Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2025 Jul 6;57(5):860–873. doi: 10.1111/jnu.70033

Factors Influencing Mistriage Based on the Korean Triage and Acuity Scale: A Retrospective Cross‐Sectional Study

Nayeon Yi 1,2, Dain Baik 1,3,✉
PMCID: PMC12420892  PMID: 40619607

ABSTRACT

Introduction

Mistriage is important because of its potential for serious consequences, notwithstanding the beneficial effects of the emergency patient classification system employed to alleviate overcrowding in emergency departments (EDs). This study aimed to assess mistriage using the Korean Triage and Acuity Scale (KTAS) and identify factors influencing it.

Design

Retrospective cross‐sectional study.

Methods

We examined the factors influencing mistriage in the KTAS and rates of under‐ and over‐triage. Participants were obtained by combining electronic health records with registry data from the National Emergency Department Information System. We assessed the eligibility of patients aged ≥ 15 years who visited the ED between July 1, 2022, and June 30, 2023. Using the KTAS classification criterion, two experienced experts determined the final acuity level. We employed multivariate logistic regression analysis to evaluate the factors that predict under‐ and over‐triage.

Results

Of 53,947 ED encounters, 1110 participants were enrolled in this study. Mistriage occurred in 207 (18.6%) patients: 88 (7.9%) had under‐triage, and 119 (10.7%) had over‐triage. In adjusted analyses, under‐triage was associated with lower mean arterial pressure (odds ratio [OR], 5.42; 95% confidence interval [CI], 1.45–20.32) and presenting complaints of immunity or fever (OR, 3.41; 95% CI, 1.38–8.45), while over‐triage was associated with advanced age (OR, 0.52; 95% CI, 0.28–0.98), pain (OR, 1.96; 95% CI, 1.18–3.25), lower KTAS experience (OR, 1.95; 95% CI, 1.08–3.51), and several specific present complaints.

Conclusions

By improving mistriage, the quality of emergency medical services may be enhanced through reduced costs, increased operational efficiency, and improved patient safety and satisfaction. Implementation of standardized criteria, validated triage tools, and enhanced provider training is crucial for achieving more accurate emergency triage. Additionally, establishing regulatory and financial incentives and developing realistic standards for mistriage management will optimize triage processes and ensure prompt, prioritized care.

Keywords: classification; emergency nursing; emergency service, hospital; patient acuity; patient safety; triage

1. Introduction

The gradual increase in the number of emergency department (ED) visits and subsequent overcrowding (National Emergency Medical Center [NEMC] 2024) has emphasized the need for accurate and rapid patient classification systems to effectively determine urgent medical needs (Park and Lim 2017). In response to this necessity, emergency triage systems identify patients requiring urgent medical intervention and efficiently allocate limited resources to securely assign individuals to appropriate care (Emergency Nurses Association [ENA] 2023; Park and Lim 2017). In EDs, triage is predominantly performed by nurses worldwide (Hinson et al. 2019), with over 85% of triage procedures carried out by nurses in South Korea (Jung and Yi 2024). Within the field of emergency nursing, triage represents a distinct professional responsibility that enables nurses to exercise clinical autonomy and deliver comprehensive, patient‐centered care (Solheim 2016).

The Korean Triage and Acuity Scale (KTAS), developed by the Korean Society of Emergency Medicine in 2012 and adapted from the Canadian Triage and Acuity Scale, which utilizes patient conditions and clinical characteristics to determine triage levels, was tailored to meet the specific needs of South Korea (Mirhaghi et al. 2015; Park and Lim 2017). From 2016 onward, all emergency medical centers across South Korea mandated KTAS, a nationally integrated classification system (Park and Lim 2017). Comprising five tiers—Level 1 (resuscitation), Level 2 (emergent), Level 3 (urgent), Level 4 (less urgent), and Level 5 (nonurgent)—the KTAS evaluates the severity and urgency of a patient's condition using various algorithms. These assessments help prioritize care, determine the level of emergency, and estimate safe waiting times, ensuring a structured approach to patient management (Park and Lim 2017).

The adoption of KTAS has led to decreased length of stay (LOS) and mortality in the ED (Kwon et al. 2019). Nonetheless, KTAS exhibited low specificity in comparison to its relatively high sensitivity (Choi et al. 2019), and false negatives, such as the hospitalization of noncritical patients categorized into KTAS Level 4 or 5, were associated with mistriage (Jang and Seo 2024; Lim et al. 2018; Oh et al. 2018). Mistriage includes under‐triage, where the classification is lower than the appropriate level, and over‐triage, where the classification is greater than the appropriate level. ED LOS was identified as a predictor of under‐triage (Chung et al. 2023). Additionally, compared to having smaller classification differences, a difference of two or more levels was observed in under‐triage, which was associated with significantly higher rates of intensive care unit admissions, mortality, and inter‐hospital transfers (Lee and Oh 2021).

Accurate triage and appropriate resource allocation contribute to reducing waiting times and medical costs (Corkery et al. 2021), while mitigating the burden caused by mistriage. However, the over‐triage of nonemergency patients results in an inappropriate resource allocation and reduces the availability of beds for those who are critically ill (Gilboy et al. 2020). In particular, under‐triage, which delays treatment, poses potential risks to patient safety, and can critically impact patient survival (Ausserhofer et al. 2020). Hence, evaluating mistriage for accurate classification is essential to enhance ED access equity, patient safety, and the quality of emergency medical care. This need warrants further emphasis in South Korea's healthcare crisis context, involving the mass resignation of medical residents (Park et al. 2024).

Previous research on mistriage using the emergency patient classification tool showed that the rate of under‐triage for adults using the Emergency Severity Index (ESI) was 3.3%–4.9%, and the rate of over‐triage was 26.0%–28.9% (Huabbangyang et al. 2023; Sax et al. 2023). Factors contributing to under‐triage include younger age, high‐risk medication use, intensive care utilization, and high comorbidity burden, while factors contributing to over‐triage include trauma and evening visits (Huabbangyang et al. 2023; Sax et al. 2023). A study based on the Manchester Triage System discovered a correlation between advanced age at acute myocardial infarction diagnosis and age (Nishi et al. 2018).

Conversely, a study using the KTAS found under‐triage to be 10.3% higher than that of ESI. Mistriage was attributed to the inappropriate utilization of the pain scale score and erroneous assessment of physical symptoms, as determined by expert opinion (Moon et al. 2019). Additionally, under‐triage was observed in patients reporting cardiac and acute abdominal pain. Under‐triage was observed to be higher in males, older adults, and during direct visits (Choi and Lee 2023; Oh and Kim 2021) but lower when patients used an ambulance or presented with chest pain and a history of cardiovascular issues (Choi and Lee 2023). Previous researchers have investigated the mistriage and predictability of the KTAS, focusing on patient‐reported pain. This study revealed that over‐triaging occurred when pain was considered during KTAS classification. This over‐triage hampers the predictive performance of the KTAS in determining patient urgency (Lee et al. 2019) (Table S1).

In summary, studies on KTAS mistriage undertaken using the emergency patient classification tool found a higher incidence of under‐triage than other scales. Moreover, these studies showed that essential clinical data and triage provider characteristics were not properly considered, with a predominant focus on specific health concerns (Choi and Lee 2023; Oh and Kim 2021). Additionally, mistriage factor analysis studies either deal with a single aspect of under‐triage (Choi and Lee 2023; Oh and Kim 2021) or rely on expert opinions to determine the cause of mistriage (Moon et al. 2019). Few studies have quantified and statistically explored the relevant factors in our review.

Hence, this study aimed to examine mistriage in adult patients in the ED without restricting it to specific diseases or symptoms using the KTAS and to statistically assess the factors influencing mistriage, including clinical data, triage provider characteristics, and the ED context. The goal was to generate implications for improving the classification system, patient safety, and quality of emergency medical services.

2. Design

This retrospective cross‐sectional study measured mistriage and examined factors influencing under‐triage and over‐triage using the KTAS. It was conducted in accordance with the principles of the Declaration of Helsinki and followed the Strengthening the Reporting of Observational Studies in Epidemiology statement, which serves as standard reporting guidelines for epidemiological observational studies (Table S2).

