development between 2012 and 2015, culminating in its official implementation across emergency medical centers throughout South Korea starting January 2016. This rapid and mandated nationwide rollout by the Ministry of Health and Welfare in 2016 signifies a strong governmental commitment to standardizing emergency care.
A critical enabler of this widespread adoption is KTAS's full integration into electronic medical record (EMR) systems in many emergency departments. This integration facilitates real-time scoring and data collection, allowing for the application of KTAS via vital signs recorded in the EMR and through face-to-face history-taking by ED nurses. This technological backbone enhances efficiency and data capture.
Accompanying the implementation, extensive training and education programs were developed for ED staff, particularly triage nurses. These programs are designed to enhance theoretical knowledge, improve decision-making capabilities, and refine emergency treatment abilities, especially for those lacking extensive clinical experience. The emphasis on training implicitly acknowledges the inherent complexity of the scale and the persistent need for skilled human interpretation, even with technological support. The success of KTAS, therefore, is a function of both systemic policy and individual competency, as even advanced algorithms require skilled human operators to navigate clinical nuances and subjective assessments effectively.
Impact on Emergency Department Efficiency and Patient Flow
Changes in Emergency Department Length of Stay (LOS)
The implementation of KTAS has profoundly impacted emergency department operations, most notably by significantly reducing the mean ED Length of Stay (LOS). Prior to KTAS, the mean LOS was 343 minutes (standard deviation = 432 minutes). Following its implementation, this figure decreased significantly to 289 minutes (SD = 333 minutes), representing a mean reduction of 54 minutes (P<0.001).
This reduction in LOS was not confined to a single area; it was observed across various medical departments, including cardiovascular, dental surgery, general surgery, internal medicine, orthopedic surgery, pediatrics, and thoracic surgery. Furthermore, the decrease in LOS was consistent across different patient dispositions, such as discharge from the ED, admission to a general ward, admission to an intensive care unit (with and without surgery), and admission to a general ward after surgery. Even for the most critical patients, those classified as Level 1, the LOS saw a significant decrease from 419 minutes to 356 minutes (
P=0.040) after KTAS implementation. This substantial reduction in ED LOS serves as a direct indicator of improved operational efficiency and patient flow, suggesting that better patient prioritization and resource allocation facilitated by KTAS actively mitigate ED overcrowding, a critical challenge in emergency medicine. By enabling quicker identification of critical patients and streamlining less urgent cases, KTAS helps prevent bottlenecks and ensures timely care, directly contributing to a more efficient patient throughput.
Influence on Patient Admission and Disposition Patterns
Beyond merely reducing the time patients spend in the ED, the implementation of KTAS also significantly altered the patterns of patient admission and disposition. Before KTAS was widely adopted, there were instances of misclassification where less serious patients might be assigned severe triage levels, while genuinely serious patients received lower classifications. This suggests that the introduction of KTAS improved the accuracy of initial patient classification, leading to more appropriate care pathways.
The observed shift in admission and disposition patterns indicates that KTAS enhanced the appropriateness of care. Accurate triage, enabled by KTAS, reduces both undertriage (where sicker patients receive delayed critical care) and overtriage (where less sick patients consume high-acuity resources). By correctly identifying patient severity, KTAS guides individuals to the most suitable care setting—whether it be the intensive care unit, a general ward, or discharge—more efficiently. This optimization of bed utilization ensures that high-acuity resources are reserved for those who truly need them, thereby improving resource stewardship and ensuring that patients receive the right level of care at the right time. This represents a critical component of high-quality emergency medicine.
Reduction in Overall Mortality Rates Post-Implementation
Perhaps the most compelling evidence of KTAS's positive impact is the statistically significant reduction in overall mortality rates following its implementation. The total mortality rate decreased from 0.36% (213 patients) in the period before KTAS to 0.28% (179 patients) in the period after its adoption (P=0.020).
