Abstract
Background:
While emergency patient triage system is effective when the pre-hospital triage acuity scale is linked with the in-hospital triage acuity scale. However, the 119 emergency medical triage system and the Korea Triage and Acuity Scale are not linked in Korea.
Objective:
This study aimed to investigate the correlation between the two triage systems and to utilize the results as basic data for the future development of a pre-hospital triage system.
Methods:
Among the 1114 patients who visited a regional emergency medical center by a 119 ambulance from April to May 2016, we analyzed the correlation between the pre-hospital and in-hospital triage systems based on the general characteristics of the patients and their reason of hospital visit (non-trauma or trauma).
Results:
Upon reclassifying the pre-hospital and in-hospital triage systems into three levels, among the 289 patients (28.1%) in level 3 of the pre-hospital triage, 79 (27.3%) were reclassified as the highest level (Resuscitation) in the in-hospital triage. The kappa coefficient as a measure of agreement between the two triage systems was very low at 0.211 (95% confidence interval, 0.164–0.258), and the kappa coefficient of the paramedic category was 0.232 (95% confidence interval, 0.161–0.303).
Conclusion:
There is a low agreement between the pre-hospital and in-hospital triage systems.
Introduction
Emergency medical care begins when a request is made to dispatch paramedics to the patient and continues in the hospital emergency room (ER). Demand for emergency medical care in Korea is on an increasing trend due to disasters, accidents, diseases, and changes in lifestyle.
As the number of emergency patients increases, triage is needed to provide treatment efficiently and to prioritize care. Accordingly, an in-hospital triage system was developed in 2012, 1 classifying patient conditions into five levels: Resuscitation, Emergent, Urgent, Less Urgent, and Non-Urgent. The system has been in use at local emergency medical centers of regional and higher levels across Korea. However, despite the triage system at the hospital level, the ERs are overcrowded due to indiscreet visits of the patients. The overcrowding of the ER adversely affects the treatment time and treatment outcome, reducing the quality of emergency medical services and hampering the performance of the ER.2,3
To solve the overcrowding and to establish an efficient emergency medical system in this situation, a good correlation of the 119 paramedics triage system, which constitutes most of the pre-hospital emergency medical care, is necessary. If the patients were transferred to the general clinic, emergency center, or trauma center according to the appropriate triage about the severity in the field, it can reduce unnecessary transport and can provide them optimal care. Currently, 119 paramedics triage system is based on the Simple Triage and Rapid Treatment (START) triage system in five levels—Emergent, Less Emergent, Potentially Emergent, Extra, and Death—according to the Standard Protocols for 119 Emergency Medical Services Providers. 4
The in-hospital triage system was based on the Canadian Triage and Acuity Scale (CTAS) after lengthy research. Although the system has been evaluated for validity and reliability 5 and has been in use nationwide in Korea, there is no study on the evaluation of the pre-hospital triage system. If triage is done appropriately in the pre-hospital stage, the outcome of patient treatment can be predicted at the in-hospital stage and the resources can be used efficiently. 4
This study aimed to analyze the correlation between the triage systems at the pre-hospital and in-hospital stages to better determine the priority of medical care for emergency patients and to utilize the results as basic data for the future development of a pre-hospital triage system.
Materials and methods
This retrospective study was conducted with data obtained from a regional emergency medical center. This study has been approved by the Research Ethics Committee of the Wonju Severance Christian Hospital, Yonsei University (approval number CR317016).
Data source
To determine the stages of the pre-hospital triage system, the paramedic activity records filled out by 119 paramedics were surveyed for the type of emergency, the age and sex of the patient, the type of patient event, the stage of patient triage, and the qualifications of the 119 emergency medical service providers (Paramedics, Emergency medical technician (EMT)-basic, Nurses, and Emergency training). To find out the stages of the in-hospital triage system, the emergency medical records of the regional emergency medical center were surveyed for the patient’s age and sex, Korea Triage and Acuity Scale (KTAS), time of visit, reason for visit, level of consciousness in ER, length of stay in the ER, outcome of ER discharge, length of hospital stay, outcome of hospital discharge, and cost of hospital treatment.
Participants
Among the 7141 patients who visited a regional emergency medical center from 1 April to 31 May 2016, 1114 patients used 119 ambulances. A total of 1028 patients were included in the study, after excluding 84 patients whose 119 paramedic activity records could not be found and 2 patients who were hospitalized during the study period.