2.1. Accuracy of the KTAS

Although the present study focused on mistriage patterns, it is important to consider the accuracy of the KTAS in light of previous research. The validity of the KTAS has been demonstrated through its predictive accuracy for outcomes such as hospital admission, intensive care unit admission, in‐hospital mortality, and 30‐day mortality (Chung et al. 2023; Lee et al. 2018; Lim et al. 2020). Reliability has been primarily evaluated through inter‐rater agreement studies, with findings indicating moderate to good consistency among trained raters (Kim et al. 2017, 2019; Park et al. 2019). Nevertheless, research on reliability has been more limited in scope compared to that on validity (Chung et al. 2023; Kim et al. 2017, 2019; Lee et al. 2018; Lim et al. 2020; Park and Lim 2017; Park et al. 2019). Despite these findings, the potential for mistriage remains due to tool limitations, evaluator judgment variability, and complex patient presentations. Thus, ongoing efforts to refine the KTAS are essential to enhance its accuracy and clinical applicability (Chung et al. 2023; Lim et al. 2020; Park et al. 2019; Zhang et al. 2024).

3. Methods

3.1. Study Setting and Participants

This study was conducted at an urban academic medical center in Seoul with over 800 beds. The emergency medical center has 25 beds in its ED and records an annual census of > 52,000 patients as of 2022.

The study included adult patients aged ≥ 15 who sought medical attention at the ED between July 2022 and June 2023. The KTAS classification system and levels categorize patients into adult and pediatric groups, with 15 years as the cutoff (Park and Lim 2017).

Consequently, participants over 15 years old were classified as adults. We excluded patients who voluntarily left the hospital without following medical advice or being seen, individuals whose reason for visiting the hospital was nonmedical, and patients with missing data for key variables. This study used data registered with the National Emergency Department Information System (NEDIS), including electronic health records, to analyze the patients' information who met the specified inclusion criteria (Figure 1).

FIGURE 1.

FIGURE 1

Open in a new tab

Flow diagram of study inclusion.

3.2. Sample and Sampling Process

We utilized the 2k 2–16k 2 formula, as employed in previous studies, to determine the sample size required to validate the under‐ and over‐triage of the KTAS with weighted kappa (Cicchetti 1981), where k represents the number of categories at the ordinal level. In this study, five KTAS levels were assigned k values. Thus, 50–400 samples were required for the weighted kappa analysis.

Logistic regression analysis was conducted to identify the factors influencing mistriage concerning KTAS. To achieve stability in this analysis, we determined the number of samples needed by setting five events per variable based on previous research (Peduzzi et al. 1996; Vittinghoff and McCulloch 2007). Applying five events per 26 predictors discovered in prior research generated 130 results. A previous study utilizing the KTAS reported a mean rate of 14.7% (Moon et al. 2019). Inserting these values into the equation for the minimum sample size in the logistic regression analysis yielded 884.35, calculated as 5 × 26 × (100/14.7). Considering a 20% dropout rate, we deemed having a minimum of 1105.44 samples necessary.

This study determined the percentages of the five KTAS levels among patients who sought medical attention in the ED between July 2022 and June 2023. We acquired 1110 samples using stratified random sampling with IBM SPSS Statistics version 29.0 (IBM Corporation, Armonk, NY, USA) based on the proportion criterion (Figure 1).

3.3. Triage System

3.3.1. KTAS Provider Qualification

Triage providers, including physicians, nurses, and paramedics, must have at least 1 year of experience in an ED setting and complete the relevant coursework and exams. On a triennial basis, participation in a reeducation program with tests is required for qualification renewal (Kim et al. 2022; Ministry of Government Legislation 2025).

3.3.2. KTAS Classification Process

Triage providers rapidly identify patients with high acuity (KTAS 1 or 2) who arrive at the ED with life‐threatening diseases or injuries. A critical first‐look evaluation, lasting 3–5 s, achieves this and promptly assigns these patients to the resuscitation or treatment area. Additionally, patients undergo primary screening for infectious diseases and are subsequently placed in isolation zones if necessary. For most patients in relatively stable conditions, the triage provider assesses symptoms of the present primary concern to determine the triage level and applies additional modifiers for selected primary symptoms. The first modifiers to be evaluated are vital signs, such as respiratory distress, hemodynamic status, level of consciousness, and temperature, as well as other factors, including pain severity, bleeding disorder, and mechanism of injury. The second modifier complements the first modifier, particularly for specific complaints such as blood glucose levels and dehydration. Applying the modifiers resulted in assigning suitable triage levels and treatment areas. Continually assessing patients in the waiting area based on the classification criteria is necessary because waiting without assigning a treatment area may worsen their symptoms (Figure 2).

FIGURE 2.

FIGURE 2

Open in a new tab

Triage system of KTAS. KTAS, Korean Triage and Acuity Scale.

3.4. Variables

Among the predictors, sex and age were identified as patient demographic characteristics, whereas ED context characteristics included arrival time, visit route, mode of transportation, and LOS. The triage providers' characteristics included their KTAS classification career and clinical experience. Additionally, we recorded the type of illness, medical history, level of consciousness, mean arterial pressure (MAP), systemic inflammatory response syndrome (SIRS), saturation of peripheral oxygen (sPO2), pain, and presenting complaints.

SIRS is defined by the presence of two or more of the following: temperature > 38 or < 36°C; heart rate > 90 beats/min; respiratory rate > 20 breaths/min or arterial carbon dioxide (PaCO2) < 32 mmHg; white blood cell count > 12000/mm3 or < 4000/mm3 or > 10% immature neutrophils (Levy et al. 2003). The SIRS score was determined based on criteria obtained after the initial assessment of the patient's vital signs at triage.

3.5. Data Collection

After receiving Institutional Review Board (IRB) approval, we removed and encoded patient‐ and nurse‐identifying information from the data extracted from the electronic health records and NEDIS registry data during the study period. These data were collected to obtain information regarding the following variables: sex, age, arrival time, visit route, arrival mode of transportation, discharge time, license number, certification number of the KTAS provider, illness type, medical history, level of consciousness, vital signs, sPO2, pain, presenting complaint, initial KTAS level, and other medical records. We conducted a review to identify the inconsistencies and outliers in the data.

3.6. Final Triage Level Assignment Process

The final classification level was determined using the following approach (Figure 3).

FIGURE 3.

FIGURE 3

Open in a new tab

Final triage level assignment process.

3.6.1. Selection of the Triage Experts

The selection criteria for the two triage experts included having over 10 years of clinical experience in the ED, being a certified current KTAS provider, having > 7 years of experience as a KTAS provider, and having a PhD registration. Additionally, they completed additional training to uphold their knowledge and skills for the KTAS classification and possessed advanced cardiovascular life support certifications. These criteria were based on the competencies of emergency nurse practitioners (American Academy of Emergency Nurse Practitioners and ENA 2021) and experts in clinical ladder models (Benner 1984; Cho et al. 2017). These triage experts were not part of the research team. This arrangement was intentionally made to ensure an objective assessment of the KTAS scores.

3.6.2. Inter‐Rater Agreement

Before the final classification‐level evaluation, experts utilized 10 datasets of individual medical records collected using a structured form to determine the triage level independently. We identified inter‐rater agreements. The experts established reliability, resulting in a high degree of agreement (Landis and Koch 1977) with a weighted kappa coefficient of 0.898 (95% confidence intervals [CI], 0.701–1.000).

3.6.3. Expert Assessment for Assigning the Triage Level

The two triage experts independently conducted a systematic assessment of 1110 individual data points gathered by the researcher to determine the acuity levels in different offices.

3.6.4. Conflict Resolution

When there was inconsistency among experts, they consulted an emergency medicine specialist who is also a KTAS instructor, reaching a consensus to resolve conflicting evaluation results.

3.6.5. Determining the Final Triage Acuity Level

The final level of triage was established upon attaining consensus between experts.

3.7. Outcomes

The primary outcomes were under‐triage and over‐triage rates, which indicate mistriage. Under‐triage refers to when the expert's final classification level is higher than the initial level assigned by the triage provider, whereas over‐triage refers to when the expert's final classification level is lower. The secondary outcome was to identify the factors associated with under‐ and over‐triage.

3.8. Statistical Analysis

The data were analyzed using R version 4.3.1 (R Foundation for Statistical Computing, Vienna, Austria). The participants' demographic and clinical characteristics, as well as characteristics of ED contexts and triage providers, were verified using frequencies, percentages, means, and standard deviations. Owing to the negative skewness of the sPO2 and LOS distributions, which indicated a higher concentration of values toward the right side, the median and interquartile ranges (IQR) were calculated.