Furthermore, mortality from causes other than cardiopulmonary resuscitation also saw a significant reduction, dropping from 0.15% to 0.09%. This suggests a broader impact on preventing patient deterioration across various medical conditions, not just immediate cardiac arrest. The reduction in mortality is likely a cumulative effect of several factors: more accurate identification of critically ill patients, which minimizes undertriage and leads to faster access to life-saving interventions; improved patient flow and reduced delays in care; and a more optimized allocation of high-acuity resources. This outcome validates KTAS's role as a critical patient safety intervention, demonstrating that a well-designed and properly implemented triage system can be a direct determinant of patient survival, reinforcing its status as a foundational element of effective emergency medical care.
Financial and Policy Implications
The National Health Insurance Service (NHIS) in Korea established an emergency medical payment system based on KTAS in 2016, integrating financial incentives with triage performance. This system provides direct compensation of approximately USD 3.3 per patient to institutions for performing KTAS triage, encouraging its widespread adoption. Additionally, it offers higher medical fees by assigning a relative weight to the basic medical fee according to the acuity and severity level determined by KTAS.
However, a critical policy misstep occurred when the Ministry of Health and Welfare (MOHW) reconfigured the 5-level KTAS for administrative simplicity. Despite the 5-level KTAS being the original and most effective form for patient disposition, the MOHW consolidated it into a 3-level tool. In this reclassification, Levels 1 and 2 were grouped as "severely emergent patients," Level 3 as "severely emergent suspected patients," and Levels 4 and 5 as "mild emergent or non-emergent patients". The NHIS further reduced this for payment purposes, only paying weighted fees for "severely emergent patients" and "severely emergent suspected patients," effectively creating a two-level payment classification.
This reduction in levels directly contradicts the global trend towards 5-level systems, which are widely accepted as more reliable. Research indicates that the 5-level KTAS has the largest relative importance for ED disposition only at its original, granular level, and attempts to reduce these levels diminish its predictive power. This policy decision creates a direct conflict between the clinical utility of a granular triage system and the desire for administrative ease, potentially undermining the very goals of patient safety and optimal resource allocation that KTAS was designed to achieve. Simplified payment structures can inadvertently disincentivize accurate, nuanced clinical classification, leading to less precise patient categorization, which in turn can result in increased undertriage (missing subtle severity) or overtriage (over-classifying for higher reimbursement). This distortion of resource allocation ultimately compromises patient safety and the overall efficiency of the emergency medical system.
Performance Evaluation of KTAS: Reliability, Validity, and Accuracy
Inter-Rater Reliability and Consistency
The Korean Triage and Acuity Scale has been evaluated as a reliable triage tool, demonstrating consistent application across different assessors. Studies have shown a high level of agreement between KTAS scores determined by emergency nurses and those assigned by expert assessors. This inter-rater reliability was quantified by a weighted kappa coefficient of 0.83 and a Pearson's r coefficient of 0.88 (P<0.001), indicating an "excellent" level of agreement.
This high reliability is a critical indicator of the system's objectivity and consistency, suggesting that different trained individuals can apply the scale with similar results. This is foundational for any standardized system, as it ensures fairness in patient prioritization and consistency of care regardless of the individual performing the triage. An earlier study conducted before the widespread introduction of KTAS reported a lower agreement (weighted kappa = 0.39), which suggests that increased familiarity and comprehensive training with KTAS have significantly improved consistency in its application over time. The robust inter-rater reliability of KTAS thus underscores its strength as a standardized tool, minimizing subjective variability in patient classification and building a reliable national dataset for emergency medical services.
Predictive Validity of KTAS
Predictive validity assesses the extent to which KTAS accurately forecasts actual patient outcomes or resource utilization, a crucial measure of its real-world effectiveness. A study analyzing KTAS disposition and emergency treatment results for patients aged 15 and older in 2016 revealed that KTAS generally exhibits high sensitivity but low specificity.