Pre-hospital triage system
The pre-hospital triage system was classified into five stages of patient triage by 119 paramedics in the Standard Protocols for 119 Emergency Medical Services Providers: Emergent (E)—patients with at least one unstable vital sign, major symptoms of chest pain, unconsciousness, dyspnea, respiratory arrest, palpitations, cardiac arrest, paralysis, severe trauma patients and patients judged by paramedics to require prompt treatment within minutes; Less Emergent (LE)—if not Emergent but need treatment within a few hours; Potentially Emergent (PE)—all patients who are not Emergent, Less Emergent, but need emergency care; Extra: this is not an emergency transfer and for outpatient visits, reserved patients, and so on; and Death (D)—signs of apparent post mortem, a purple spot, head amputation, torso amputation, or apparent death of corruption, and patients with cardiac arrest are not eligible for this indication. The Extra classification, including cases without emergency patient transfers, was not included. In total, 119 Providers are Emergency Medical Services which consist of Paramedics, EMT-basic, and Nurse who have been trained through regular courses to provide Emergency Medical Services.
Hospital triage system: KTAS
The in-hospital triage system used five stages of the KTAS: Resuscitation (KTAS 1), Emergent (KTAS 2), Urgent (KTAS 3), Less Urgent (KTAS 4), and Non-Urgent (KTAS 5). KTAS is categorized when the patient enters the ER on a 119 ambulance and when the initial physician and triage nurse first see the patient. Patient classification is done by the nurse who has completed the KTAS education for more than 2 years in the ER. If the judgment is unclear, discuss it with the initial doctor and make a decision.
Statistics
We performed Fisher’s chi-square test on the triage based on the general characteristics and the reason for the hospital visit (non-trauma, trauma), admission rate, and duration of stay in the ER according to the triage level. The inter-rater agreement was used to analyze the correlation between the pre-hospital and in-hospital triage systems. The significance level was judged to be statistically significant when it was less than 0.05. Statistical analysis was performed using MedCalc Statistical Software version 12.7.0 (MedCalc Software, Ostend, Belgium) and PASW IBM SPSS Statistics version 21.0 (SPSS Inc., Chicago, IL, USA).
Results
From April to May 2016, a total of 7141 patients visited the medical center and 1114 patients used 119 ambulances. The total number of patients included was 1028. The mean age was 56.4 ± 22.7 years, and 596 patients (58.0%) were male. As for the reason of hospital visit, 672 patients (65.4%) had non-trauma-related reasons and 354 patients (34.4%) had trauma reasons; thus, those experiencing non-trauma conditions were 31% greater in number. As for reasons of departing the ER, the most common was 476 patients (46.3%) returning home with recovery, followed by 424 patients (41.2%) admitted (Table 1). As for reasons for hospital visit, both non-trauma and trauma were most commonly classified into the highest emergency level (non-trauma, 41.5%; trauma, 38.7%) at the pre-hospital stage. At the in-hospital stage, most disease cases were classified into KTAS 3 (46.1%) and most trauma cases into KTAS 1 (35.0%) (Table 2). At the time of visit to the ER, there were 814 Alert, 126 Verbal, 40 Painful, and 48 Unresponse in the hospital. When Mental was Alert, KTAS had the highest levels 3 and 2, but the pre-triage mostly had E and D levels. Of the 15 patients (33.3%) classified as KTAS, 5 were triaged with pre-triage E and D levels. When Mental was verbal, KTAS mostly had level 2 and pre-triage had E and D levels. When Mental was Pain and KTAS, 39 (97.5%) were classified as level 3 or more, 8 (22.5%) as pre-triage, and 2 (5.0%) as PE (Table 3). At the time of discharge in ER, 477 ER discharge, 36 Death in ER, 424 Admission, 57 Transfer, and 34 Operation. KTAS 3 and pre-triage LE were the most common at the time of ER discharge. All 36 patients in Death in ER were classified as KTAS 1 and 2, but 2 patients were classified as pre-triage PE (5.6%) (Table 4). There were 366 Paramedic patients, 548 EMT patients, and 111 Nurse patients according to pre-hospital triage for pre-hospital provider qualification. In the Paramedic classification, 12 patients (21.4%) of 56 patients with KTAS 1 and 29 patients (27.4%) of 106 patients with KTAS 2 were Undertriaged. In EMT-basic classification, 24 patients (23.8%) of 101 patients with KTAS 4 and 3 patients (30.0%) of 10 patients with KTAS 5 were Overtriaged. In Nurse classification, 11 (28.9%) of 38 patients with KTAS 2 were Undertriaged (Table 5). In the evaluation of agreement between the two triage systems, the kappa correlation coefficient was very low at 0.211 (95% confidence interval [CI], 0.164–0.258). In addition, for the qualification of pre-hospital triage personnel, the kappa value was 0.232 (95% CI, 0.161–0.303) for Paramedics, 0.205 (95% CI, 0.140–0.271) for EMT-basic, and 0.216 (95% CI, 0.086–0.346) for Nurses (Table 6).