We employed crosstables and weighted kappa statistics to assess the degree of under‐ and over‐triage by KTAS level and compare triage providers with experts. Additionally, chi‐square or Fisher's exact tests were conducted to compare differences in under‐ and over‐triage based on the demographic and clinical characteristics of the participants, ED context characteristics, and triage provider characteristics. Univariate and multivariate logistic regression analyses were performed to identify factors influencing mistriage. Under‐ and over‐triage were compared with accurate triage. Among the 1110 participants, seven (0.63% of total data) had missing values for key variables crucial for logistic regression analysis. These seven cases were included in the descriptive statistics by categorizing the missing values into separate categories within the relevant variables. However, they were excluded from the logistic regression analysis to prevent potential bias in the regression estimates.

3.9. Ethical Approval

This retrospective study utilized electronic health records and NEDIS registry data. The IRB of the participating medical facilities approved the study and granted a waiver of informed consent (S**** 2023–10‐020).

4. Results

4.1. Characteristics of the Participant, ED Contexts, and Triage Provider

The study included 1110 individuals, with 59.7% being female, and the average age of the participants was 51.47 ± 20.94 years. The evening time (15:00–22:59) had the highest number of participants visiting the ED, accounting for 39.8% of all encounters, and 87.9% were direct visits. In total, 26.0% used ambulances for ED visits. The median ED LOS was 122.5 min (IQR, 68.0, 206.0) (Table 1).

TABLE 1.

Characteristics of participants, ED contexts, and triage providers in the triage groups.

Characteristics Overall Correct Triage Under‐Triage Over‐Triage
(N = 1110) (n = 903) (n = 88) (n = 119)
Sex
Male 447 (40.3) 352 (39.0) 45 (51.1) 50 (42.0)
Female 663 (59.7) 551 (61.0) 43 (48.9) 69 (58.0)
Age (years) 51.5 ± 20.9 52.1 ± 20.8 53.1 ± 23.4 45.2 ± 19.0
< 45 469 (42.3) 370 (41.0) 35 (39.8) 64 (53.8)
45–64 289 (26.0) 240 (26.6) 16 (18.2) 33 (27.7)
> 64 352 (31.7) 293 (32.4) 37 (42.0) 22 (18.5)
Arrival time
Day (07:00–14:59) 410 (36.9) 344 (38.1) 27 (30.7) 39 (32.8)
Evening (15:00–22:59) 442 (39.8) 351 (38.9) 39 (44.3) 52 (43.7)
Night (23:00–06:59) 258 (23.2) 208 (23.0) 22 (25.0) 28 (23.5)
ED visits
Direct 976 (87.9) 793 (87.8) 77 (87.5) 106 (89.1)
Referrals 134 (12.1) 110 (12.2) 11 (12.5) 13 (10.9)
Arrival mode of transportation
Ambulance 289 (26.0) 235 (26.0) 33 (37.5) 21 (17.6)
Others 821 (74.0) 668 (74.0) 55 (62.5) 98 (82.4)
ED‐LOS (min) 122.5 (68.0, 206.0) 122.0 (68.0, 207.0) 169.5 (88.0, 260.5) 107.0 (46.5, 146.0)
< 300 1013 (91.3) 823 (91.1) 75 (85.2) 115 (96.6)
≧ 300 97 (8.7) 80 (8.9) 13 (14.8) 4 (3.4)
KTAS experience (years)
< 2 409 (36.8) 322 (35.7) 30 (34.1) 57 (47.9)
2–3 425 (38.3) 351 (38.9) 30 (34.1) 44 (37.0)
> 3 276 (24.9) 230 (25.5) 28 (31.8) 18 (15.1)
Clinical nursing experience (years)
< 3 49 (4.4) 35 (3.9) 3 (3.4) 11 (9.2)
≧3 1061 (95.6) 868 (96.1) 85 (96.6) 108 (90.8)
Open in a new tab

Note: Data are reported as n (%) or mean ± SD or median (IQR).

Abbreviations: ED, emergency department; IQR, interquartile range; KTAS, The Korean Triage and Acuity Scale; LOS, length of stay; SD, standard deviation.

Nontrauma patients comprised the majority (82.3%) of the participants, and 46.1% had a medical history. At the ED visit, 96.3% of patients were conscious, and 1.2% had MAP below 65 mmHg. More than two SIRS criteria were matched in 11.9%, and the median value of sPO2 was 98% (IQR, 96, 99). At the time of the visit, 60.4% of patients experienced pain. The most common complaint was in the gastrointestinal tract (23.7%), whereas the least common complaint was in the respiratory system (6.1%) (Table 2).

TABLE 2.

Clinical characteristics of participants in the triage groups.

Characteristics Overall Correct Triage Under‐Triage Over‐Triage
(N = 1110) (n = 903) (n = 88) (n = 119)
Illness type
Non‐trauma 913 (82.3) 748 (82.8) 69 (78.4) 96 (80.7)
Trauma 197 (17.7) 155 (17.2) 19 (21.6) 23 (19.3)
Medical history
Yes 512 (46.1) 423 (46.8) 48 (54.5) 41 (34.5)
No 598 (53.9) 480 (53.2) 40 (45.5) 78 (65.5)
Level of consciousness
Alert 1069 (96.3) 869 (96.2) 82 (93.2) 118 (99.2)
Verbal response 22 (2.0) 16 (1.8) 5 (5.7) 1 (0.8)
Pain response 12 (1.1) 11 (1.2) 1 (1.1) 0 (0.0)
Unresponsiveness 7 (0.6) 7 (0.8) 0 (0.0) 0 (0.0)
MAP (mmHg) a
< 65 13 (1.2) 8 (0.9) 5 (5.7) 0 (0.0)
≧ 65 1090 (98.2) 888 (98.3.) 83 (94.3) 119 (100)
N/A 7 (0.6) 7 (0.8) 0 (0.0) 0 (0.0)
SIRS
0 646 (58.2) 535 (59.2) 40 (45.5) 71 (59.7)
1 325 (29.3) 261 (28.9) 21 (23.9) 43 (36.1)
2–3 132 (11.9) 100 (11.1) 27 (30.7) 5 (4.2)
N/A 7 (0.6) 7 (0.8) 0 (0.0) 0 (0.0)
sPO 2 (%) 98 (96, 99) 98 (96, 99) 97 (96, 99) 98 (97, 99)
< 92 12 (1.1) 9 (1.0) 2 (2.3) 1 (0.8)
92–94 15 (1.4) 14 (1.6) 1 (1.1) 0 (0.0)
> 94 214 (19.3) 168 (18.6) 22 (25.0) 24 (20.2)
Unmeasured 869 (78.3) 712 (78.8) 63 (71.6) 94 (79.0)
Impact of pain
Yes 670 (60.4) 528 (58.5) 50 (56.8) 92 (77.3)
No 440 (39.6) 375 (41.5) 38 (43.2) 27 (22.7)
Presenting complaint
Cardiovascular 105 (9.5) 82 (9.1) 11 (12.5) 12 (10.1)
Gastrointestinal 263 (23.7) 213 (23.6) 11 (12.5) 39 (32.8)
Immunity or fever 81 (7.3) 60 (6.6) 19 (21.6) 2 (1.7)
Musculoskeletal or skin 192 (17.3) 150 (16.6) 16 (18.2) 26 (21.8)
Neurological 208 (18.7) 187 (20.7) 14 (15.9) 7 (5.9)
Respiratory 68 (6.1) 50 (5.5) 10 (11.4) 8 (6.7)
Others 193 (17.4) 161 (17.8) 7 (8.0) 25 (21.0)
Open in a new tab

Note: Data are reported as n (%) or median (IQR).

Abbreviations: MAP, mean arterial pressure; N/A, not applicable; SIRS, systemic inflammatory response syndrome; sPO2, saturation of peripheral oxygen.

a

MAP was estimated from blood pressure (MAP = diastolic pressure plus one‐third of the difference between systolic and diastolic pressure).

Most (95.6%) triage providers for emergency patients had over 3 years of clinical nursing experience. However, a considerable proportion (75.1%) had < 3 years of experience as KTAS providers (Table 1).

4.2. Mistriage

4.2.1. Mistriage Rate

As triage providers, ED nurses correctly classified 903 (81.4%) of the 1110 participants, whereas 207 (18.6%) were misclassified. Under‐triage accounted for 88 (7.9%) of total data points, whereas over‐triage accounted for 119 (10.7%) (Tables 1 and 2).