To illustrate these findings, consider two scenarios:
Scenario 1: Death in the emergency department (ED) or intensive care unit (ICU) admission considered an emergency
Sensitivity: 0.916. This indicates that 91.6% of patients who either died in the ED or were admitted to the ICU were correctly classified by KTAS as emergency levels 1, 2, or 3. This high sensitivity is a strong point, demonstrating the system's effectiveness in identifying truly critical patients and minimizing undertriage.
Specificity: 0.581. This means that only 58.1% of patients who were admitted to a general ward or discharged were correctly classified as non-emergency levels 4 or 5 by KTAS. This low specificity suggests a tendency to over-classify less severe patients.
Positive Predictive Value (PPV): 0.097. Among patients classified by KTAS as emergency levels 1, 2, or 3, only 9.7% were ultimately admitted to the ICU or died in the ED. A low PPV points to a high rate of overtriage, meaning many non-urgent patients are classified as urgent.
Negative Predictive Value (NPV): 0.993. Among patients classified as non-emergency levels 4 or 5 by KTAS, an impressive 99.3% were indeed admitted to a general ward or discharged. This excellent NPV provides strong reassurance that patients classified as non-urgent are highly unlikely to be critically ill.
Scenario 2: Death in the ED, ICU, or non-ICU admission considered an emergency
Sensitivity: 0.700.
Specificity: 0.642.
Positive Predictive Value: 0.391.
Negative Predictive Value: 0.867.
The predictive validity profile of KTAS reveals a strategic bias towards patient safety (high sensitivity, high NPV) at the cost of efficiency (low specificity, low PPV). While this design minimizes the risk of missing critical patients, it simultaneously contributes to resource strain and overcrowding due to overtriage. This highlights a fundamental trade-off in triage system design and underscores the need for continuous refinement to balance safety with resource optimization.
Analysis of Mistriage: Under-triage and Over-triage
Mistriage, defined as the incorrect classification of a patient's severity, can manifest as either undertriage (underestimation of severity) or overtriage (overestimation of severity). In one study involving 1267 patient records, 186 cases (14.7%) exhibited disagreement between nurse-assigned KTAS scores and expert assessments. Of these, 131 cases were undertriage, and 55 were overtriage, suggesting that undertriage was a more frequent issue in that specific dataset. The consequences of mistriage are significant, potentially extending ED length of stay and increasing patient mortality. Overtriage, specifically, leads to increased resource consumption, exacerbates ED overcrowding, and prolongs LOS.
Common Causes of Triage Errors:
Incorrect Application of Pain Scale (NRS): The most frequent reason for triage error, accounting for 34.4% of all disagreement cases, was the misapplication of the numerical rating scale (NRS) for pain when using KTAS criteria. For example, a nurse might assign a KTAS score of 2 to a patient with acute abdominal pain and a pain score of 6, even when experts deemed KTAS 3 more appropriate given stable vital signs. This highlights a critical interface between subjective patient experience and objective triage criteria. While designed for objectivity, KTAS's reliance on subjective inputs like pain, without sufficiently standardized interpretation guidelines, can lead to significant classification errors.
Misjudgment of Physical Symptoms: The second most frequent cause of mistriage, observed in 25.3% of cases, was the misjudgment of physical symptoms related to the chief complaint. An illustrative example involves a 70-year-old male presenting with diffuse chest discomfort, diaphoresis, and nausea, but stable vital signs. A triage nurse might assign a KTAS 3, while experts would assess this as undertriage and assign a KTAS 2 due to the high suspicion of cardiogenic disease.
The identified causes of mistriage, particularly the misapplication of the pain scale, reveal a systemic vulnerability in KTAS. This underscores the need for enhanced training or revised criteria for such subjective modifiers to minimize interpretive variability and improve overall accuracy.
Performance in Specific Patient Populations
Considerations and Challenges for Elderly Patients
The increasing elderly population visiting emergency departments presents unique challenges for triage systems. Currently, KTAS applies the same classification system to elderly patients as it does to adults, without specific considerations for the physiological and pathological complexities often seen in older adults.
Studies indicate that KTAS is less accurately associated with severity in the elderly compared to adults. The proportion of KTAS "up-triage," where patients are classified as more severe than their actual condition warrants, was significantly higher in the elderly group (1.9% vs. 1.2%,
P<0.001). Furthermore, the predictive ability (measured by Area Under the Receiver Operating Characteristic curve - AUROC) for critical outcomes such as overall admission rate, ICU admission, and in-hospital mortality was consistently lower in the elderly group compared to adults. For instance, the AUROC for overall admission was 0.667 in the elderly versus 0.686 in adults; for ICU admission, it was 0.767 versus 0.842; and for in-hospital mortality, it was 0.711 versus 0.809. Old age was identified as the most influential independent factor contributing to up-triage.
This diminished predictive accuracy and higher rate of up-triage in elderly patients reveal a significant systemic inadequacy within KTAS for this vulnerable demographic. This is likely due to the complex interplay of physical frailty, multiple chronic comorbidities, and atypical symptom presentation common in older adults, which the current adult-centric KTAS may not adequately capture. The findings highlight an urgent unmet need for specialized geriatric triage guidelines to ensure optimal care for this rapidly growing patient population.
Pediatric Patient Guidelines and the Revised KTAS (rKTAS) for Enhanced Prediction
KTAS acknowledges the unique nature of pediatric patients by including 165 main complaints specifically for children, with pediatric KTAS applied to individuals under 15 years old. The Korean Society of Emergency Medicine developed KTAS with the aim of improving patient safety through real-time severity scoring for pediatric patients.
A "revised Korean Triage and Acuity Scale (rKTAS)" has been developed and studied specifically for pediatric patients, demonstrating a proactive approach to refining KTAS for this population. The rKTAS "elevates acuity by one step in relation to anticipated resource utilization". Studies comparing rKTAS with the conventional KTAS for pediatric patients, such as those presenting with gastrointestinal tract foreign bodies, have shown that rKTAS possesses higher accuracy in predicting critical clinical outcomes. These outcomes include overall hospitalization, admission to general wards, ICU admission, and operating room use. Furthermore, the emergency department length of stay (EDLOS) with rKTAS was observed to be longer at levels 1, 2, and 3 compared to the conventional KTAS.
The development and study of rKTAS for pediatric patients signify a move towards "precision triage." This approach acknowledges that pediatric patients can deteriorate rapidly and often require more intensive resources for stabilization or definitive care, even if their initial presentation seems less severe. By elevating acuity based on anticipated resource utilization, rKTAS implicitly factors in the higher threshold for intervention or specialized care needed for children. This proactive, resource-sensitive approach in pediatric triage is crucial for optimizing outcomes in a population with unique physiological vulnerabilities and highlights a valuable model for future KTAS refinements for other complex patient groups.
Challenges and Limitations in KTAS Application
Impact of Public Health Crises: Lessons from the COVID-19 Pandemic
The COVID-19 pandemic presented significant challenges to the accuracy and application of KTAS, exposing critical systemic vulnerabilities in its design when confronted with novel, rapidly evolving public health crises.
A primary limitation observed was the low accuracy of KTAS in predicting critical conditions, particularly for patients presenting with symptoms related to viral diseases. Triage became exceptionally challenging due to the convergence of individuals with COVID-19 symptoms, asymptomatic individuals with a history of contact, and those with other unrelated medical issues. This often led to
overclassification, partly because of the lack of standardized objective evaluation methods for viral infection symptoms and risk factors. The existing KTAS classifications, optimized for routine emergencies,
lacked the specialization required to accurately differentiate and classify patients exhibiting generalized flu-like symptoms, which could mask underlying critical conditions or lead to unnecessary resource consumption.