4.2.2. Mistriage by KTAS Level

From the outcomes of mistriage by KTAS level, under‐triage was most prevalent in Level 4 (49 cases, 23.7% of all mistriaged encounters), whereas over‐triage was highest in Level 3 (86 cases, 41.5% of all mistriaged encounters) (Figure 4, Table 3).

FIGURE 4.

FIGURE 4

Open in a new tab

Mistriage by KTAS level. KTAS, Korean Triage and Acuity Scale.

TABLE 3.

Mistriage by KTAS level.

KTAS 1 KTAS 2 KTAS 3 KTAS 4 KTAS 5 Total
Triage provider level KTAS 1 9 0 0 0 0 9
KTAS 2 2 66 11 1 0 80
KTAS 3 0 32 474 74 12 592
KTAS 4 0 2 47 234 21 304
KTAS 5 0 0 2 3 120 125
Total 11 100 534 312 153 1110
Open in a new tab

Note: Weighted kappa = 0.771; 95% CI 0.741–0.802.

4.3. Multivariate Analysis of Factors Associated With Mistriage

Multivariate analysis revealed that the factors influencing mistriage included age, MAP, pain, presenting complaints, and classification experience (Table 4). Among them, we confirmed that under‐triage increased in the MAP of < 65 mmHg 5.42 times (95% CI, 1.45–20.32) compared to ≥ 65 mmHg, and the presenting complaint associated with immunity or fever increased 3.41 times (95% CI, 1.38–8.45) compared to neurological symptoms. Conversely, over‐triage caused a 48% decrease (95% CI, 0.28–0.98) among those aged ≥ 65 compared to those under 45. Thus, individuals under the age of 45 years are more likely to be over‐triaged than those aged ≥ 65 years. Patients with pain experienced an increase in over‐triage 1.96 times (95% CI, 1.18–3.25) compared to those without pain. Compared to those with > 3 years, triage providers with < 2 years of KTAS experienced an increase of 1.95 times (95% CI, 1.08–3.51). In comparison to neurological symptoms, the over‐triage for respiratory symptoms increased by 6.05 times (95% CI, 1.97–18.60), for cardiovascular symptoms by 3.78 times (95% CI, 1.40–10.17), for other symptoms by 3.74 times (95% CI, 1.53–9.16), for gastrointestinal symptoms by 3.63 times (95% CI, 1.54–8.56), and for musculoskeletal or skin symptoms by 3.10 times (95% CI, 1.27–7.57).

TABLE 4.

Multivariate analysis of factors associated with mistriage.

Variables Under‐Triage Over‐Triage
OR adj (95% CI) p OR adj (95% CI) p
Sex
Male 1 (Reference) 1 (Reference)
female 0.70 (0.44–1.12) 0.135 0.80 (0.53–1.21) 0.288
Age (years)
< 45 1 (Reference) 1 (Reference)
45–64 0.82 (0.42–1.61) 0.556 0.95 (0.59–1.55) 0.845
> 64 1.08 (0.55–2.13) 0.830 0.52 (0.28–0.98) 0.042*
Arrival mode of transportation
Others 1 (Reference) 1 (Reference)
Ambulance 1.40 (0.82–2.39) 0.215 0.91 (0.53–1.56) 0.733
Medical history
No 1 (Reference) 1 (Reference)
Yes 1.17 (0.63–2.18) 0.625 0.90 (0.54–1.48) 0.666
Level of consciousness
Alert 1 (Reference) 1 (Reference)
Verbal response 1.65 (0.48–5.59) 0.425 1.43 (0.16–12.53) 0.748
Pain response 0.63 (0.07–5.36) 0.675 0.985
MAP (mmHg) a
≧ 65 1 (Reference) 1 (Reference)
< 65 5.42 (1.45–20.32) 0.012* 0.987
SIRS
0 1 (Reference) 1 (Reference)
1 0.97 (0.54–1.73) 0.919 1.24 (0.81–1.90) 0.327
2–3 1.84 (0.91–3.74) 0.092 0.50 (0.18–1.39) 0.184
Impact of pain
No 1 (Reference) 1 (Reference)
Yes 1.29 (0.76–2.20) 0.351 1.96 (1.18–3.25) 0.010*
Presenting complaint
Neurologic 1 (Reference) 1 (Reference)
Cardiovascular 1.99 (0.83–4.77) 0.121 3.78 (1.40–10.17) 0.009**
Gastrointestinal 0.74 (0.31–1.78) 0.507 3.63 (1.54–8.56) 0.003**
Immunity or fever 3.41 (1.38–8.45) 0.008** 1.23 (0.23–6.63) 0.809
Musculoskeletal or skin 1.64 (0.72–3.73) 0.235 3.10 (1.27–7.57) 0.013*
Respiratory 2.02 (0.78–5.25) 0.150 6.05 (1.97–18.60) 0.002**
Others 0.84 (0.32–2.24) 0.731 3.74 (1.53–9.16) 0.004**
KTAS experience (years)
> 3 1 (Reference) 1 (Reference)
< 2 0.64 (0.35–1.16) 0.140 1.95 (1.08–3.51) 0.026*
2–3 0.63 (0.36–1.12) 0.119 1.59 (0.88–2.87) 0.125
Clinical nursing experience (years)
≧ 3 1 (Reference) 1 (Reference)
< 3 1.13 (0.31–4.08) 0.848 1.99 (0.91–4.34) 0.085
Open in a new tab

Abbreviations: CI, confidence interval; KTAS, the Korean Triage and Acuity Scale; MAP, mean arterial pressure; OR, odds ratio; SIRS, systemic inflammatory response syndrome.

a

MAP was estimated from blood pressure (MAP = diastolic pressure plus one‐third of the difference between systolic and diastolic pressure).

*p < 0.05; **p < 0.01.

5. Discussion

Mistriage was identified by comparing initial and final KTAS classification levels in 1110 samples. The over‐triage rate was higher than the under‐triage rate, aligning with the findings of prior studies employing ESI (Huabbangyang et al. 2023; Sax et al. 2023). However, a previous study conducted shortly after introducing the KTAS nationally in 2016 found a higher rate of under‐triage, which contradicts our findings (Moon et al. 2019). This discrepancy may arise from differences in the study period or characteristics of the medical facilities involved. Future studies should validate these findings using extensive multicenter data.

At KTAS Levels 3 and 4, 188 mistriaged cases (90.8% of all mistriage) were identified. Most patients in the ED (53.1%–81.1%) are assigned this level (Centers for Disease Control and Prevention 2023; Moon et al. 2019; Sax et al. 2023). Notably, at Level 3, significant variability arises within the same triage level because of the coexistence of critical and noncritical patients (Kang et al. 2020). In KTAS, Level 3 refers to “a condition requiring first aid and being at a potential risk of escalating to a serious issue” (Choi et al. 2019). Briefly, although it does not pose a threat to life, it indicates that a swift evaluation and intervention are necessary. Although it is crucial for the triage system to effectively differentiate between patients with high and low emergency levels, initial evaluations are constrained by information and time limitations. Additionally, the KTAS, which is based on the present symptoms, makes it difficult to reflect potential variability in patient status. Therefore, the broad scope of Level 3, aimed at ensuring patient safety by early patient identification who may deteriorate rapidly based on their present medical status, is regarded as involving overlapping emergency and nonemergency patients (Kang et al. 2020). This reflects the complex nature of emergency care. However, this allowance for flexibility may lead to uncertainty in the triage decision‐making process, potentially compromising accuracy, by prioritizing over‐triage to minimize the likelihood of under‐triage (Lee et al. 2019). Consequently, differentiated classification approaches at Level 3 should be developed, to minimize mistriage in this triage level.

The criteria differentiating Level 4, which exhibited the largest under‐triage in the KTAS, from level ≥ 3 were distinct from the ESI. The ESI algorithm uses relatively straightforward decision points, such as required resources or vital indicators (ENA 2023). Conversely, the KTAS categorizes patients based on 155 primary symptoms in adults. It uses a complex classification process that includes objective indicators such as vital signs, shortness of breath, the Glasgow Coma Scale score, mechanism of injury, and blood sugar levels, along with subjective indicators such as pain severity, “unwell appearance” in fever, and unquantified bleeding (Park and Lim 2017). Selected complaints and modifiers can lead to variability in the classification rates (8.1%–50.5%) (Park et al. 2019).

Level 3 may be over‐triaged when providers choose an intermediate level for challenging cases not clearly defined in the KTAS algorithm or are unsure about the severity of the patient's condition (Levis‐Elmelech et al. 2022). In South Korea, unlike the KTAS Levels 4 and 5 with a 50%–60% out‐of‐pocket cost ratio for ED visits during the study period, levels ≥ 3, which are covered by extensive national health insurance with an out‐of‐pocket cost ratio of 20% (Health Insurance Review and Assessment Service 2024; Ministry of Health and Welfare [MOHW] 2015), may influence this. Additionally, in this study, a classification experience of < 3 years was the majority (75.1%), and limited classification experience may be associated with increased over‐triage (Levis‐Elmelech et al. 2022).

The triage provider determines the pace at which the patients receive emergency medical care. The likelihood of serious risks can be decreased by implementing highly sensitive criteria to minimize under‐triage. As Level 3 often requires patients to wait in a crowded waiting room (Ding et al. 2010), and extended waiting periods can gradually exacerbate their condition. Precise methods are needed to accurately identify patients categorized as Level 3 to improve the distribution of patients in emergency and reduce mistriage.

This study aimed to use statistical methods to discover the factors contributing to mistriage using the KTAS. Identifying the causes of mistriage is an essential initial step toward the major goal of emergency patient classification, which is to allocate limited resources to critical patients preferentially. We identified factors contributing to mistriage, including age, MAP, pain, presenting complaints, and KTAS experience, according to the individual characteristics of the patient and triage provider.

Regarding mistriage, the demographic characteristics of the participants revealed that the younger age group was more prone to over‐triage than the older age group, which is consistent with previous studies (Huabbangyang et al. 2023; Sax et al. 2023). Younger patients tend to exhibit more intense symptoms and heightened pain sensitivity (Arslanian‐Engoren 2000). Over‐triage can be affected, but further research is required to fully understand this effect. This study indicates that age should be considered in classification algorithms because it demonstrates that variations in age may result in discrepancies in over‐triage. Young and older adults differ in resource utilization, hospitalization rates, and mortality in EDs despite exhibiting similar symptoms and falling into the same categories (Ginsburg et al. 2021). This finding is supported by the statement that classification should consider the unique challenges of older adults, such as physiological shifts (Bullard et al. 2017).

Regarding mistriage, the participants' clinical characteristics included MAP. Low MAPs are the sole objective indicators of physiological indices linked to under‐triage. This study found that lower MAPs can be linked to under‐triage despite the general understanding that lower MAPs are associated with higher levels of acuity (Petruniak et al. 2018). Although KTAS, such as ESI, evaluates hemodynamic stability along with signs of hypoperfusion and hypotension (ENA 2023; Kim et al. 2022), this may be related to not incorporating blood pressure thresholds below the normal range in its algorithm. In this situation, patient‐specific criteria, such as Levels 2 (lower than the expected blood pressure for a patient) and 3 (different from the patient's normal values) (Kim et al. 2022)—in other words, the ambiguous difference between these two levels and inaccurate recognition by triage providers—can be associated with under‐triage. Nevertheless, the triage provider must recognize the potential physiological impairments in a relatively normotensive compensated shock state. Noncritical hypotension is associated with the incidence of sepsis and higher mortality (Coeckelenbergh et al. 2019). Awareness of these potential threats is essential to minimize under‐triage. Blood pressure is a crucial and quickly obtainable objective component of ED. Thus, it is vital to revise the ambiguous criteria to improve the validity of the guidelines.

In this study, compared to those without pain, patients experiencing pain were approximately twice as likely to be over‐triaged. Previous research has identified pain as a prominent factor contributing to over‐triage, which aligns with our findings (Choi and Lee 2023; Davis et al. 2022; Lee et al. 2019; Moon et al. 2019). Pain, one of the major modifiers of the KTAS algorithm, relies on subjective representation and evaluation by patients and triage providers, with patients expressing the intensity of their pain on the 10‐point Likert scale (Park and Lim 2017). If a patient exaggerates their pain or misunderstands the pain score concept, reliable estimates may not be obtained (Sampson et al. 2019). Additionally, anxiety may escalate in relation to acute pain, impacting perceived pain levels when classified (Kapoor et al. 2016). Thus, the triage provider sometimes adjusts the pain score arbitrarily to raise or lower the priority of treatment, considering other clinical factors, such as the patient's urgency or availability of ED beds.

Although clinical judgment by medical staff yields a pain score that better predicts hospitalization than self‐reported pain assessments in the ED (Ku et al. 2023), variability in the competencies of medical personnel poses limitations. The KTAS pain classification system categorizes pain into three levels according to severity: mild (score 1–3), moderate (score 4–7), and severe (score 8–10). Each level is further subdivided into central and peripheral categories based on pain location. Additionally, the system stratifies pain into different acuity levels based on its chronicity (Park and Lim 2017). This complex pain classification system can lead to incorrect use by triage providers (Choi and Lee 2023; Moon et al. 2019).

Standards must be established to exclude subjectivity in pain assessments. Currently, the KTAS guidelines define the degree of dyspnea and dehydration differently based on the observed appearance and patterns exhibited by patients. This guideline recommends considering other clinical indices when determining the classification levels. However, although a high pain score can be classified beyond the presenting complaint and vital signs, the evaluation primarily considers the pain score, location, and chronicity. Incorporating observable, objective data and physiological indices combined with the patient‐reported pain level could improve the current classification criteria based on self‐reporting, thereby supporting the classification provider's clinical decision‐making process.

This study identified the factors leading to mistriage that were associated with various symptoms. Compared to neurological symptoms, an increased risk of over‐triage was observed for respiratory, cardiovascular, gastrointestinal, and musculoskeletal or skin symptoms. Symptoms of the respiratory, cardiovascular, gastrointestinal, and trauma systems are prioritized in the ED (Giri et al. 2022; Moura et al. 2022). In the ED, dyspnea is a major factor in 30‐day mortality and readmission (Sørensen et al. 2021), and 10% of patients presenting with chest pain are diagnosed with acute coronary syndrome (Fanaroff et al. 2015). Thus, over‐triage may be explained by the triage provider's awareness of high‐risk and time‐sensitive conditions and medical institution's permitted pathways to exclude severe conditions early and prevent delayed treatment of these potentially critical conditions. Additionally, many gastrointestinal and musculoskeletal symptoms are pain‐related (Lee et al. 2019) and may be assigned a higher level when sufficient clinical data cannot be collected or pain is considered a significant modifier.

Conversely, under‐triage was more prevalent among those with complaints such as fever and immune‐related symptoms. In the KTAS, specific criteria are specified to differentiate acuity levels within the presenting complaints of fever, including circulatory, respiratory, and neurological impairments and SIRS criteria. Despite not being designed to assign patients to a lower level, these results can be interpreted from two perspectives: the timing of data collection and inappropriate application of the SIRS criteria. The decreasing severity of infectious diseases associated with the spread of the omicron variant of coronavirus disease 2019 and World Health Organization's declaration of the end of the pandemic may have been reflected here. In conjunction with this unique situation, triage providers may have prioritized clinical judgment over the SIRS criteria to accommodate many patients with fevers and limited resources.

It is necessary to consider whether applying the SIRS criteria to patients with fever is valid in determining the KTAS classification level. During the initial triage, only temperature, pulse, and respiratory rate can be utilized from the SIRS criteria, excluding the white blood cell count and PaCO2 value (a limitation in most cases). Applying these incomplete SIRS criteria, which failed to predict intensive care unit admissions and emergent interventions (Sun et al. 2019), highlights the need to improve the validity of the alignment between the SIRS criteria and KTAS levels or to discuss alternative criteria. Additionally, patients in an emergency often present with undifferentiated symptoms rather than diagnosed conditions, and these symptoms sometimes appear in more complex combinations than single ones. Therefore, refining triage tools and providing standardized criteria for quantifying nonspecific presentations is essential.

Experience with the KTAS classification was a factor in over‐triage, which was attributed to the characteristics of the triage provider. Specifically, this study found that nurses with < 2 years of KTAS experience were more likely to over‐triage compared to nurses with ≥ 4 years of experience. Triage providers' classification competency is the most important facilitator of efficient triage (Moon et al. 2021). A previous study has suggested that classification anxiety from limited experience can lead to over‐triage (Levis‐Elmelech et al. 2022), supporting the finding that limited classification experience is a factor in over‐triage.