Challenges extended to both patients and medical staff. Patients often struggled to objectively evaluate their own symptoms due to inaccurate medical knowledge about emerging viral diseases, while medical staff faced difficulties in performing appropriate evaluations given the imperative for quick classification in short timeframes. Operationally, during the COVID-19 pandemic, emergency department patient numbers decreased, notably for KTAS level 3 patients (a 39.7% decrease). However, paradoxically, deaths increased by 23.0%, while admissions and discharges decreased. Initial wait times increased for KTAS level 2 patients (by 5.69
± 4.60 minutes), and the length of ED stay significantly increased for KTAS levels 1, 2, and 3. This suggests that even with fewer overall patients, the system struggled to efficiently process and treat the most critical cases, possibly due to resource strain, altered pathways, or the inherent difficulty in distinguishing true severity in a pandemic context. The experience highlights the urgent need for dynamic, adaptable triage protocols for future pandemics that can rapidly integrate new epidemiological data and specific clinical criteria for novel pathogens.
Discrepancies and Lack of Linkage with Pre-hospital Triage Systems
South Korea's emergency medical system is characterized by a bifurcation into pre-hospital (focused on structure and transport) and hospital (focused on treatment) stages, which largely operate independently. KTAS is primarily designed as a hospital-stage tool. While a pre-hospital KTAS has been developed in cooperation with the Ministry of Health and Welfare and the National Fire Agency, a significant systemic inefficiency exists: the pre-hospital triage system (e.g., the 119 emergency medical triage system) and the in-hospital KTAS are not linked and do not seamlessly share or interoperate information.
This lack of linkage represents a critical patient safety gap. Pre-hospital triage often undertriages patients compared to KTAS, possibly because it relies on a more limited set of factors such as main symptoms, level of consciousness, and general symptoms, whereas KTAS incorporates a broader array of considerations including vital signs, pain, and injury severity. Without seamless information transfer and consistent acuity assessment from the field to the ED, emergency departments cannot anticipate patient arrivals with accurate acuity levels. This hinders proactive resource preparation and can lead to delayed or inappropriate care upon arrival, as patients might arrive with a lower perceived urgency than their actual condition warrants. This creates a "handoff" problem where critical information or consistent assessment is lost, potentially compromising patient outcomes and contributing to ED bottlenecks. Bridging this gap through a unified registry and interoperable systems is essential to ensure a continuous, accurate assessment of patient acuity from the point of contact to definitive care, thereby enhancing patient safety and optimizing resource flow across the entire emergency care continuum.
Subjectivity in Assessment and the Need for Continuous Training
As previously discussed, subjective elements within the KTAS assessment process, such as a patient's self-reported pain level or subjective evaluations made by medical staff (e.g., assessing breathing effort or paleness) , can introduce variability into KTAS results. This variability is often dependent on the individual classifier's experience and ability.
This persistent challenge underscores that even a structured algorithm cannot entirely eliminate the need for nuanced human judgment. While KTAS provides a framework, clinical assessment frequently involves interpreting subtle cues that are not purely quantitative. This interpretive layer introduces subjectivity. To mitigate this, the system requires specialized personnel, significant time, and financial investment for initial training, as well as periodical training to maintain qualifications and ensure consistent application. Without continuous, high-quality training, inter-rater reliability can degrade, leading to inconsistent triage decisions. This highlights that the "human element" remains a critical variable in the system's performance, necessitating ongoing investment in professional development alongside efforts to refine the scale's objective criteria.
Addressing Emergency Department Overcrowding and Resource Strain
Emergency department overcrowding is a complex, multifaceted problem that arises when the demand for emergency services consistently exceeds available capacity, whether in terms of beds, staffing, or other resources. This phenomenon leads to a cascade of negative consequences, including delays in care, prolonged patient waiting times, patient dissatisfaction, increased medical errors, higher mortality rates, and a greater number of patients leaving without being seen.