While this study found differences in over‐triage based on nurses' overall clinical experience, it was not a significant factor. This finding indicates that triage is an independent specialty, even within emergency nursing, and triage competency is more significantly influenced by direct experience in triaging emergency patients than overall clinical experience. Additionally, the current KTAS training consists of 4.5 h of lectures and an exam to obtain certification, which must be renewed every 3 years through refresher training and an exam. Adult Learning Theory reinforces these experiences along with the refresher training. This is because particular objectives are established, grounded in an individual class of experiences and the promotion of an educational strategy, focused on classifying the providers such that they can immediately implement this in reality (Knowles 1978). Therefore, nurses with > 4 years of triage experience, who have completed the refresher training and examination, may exhibit variations in KTAS classification accuracy. This suggests that the retraining could be a significant factor in the differences observed among nurse groups based on their experience with the KTAS classification.

The accuracy of the individual providers supports the overall quality. Therefore, ongoing accuracy assessments should be implemented, with adjustments and additional training provided as needed (ENA 2018). Additionally, to reduce mistriage and strengthen nurses' competency, a process integrating regular competency evaluations with related educational programs should be established.

5.1. Limitations

This study is significant as it statistically evaluated mistriage using the KTAS, considering clinical data, triage providers, and ED context. However, it has some limitations. First, the retrospective study design using NEDIS data limited the number of predictive variables that could be identified. Although electronic health records complemented this, the single‐center design may have caused the results to reflect the characteristics of that institution. Second, limitations included the inability to obtain visual information, such as the “critical first look” and data collected from conversations with patients, which can only be acquired during the emergency triage process. Finally, although sPO2 is an important variable in determining KTAS levels, it was not mandatorily measured in all patients, limiting the analysis. Therefore, future research should involve multicenter prospective studies to determine the classification levels and evaluate predictive factors more accurately.

5.2. Finding Implications

Improvements in KTAS mistriage may have implications in the emergency medical setting across multiple dimensions. Economically, medical costs can be reduced by curtailing superfluous treatments, examinations, and LOS. Operationally, this promotes the efficient allocation of scarce resources. From the perspective of patient safety, it enables early intervention for critically ill individuals. From the perspective of patient satisfaction, it can decrease waiting times and enhance the quality of care. These interconnected elements shape the overall quality and functionality of emergency medical services. Although specific data quantifying the impacts of improving mistriage in KTAS are currently unavailable, further studies should explore the potential cost savings, reductions in waiting times, and enhanced patient outcomes associated with such improvements.

An ideal emergency patient triage system should ensure that the same triage level represents the same urgency (Lee et al. 2019). To achieve this, improving the KTAS tool by addressing the factors contributing to mistriage identified in this study and comparing them with additional prospective studies is necessary. Additionally, regular training to enhance the competency of triage providers, along with feedback through monitoring mistriage to improve accuracy, is essential. Although legal standards for emergency patient acuity classification and triage fee calculation have been established in Korea (MOHW 2016), the evaluation results are solely linked to emergency medical fees based on recorded medical data (NEMC 2022), without specific measures for managing and improving mistriage. Therefore, establishing cutoff criteria for managing mistriage and enhancing regular audit evaluation methods through management guidelines that reflect real‐world conditions is necessary. This approach will form the basis for continuously maintaining appropriate triage levels and improving the emergency patient classification system using the KTAS.

6. Conclusion

By improving mistriage, the quality of emergency medical services may be enhanced through reduced costs, increased operational efficiency, and improved patient safety and satisfaction. Applying simple and objective standardized criteria to control for the factors contributing to mistriage, validation of triage tools, and enhancing provider training is essential to improve the accuracy of emergency patient triage. Establishing appropriate regulatory and financial incentives for the triage process and developing more realistic standards for managing mistriage will optimize triage and ensure prompt, prioritized care.

Conflicts of Interest

The authors declare no conflicts of interest.

Clinical Resources

Supporting information

Table S1. Summary of the studies.

JNU-57-860-s002.docx (31.7KB, docx)

Table S2. Strobe Statement.

JNU-57-860-s001.pdf (156.9KB, pdf)

Acknowledgments

This work was supported by the Korean Association of Advanced Practice Nurses (KAAPN) fund in 2024 (No. 2024‐02). We would also like to thank Jinseob Kim, M.D., M.P.H., the CEO of Zarathu, for his valuable assistance with the statistical analysis.

Funding: The authors received no specific funding for this work.

Contributor Information

Nayeon Yi, Email: n990011@gmail.com.

Dain Baik, Email: lafore103@naver.com.

Data Availability Statement

Data sharing is not applicable to this article due to privacy or ethical restrictions.