While the implementation of KTAS has demonstrably improved internal ED efficiency by reducing patient length of stay and enhancing patient flow, the persistence of overcrowding indicates that KTAS alone cannot fully resolve this systemic issue. Overcrowding is often a symptom of broader systemic problems within the healthcare system, extending beyond the ED walls. These include hospital capacity limitations, particularly a shortage of inpatient beds, misaligned payment incentives, and overall insufficient staffing. Overtriage, a known issue with KTAS, further exacerbates overcrowding by directing non-urgent patients to resource-intensive pathways, consuming valuable resources that could otherwise be allocated to genuinely critical cases.
Therefore, while KTAS serves as an essential internal optimization tool for EDs, its impact on overcrowding is limited by these external systemic factors. True alleviation of ED overcrowding necessitates a multi-pronged approach that addresses hospital-wide patient flow management, appropriate utilization of emergency services, and continuous refinement of triage accuracy to minimize overtriage.
Future Directions and Innovations for KTAS
Proposed Revisions and Adaptations
Enhancing Criteria for Specific Symptoms and Conditions
Continuous review and revision of the KTAS level classification are imperative to address persistent errors of under- and over-triage. This involves a dynamic process of adapting the scale to evolving medical knowledge and patient presentation patterns. A critical area for refinement is the establishment of more precise and reasonable criteria for the symptoms and signs of specific viruses, particularly in light of lessons learned from the COVID-19 pandemic. Such enhanced criteria are necessary to enable medical personnel to perform consistent evaluations during future public health crises.
Furthermore, a re-evaluation of how pain is considered within the KTAS framework is essential to reduce the overestimation of patient severity. The subjective nature of pain, when given significant weight, can distort the overall acuity assessment, leading to inefficient resource allocation. KTAS is fundamentally based on the "severity of symptoms, vital signs, and chief complaint". Future revisions should actively incorporate diverse clinical experiences and the latest academic research findings to ensure the scale reflects medical reality as accurately as possible.
Developing Specialized Triage for Vulnerable Populations (e.g., Frailty Scale Integration for Elderly)
The current KTAS has shown limitations in its ability to accurately assess severity in elderly patients, with a higher propensity for up-triage. This highlights a critical need for specialized triage approaches for vulnerable populations. A promising direction involves integrating tools like the Clinical Frailty Scale (CFS) with KTAS. Studies have demonstrated that combining CFS with KTAS significantly improves predictive performance for older patients regarding critical outcomes such as hospital admission, ICU admission, and in-hospital mortality. The CFS is a simple, time-efficient, and validated tool for prehospital frailty screening in patients over 65 years of age.
This move towards integrating frailty scales for the elderly signifies a shift towards "precision triage." It acknowledges that a one-size-fits-all approach is insufficient for diverse patient populations and that future iterations of KTAS must incorporate more nuanced, population-specific criteria to enhance accuracy and optimize care. This adaptive refinement, exemplified by the integration of frailty scales, will be crucial for maintaining the scale's relevance and effectiveness in an increasingly complex healthcare landscape.
Leveraging Artificial Intelligence (AI) and Natural Language Processing (NLP)
Traditional triage systems, heavily reliant on human judgment, are susceptible to issues such as under-triage, over-triage, inherent variability, and bias. Emerging research suggests that Machine Learning (ML) and Natural Language Processing (NLP) can significantly enhance triage accuracy and consistency, offering a pathway to overcome these limitations.
AI-Powered Severity Classification from Clinical Conversations
AI-powered NLP algorithms demonstrate considerable potential in analyzing real-world conversations between medical staff and patients to automatically categorize patient severity. Models such as Support Vector Machine (SVM) and Multilayer Perceptron (MLP) have shown promising performance, with AUROC values of 0.764 and 0.759 respectively, in classifying severity even from "nonsensical and confused content" within clinical dialogues. This innovative approach holds the potential to alleviate emergency department overcrowding, reduce patient waiting times, and facilitate more fluid and immediate responses to urgent situations by providing rapid, objective severity assessments.
Federated Learning for Collaborative Triage Enhancement Across Institutions
Federated learning (FL) offers a transformative approach to enhancing triage accuracy by enabling collaborative AI model training across multiple healthcare institutions while rigorously maintaining patient privacy. This method is particularly effective in overcoming data disparities that often exist between different emergency medical centers. A clinical support system based on FL for KTAS has been developed to predict the need for KTAS revision during a patient's ED stay, with the explicit goal of preventing both under-triage and over-triage. This model has demonstrated improved outcomes, especially in Local Emergency Medical Institutions (LEMIs) that typically operate with fewer resources, showcasing its capacity to democratize advanced triage capabilities.
Clinical Decision Support Systems (CDSS) for Improved Triage Accuracy
The integration of AI algorithms into Clinical Decision Support Systems (CDSS) is proving to be a powerful tool for improving triage accuracy. These systems can integrate multiple data modalities, including vital signs, laboratory results, and imaging findings from electronic health records, to predict clinical deterioration more effectively than traditional logistic regression models. Such AI-powered CDSS can significantly reduce variability in risk classification compared to human nursing assessments, decrease instances of both under- and over-triage, and provide enhanced clinical decision-making support to emergency department staff. The ability of these systems to capture complex patterns in large volumes of structured and unstructured data, especially through advanced approaches like Deep Neural Networks (DNNs), positions them as a key innovation for the future of KTAS, leading to more precise and consistent patient prioritization.
Conclusions
The Korean Triage and Acuity Scale (KTAS) has undeniably transformed emergency medical services in South Korea since its national implementation in 2016. Its development, rooted in the internationally recognized Canadian Triage and Acuity Scale (CTAS) and driven by the Korean Society of Emergency Medicine (KSEM), addressed a critical need for standardization in a previously fragmented system. The adoption of KTAS has yielded tangible benefits, including a significant reduction in emergency department length of stay and, most importantly, a decrease in overall patient mortality rates. This demonstrates its profound impact on patient safety and operational efficiency. The scale's high inter-rater reliability further validates its consistent application by trained medical professionals.
However, the analysis also reveals inherent complexities and limitations that warrant continuous attention and refinement. The persistent issue of mistriage, particularly influenced by the subjective interpretation of pain, highlights a fundamental challenge in balancing comprehensive patient assessment with objective classification. The performance of KTAS during the COVID-19 pandemic underscored its limitations in adapting to novel infectious disease outbreaks, revealing a need for more dynamic and specialized criteria for such crises. Furthermore, the diminished predictive accuracy for elderly patients points to an unmet need for geriatric-specific triage considerations. A critical policy concern remains the administrative reconfiguration of the 5-level KTAS into a simplified 3-level or 2-level system for payment purposes. This administrative simplification, contrary to global best practices, risks undermining the nuanced clinical utility and predictive power that the original 5-level scale was designed to provide.
Moving forward, the evolution of KTAS must embrace a multi-faceted approach. This includes continuous refinement of its clinical criteria, particularly for subjective elements like pain and for specific vulnerable populations such as the elderly, potentially through the integration of specialized assessment tools like frailty scales. Crucially, leveraging advanced technologies such as artificial intelligence (AI) and natural language processing (NLP) offers a promising pathway to enhance triage accuracy, consistency, and efficiency. AI-powered systems, including those utilizing federated learning and clinical decision support, can analyze complex data, including clinical conversations, to provide more objective severity classifications, thereby alleviating overcrowding and optimizing resource allocation.
Ultimately, the future success of KTAS hinges on a sustained commitment to evidence-based adaptation, technological integration, and a clear prioritization of clinical efficacy over administrative convenience. Such a comprehensive strategy will ensure that KTAS continues to serve as a robust and adaptable tool, capable of meeting the evolving demands of emergency medicine and safeguarding patient outcomes in South Korea.