References

  1. American Academy of Emergency Nurse Practitioners , and Emergency Nurses Association . 2021. Emergency Nurse Practitioner Competencies. AAENP & ENA. https://www.ena.org/docs/default‐source/education‐document‐library/enpcompetencies_final.pdf?sfvrsn=f75b4634_0. [Google Scholar]
  2. Arslanian‐Engoren, C. 2000. “Gender and Age Bias in Triage Decisions.” Journal of Emergency Nursing 26, no. 2: 117–124. 10.1016/S0099-1767(00)90053-9. [DOI] [PubMed] [Google Scholar]
  3. Ausserhofer, D. , Zaboli A., Pfeifer N., Siller M., and Turcato G.. 2020. “Performance of the Manchester Triage System in Patients With Dyspnoea: A Retrospective Observational Study.” International Emergency Nursing 53: 100931. 10.1016/j.ienj.2020.100931. [DOI] [PubMed] [Google Scholar]
  4. Benner, P. 1984. From Novice to Expert: Excellence and Power in Clinical Nursing Practice. Addison‐Wesley. [Google Scholar]
  5. Bullard, M. J. , Melady D., Emond M., et al. 2017. “Guidance When Applying the Canadian Triage and Acuity Scale (CTAS) to the Geriatric Patient: Executive Summary.” CJEM 19, no. Suppl. 2: S28–S37. 10.1017/cem.2017.363. [DOI] [PubMed] [Google Scholar]
  6. Centers for Disease Control and Prevention . 2023. National Hospital Ambulatory Medical Care Survey: Emergency Department Summary Tables. CDC. https://www.cdc.gov/nchs/ahcd/web_tables.htm#2021. [Google Scholar]
  7. Cho, M. S. , Kwon I. G., Kim K. H., Kim M. S., and Cho Y. A.. 2017. “Validity and Applicability of Clinical Ladder System Model for Nurses.” Journal of Korean Clinical Nursing Research 23, no. 3: 281–292. 10.22650/JKCNR.2017.23.3.281. [DOI] [Google Scholar]
  8. Choi, H. , Ok J. S., and An S. Y.. 2019. “Evaluation of Validity of the Korean Triage and Acuity Scale.” Journal of Korean Academy of Nursing 49, no. 1: 26–35. 10.4040/jkan.2019.49.1.26. [DOI] [PubMed] [Google Scholar]
  9. Choi, Y. , and Lee H.. 2023. “Factors Related to Under‐Triage of Patients With Acute Coronary Syndrome in the Emergency Department: A Retrospective Study.” International Emergency Nursing 69: 101316. 10.1016/j.ienj.2023.101316. [DOI] [PubMed] [Google Scholar]
  10. Chung, H. S. , Namgung M., Lee D. H., Choi Y. H., and Bae S. J.. 2023. “Validity of the Korean Triage and Acuity Scale in Older Patients Compared to the Adult Group.” Experimental Gerontology 175: 112136. 10.1016/j.exger.2023.112136. [DOI] [PubMed] [Google Scholar]
  11. Cicchetti, D. V. 1981. “Testing the Normal Approximation and Minimal Sample Size Requirements of Weighted Kappa When the Number of Categories Is Large.” Applied Psychological Measurement 5, no. 1: 101–104. 10.1177/014662168100500114. [DOI] [Google Scholar]
  12. Coeckelenbergh, S. , Van Nuffelen M., and Mélot C.. 2019. “Sepsis Is Frequent in Initially Non‐Critical Hypotensive Emergency Department Patients and Is Associated With Increased Mortality.” American Journal of Emergency Medicine 37, no. 12: 2242–2245. 10.1016/j.ajem.2019.158360. [DOI] [PubMed] [Google Scholar]
  13. Corkery, N. , Avsar P., Moore Z., O'Connor T., Nugent L., and Patton D.. 2021. “What Is the Impact of Team Triage as an Intervention on Waiting Times in an Adult Emergency Department?–A Systematic Review.” International Emergency Nursing 58: 101043. 10.1016/j.ienj.2021.101043. [DOI] [PubMed] [Google Scholar]
  14. Davis, S. , Ju C., Marchandise P., Diagne M., and Grant L.. 2022. “Impact of Pain Assessment on Canadian Triage and Acuity Scale Prediction of Patient Outcomes.” Annals of Emergency Medicine 79, no. 5: 433–440. 10.1016/j.annemergmed.2022.01.014. [DOI] [PubMed] [Google Scholar]
  15. Ding, R. , McCarthy M. L., Desmond J. S., Lee J. S., Aronsky D., and Zeger S. L.. 2010. “Characterizing Waiting Room Time, Treatment Time, and Boarding Time in the Emergency Department Using Quantile Regression.” Academic Emergency Medicine 17, no. 8: 813–823. 10.1111/j.1553-2712.2010.00812.x. [DOI] [PubMed] [Google Scholar]
  16. Emergency Nurses Association . 2018. Position Statement: Triage Qualifications and Competency. ENA. https://media.emscimprovement.center/documents/triagequalificationscompetency.pdf. [Google Scholar]
  17. Emergency Nurses Association . 2023. Emergency Severity Index Handbook. 5th ed. ENA. https://californiaena.org/wp‐content/uploads/2023/05/ESI‐Handbook‐5th‐Edition‐3‐2023.pdf. [Google Scholar]
  18. Fanaroff, A. C. , Rymer J. A., Goldstein S. A., Simel D. L., and Newby L. K.. 2015. “Does This Patient With Chest Pain Have Acute Coronary Syndrome?: The Rational Clinical Examination Systematic Review.” JAMA 314, no. 18: 1955–1965. 10.1001/jama.2015.12735. [DOI] [PubMed] [Google Scholar]
  19. Gilboy, N. , Tanabe P., Travers D., and Rosenau A. M.. 2020. Emergency Severity Index (ESI): A Triage Tool for Emergency Department Care Version 4. (2020 ed.). Emergency Nurses Association. [Google Scholar]
  20. Ginsburg, A. D. , Oliveira J. E., Silva L., et al. 2021. “Should Age Be Incorporated Into the Adult Triage Algorithm in the Emergency Department?” American Journal of Emergency Medicine 46: 508–514. 10.1016/j.ajem.2020.10.075. [DOI] [PubMed] [Google Scholar]
  21. Giri, S. , Watts M., LeVine S., and Tshering U.. 2022. “Characteristics and Outcomes of Patients Triaged as Critically Ill in the Emergency Department of a Tertiary Care Hospital in Bhutan.” International Journal of Emergency Medicine 15: 64. 10.1186/s12245-022-00468-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Health Insurance Review and Assessment Service . 2024. Out‐Of‐Pocket Cost Criteria for Health Insurance. Health Insurance Review and Assessment Service. https://www.hira.or.kr/dummy.do?pgmid=HIRAA030056020100. [Google Scholar]
  23. Hinson, J. S. , Martinez D. A., Cabral S., et al. 2019. “Triage Performance in Emergency Medicine: A Systematic Review.” Annals of Emergency Medicine 74, no. 1: 140–152. 10.1016/j.annemergmed.2018.09.022. [DOI] [PubMed] [Google Scholar]
  24. Huabbangyang, T. , Rojsaengroeng R., Tiyawat G., et al. 2023. “Associated Factors of Under and Over‐Triage Based on the Emergency Severity Index: A Retrospective Cross‐Sectional Study.” Archives of Academic Emergency Medicine 11, no. 1: e57. 10.22037/aaem.v11i1.2076. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Jang, K. , and Seo Y. H.. 2024. “Characteristics of Undertriaged Older Patients in the Emergency Department: Retrospective Study.” International Emergency Nursing 75: 101477. 10.1016/j.ienj.2024.101477. [DOI] [PubMed] [Google Scholar]
  26. Jung, S. , and Yi Y.. 2024. “Incidence of Overtriage and Undertriage and Associated Factors: A Cross‐Sectional Study Using a Secondary Data Analysis.” Journal of Advanced Nursing 80, no. 4: 1405–1416. 10.1111/jan.15895. [DOI] [PubMed] [Google Scholar]
  27. Kang, D. Y. , Cho K. J., Kwon O., et al. 2020. “Artificial Intelligence Algorithm to Predict the Need for Critical Care in Prehospital Emergency Medical Services.” Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine 28: 17. 10.1186/s13049-020-0713-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Kapoor, S. , White J., Thorn B. E., and Block P.. 2016. “Patients Presenting to the Emergency Department With Acute Pain: The Significant Role of Pain Catastrophizing and State Anxiety.” Pain Medicine 17, no. 6: 1069–1078. 10.1093/pm/pnv034. [DOI] [PubMed] [Google Scholar]
  29. Kim, H. I. , Oh S. B., and Choi H. J.. 2019. “Inter‐Rater Agreement of Korean Triage and Acuity Scale Between Emergency Physicians and Nurses.” Journal of the Korean Society of Emergency Medicine 30, no. 4: 309–317. [Google Scholar]
  30. Kim, J. Y. , Hong D. Y., Kim S. Y., et al. 2017. “Reliability of Korean Triage and Acuity Scale‐Based Triage System as a Severity Index in Emergency Patients.” Journal of the Korean Society of Emergency Medicine 28, no. 6: 552–556. [Google Scholar]
  31. Kim, S. W. , Kim Y. W., Min Y. H., et al. 2022. “Development and Validation of Simple Age‐Adjusted Objectified Korean Triage and Acuity Scale for Adult Patients Visiting the Emergency Department.” Yonsei Medical Journal 63, no. 3: 272–281. 10.3349/ymj.2022.63.3.272. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Knowles, M. S. 1978. “Andragogy: Adult Learning Theory in Perspective.” Community College Review 5, no. 3: 9–20. 10.1177/009155217800500. [DOI] [Google Scholar]
  33. Ku, N. W. , Cheng M. T., Liew C. Q., et al. 2023. “Prospective Study of Pain and Patient Outcomes in the Emergency Department: A Tale of Two Pain Assessment Methods.” Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine 31: 56. 10.1186/s13049-023-01130-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Kwon, H. , Kim Y. J., Jo Y. H., et al. 2019. “The Korean Triage and Acuity Scale: Associations With Admission, Disposition, Mortality and Length of Stay in the Emergency Department.” International Journal for Quality in Health Care 31, no. 6: 449–455. 10.1093/intqhc/mzy184. [DOI] [PubMed] [Google Scholar]
  35. Landis, J. R. , and Koch G. G.. 1977. “The Measurement of Observer Agreement for Categorical Data.” Biometrics 33, no. 1: 159–174. 10.2307/2529310. [DOI] [PubMed] [Google Scholar]
  36. Lee, E. S. , and Oh H.. 2021. “Re‐Evaluation Characteristics of the Korean Triage and Acuity Scale (KTAS): The Relationship Between Overcrowding and KTAS re‐Evaluation.” Journal of the Korean Society of Emergency Medicine 32, no. 2: 179–188. [Google Scholar]
  37. Lee, I. , Kim O., Kim C., et al. 2018. “Validity Analysis of Korean Triage and Acuity Scale.” Journal of the Korean Society of Emergency Medicine 29, no. 1: 13–20. [Google Scholar]
  38. Lee, J. H. , Park Y. S., Park I. C., et al. 2019. “Over‐Triage Occurs When Considering the Patient's Pain in Korean Triage and Acuity Scale (KTAS).” PLoS One 14, no. 5: e0216519. 10.1371/journal.pone.0216519. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Levis‐Elmelech, T. , Schwartz D., and Bitan Y.. 2022. “The Effect of Emergency Department Nurse Experience on Triage Decision Making.” Human Factors in Healthcare 2: 100015. 10.1016/j.hfh.2022.100015. [DOI] [Google Scholar]
  40. Levy, M. M. , Fink M. P., Marshall J. C., et al. 2003. “2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference.” Critical Care Medicine 31, no. 4: 1250–1256. 10.1097/01.CCM.0000050454.01978.3B. [DOI] [PubMed] [Google Scholar]
  41. Lim, C. Y. , Park S. Y., Park K. H., Park H. Y., and Kim J. E.. 2018. “The Predictive Factors for Hospitalization of Nonurgent Patients Visiting an Emergency Department in an Urban Area: A Single Center Study.” Journal of the Korean Society of Emergency Medicine 29, no. 2: 152–159. [Google Scholar]
  42. Lim, Y. D. , Lee D. H., Lee B. K., Cho Y. S., and Choi G.. 2020. “Validity of the Korean Triage and Acuity Scale for Predicting 30‐Day Mortality due to Severe Trauma: A Retrospective Single‐Center Study.” European Journal of Trauma and Emergency Surgery 46, no. 4: 895–901. 10.1007/s00068-018-1048-y. [DOI] [PubMed] [Google Scholar]
  43. Ministry of Government Legislation . 2025. Enforcement Rule of the Emergency Medical Service Act 18‐3. Korea Ministry of Government Legislation. https://www.law.go.kr/법령/응급의료에관한법률시행규칙/(20250311,01096,20250311)/제18조의3. [Google Scholar]
  44. Ministry of Health and Welfare . 2015. The Ministry of Health and Welfare Announcement (No. 2015‐241). MOHW. https://www.mohw.go.kr/board.es?mid=a10409020000&bid=0026&tag=&act=view&list_no=329275. [Google Scholar]
  45. Ministry of Health and Welfare . 2016. The Ministry of Health and Welfare Announcement (No. 2016–275). MOHW. https://www.mohw.go.kr/board.es?mid=a10409020000&bid=0026&act=view&list_no=337928&tag=&nPage=1. [Google Scholar]
  46. Mirhaghi, A. , Heydari A., Mazlom R., and Ebrahimi M.. 2015. “The Reliability of the Canadian Triage and Acuity Scale: Meta‐Analysis.” North American Journal of Medical Sciences 7, no. 7: 299–305. 10.4103/1947-2714.161243. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Moon, S. H. , Jeon M. K., and Ju D.. 2021. “Facilitators and Barriers of the Triage Process Based on Emergency Nurses' Experience With the Korean Triage and Acuity Scale: A Qualitative Content Analysis.” Asian Nursing Research 15, no. 4: 255–264. 10.1016/j.anr.2021.08.001. [DOI] [PubMed] [Google Scholar]
  48. Moon, S. H. , Shim J. L., Park K. S., and Park C. S.. 2019. “Triage Accuracy and Causes of Mistriage Using the Korean Triage and Acuity Scale.” PLoS One 14, no. 9: e0216972. 10.1371/journal.pone.0216972. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Moura, B. R. S. , Oliveira G. N., Medeiros G., Vieira A. D. S., and Nogueira L. D. S.. 2022. “Rapid Triage Performed by Nurses: Signs and Symptoms Associated With Identifying Critically Ill Patients in the Emergency Department.” International Journal of Nursing Practice 28, no. 1: e13001. 10.1111/ijn.13001. [DOI] [PubMed] [Google Scholar]
  50. National Emergency Medical Center . 2022. Guidelines for the Evaluation of Emergency Medical Institutions 2023. National Emergency Medical Center. [Google Scholar]
  51. National Emergency Medical Center . 2024. Emergency Medical Statistical Annual Report, 2014–2023. National Emergency Medical Center. https://www.e‐gen.or.kr/nemc/statistics_annual_report.do. [Google Scholar]
  52. Nishi, F. A. , Polak C., and Cruz D. A. L. M. D.. 2018. “Sensitivity and Specificity of the Manchester Triage System in Risk Prioritization of Patients With Acute Myocardial Infarction Who Present With Chest Pain.” European Journal of Cardiovascular Nursing 17, no. 7: 660–666. 10.1177/1474515118777402. [DOI] [PubMed] [Google Scholar]
  53. Oh, B. Y. , and Kim K.. 2021. “Factors Associated With the Undertriage of Patients With Abdominal Pain in an Emergency Room.” International Emergency Nursing 54: 100933. 10.1016/j.ienj.2020.100933. [DOI] [PubMed] [Google Scholar]
  54. Oh, M. T. , Lee S. H., Park S. W., et al. 2018. “Factors Associated Hospital Admission in Patients With Low Acuity Visiting Emergency Department.” Journal of the Korean Society of Emergency Medicine 29, no. 5: 408–414. [Google Scholar]
  55. Park, J. , and Lim T.. 2017. “Korean Triage and Acuity Scale (KTAS).” Journal of the Korean Society of Emergency Medicine 28, no. 6: 547–551. [Google Scholar]
  56. Park, J. , Shin C. H., and Lee J. Y.. 2024. “Why Did All the Residents Resign? Key Takeaways From the Junior Physicians' Mass Walkout in South Korea.” Journal of Graduate Medical Education 16, no. 4: 402–406. 10.4300/JGME-D-24-00227.1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Park, J. B. , Lee J., Kim Y. J., Lee J. H., and Lim T. H.. 2019. “Reliability of Korean Triage and Acuity Scale: Interrater Agreement Between Two Experienced Nurses by Real‐Time Triage and Analysis of Influencing Factors to Disagreement of Triage Levels.” Journal of Korean Medical Science 34, no. 28: e189. 10.3346/jkms.2019.34.e189. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Peduzzi, P. , Concato J., Kemper E., Holford T. R., and Feinstein A. R.. 1996. “A Simulation Study of the Number of Events Per Variable in Logistic Regression Analysis.” Journal of Clinical Epidemiology 49, no. 12: 1373–1379. 10.1016/S0895-4356(96)00236-3. [DOI] [PubMed] [Google Scholar]
  59. Petruniak, L. , El‐Masri M., and Fox‐Wasylyshyn S.. 2018. “Exploring the Predictors of Emergency Department Triage Acuity Assignment in Patients With Sepsis.” Canadian Journal of Nursing Research 50, no. 2: 81–88. 10.1177/0844562118766178. [DOI] [PubMed] [Google Scholar]
  60. Sampson, F. C. , Goodacre S. W., and O'Cathain A.. 2019. “The Reality of Pain Scoring in the Emergency Department: Findings From a Multiple Case Study Design.” Annals of Emergency Medicine 74, no. 4: 538–548. 10.1016/j.annemergmed.2019.02.018. [DOI] [PubMed] [Google Scholar]
  61. Sax, D. R. , Warton E. M., Mark D. G., et al. 2023. “Evaluation of the Emergency Severity Index in US Emergency Departments for the Rate of Mistriage.” JAMA Network Open 6, no. 3: e233404. 10.1001/jamanetworkopen.2023.3404. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Solheim, J. 2016. Emergency Nursing: The Profession, the Pathway, the Practice. Sigma Theta Tau International. [Google Scholar]
  63. Sørensen, S. F. , Ovesen S. H., Lisby M., Mandau M. H., Thomsen I. K., and Kirkegaard H.. 2021. “Predicting Mortality and Readmission Based on Chief Complaint in Emergency Department Patients: A Cohort Study.” Trauma Surgery & Acute Care Open 6, no. 1: e000604. 10.1136/tsaco-2020-000604. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Sun, J. , Chung H., Jang H., Kim S., Lee Y., and Park J.. 2019. “Validation of Systemic Inflammatory Response Syndrome Criteria Without White Blood Cell Count in Korean Triage and Acuity Scale.” Journal of the Korean Society of Emergency Medicine 30, no. 3: 232–238. [Google Scholar]
  65. Vittinghoff, E. , and McCulloch C. E.. 2007. “Relaxing the Rule of Ten Events Per Variable in Logistic and Cox Regression.” American Journal of Epidemiology 165, no. 6: 710–718. 10.1093/aje/kwk052. [DOI] [PubMed] [Google Scholar]
  66. Zhang, W. , Zhang M., Yang P., Zhou W., Zheng J., and Zhang Y.. 2024. “The Reliability and Validity of Triage Tools in Geriatric Emergency Departments: A Scoping Review.” International Emergency Nursing 77: 101509. 10.1016/j.ienj.2024.101509. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Table S1. Summary of the studies.

JNU-57-860-s002.docx (31.7KB, docx)

Table S2. Strobe Statement.

JNU-57-860-s001.pdf (156.9KB, pdf)

Data Availability Statement

Data sharing is not applicable to this article due to privacy or ethical restrictions.

Articles from Journal of Nursing Scholarship are provided here courtesy of Wiley

RESOURCES