Characteristics of patients enrolled in this study (n = 1028).
ED: emergency department; SD: standard deviation.
Triage scale of non-trauma and trauma groups.
KTAS: Korea Triage and Acuity Scale.
Comparison of KTAS and pre-hospital triage level of consciousness in emergency room.
E, D: Emergent, Death; LE: Less Emergent; PE: Potentially Emergent; KTAS: Korea Triage and Acuity Scale.
Comparison of KTAS and pre-hospital triage in emergency room discharge.
E, D: Emergent, Death; LE: Less Emergent; PE: Potentially Emergent; KTAS: Korea Triage and Acuity Scale; ER: emergency room.
Comparison of KTAS and pre-hospital triage of the qualification personnel.
E, D: Emergent, Death; LE: Less Emergent; PE: Potentially Emergent; KTAS: Korea Triage and Acuity Scale.
Correlation between the Korean Triage and Acuity Scale (KTAS) and pre-hospital triage for pre-hospital provider qualification.
Discussion
Triage is an indispensable method for accurately identifying patients and transferring them efficiently in complex and latency situations, as well as in disaster situations where there are large numbers of emerging patients.
The widely used in-hospital triage systems include the Australasian Triage Scale (ATS) in Australia, the Emergency Severity Index (ESI) in the United States, and the CTAS in Canada. The KTAS was developed 1 in 2012 by applying the Delphi method based on the CTAS and conducting research on 1000 nurses and paramedics. The KTAS showed fair to moderate reliability in reliability and validity studies and was evaluated as a valid tool. 5
The pre-hospital triage systems in use include the Guidelines for Field Triage of Injured Patients and Prehospital CTAS. However, 119 paramedics triage system is based on the START triage system in five levels—Emergent, Less Emergent, Potentially Emergent, Extra, and Death—according to the Standard Protocols for 119 Emergency Medical Services Providers. In fact, there are three levels—Emergent, Less Emergent, and Potentially Emergent—and there are few advanced studies related to validity and reliability.
There is a growing need for efficient triage with the gradual increase in the number of patients receiving emergency medical care.6,7 Although there is a wide range of evaluations done on the degree of agreement between in-hospital triage systems,8–11 there are not many studies on the degree of agreement between pre-hospital and in-hospital triage systems. If the patients are not triaged at the pre-hospital stage and the patients with severe symptoms and those with mild symptoms are mixed up in the ER, the waiting time for the treatment would become longer and the satisfaction is lowered. 12
When comparing the pre-hospital triage with the hospital triage to the state of consciousness at the time of visit in ER, Overtriage tendency was shown in pre-hospital. Meanwhile, the patients with consciousness Alert was classified as level 5 in KTAS, and they were Undertriaged as E in pre-triage. The symptoms of the patients were gait disturbance, urticaria after medication, convulsions due to hyperventilation, and hiccups that did not stop, but they were Overtriaged inappropriately.
Patients from conscious state of Pain were classified as KTAS 4, but pre-triage was classified as Emergent. The patients had a history of traumatic subarachnoid hemorrhage (SAH) and hematuria. Despite some patients in Alert group were classified as KTAS 1, they were Undertriaged as PE in pre-triage. The patients had symptoms such as nausea, dyspepsia, dizziness, sputum, cough, and burn. While KTAS was considered by the various factors such as main symptom, vital signs, pain level, and injury severity, pre-triage was classified by only main symptom. Due to the difference between them, pre-triage was Undertriaged (Table 3). This may be due to the characteristics of this study in pre-hospital and hospital severity classifications as pre-triage is based on the patient’s chief complaint, level of consciousness, and symptoms.
Patients classified as PE in the pre-hospital triage, but classified as KTAS 1, 119 Emergency Medical Services Providers were the first to see the patient because the vital sign was stable and the consciousness level was high.
At this time, according to the regional characteristics of the study, the hospital was classified as level 1 or 2 because of the long transfer time or the state of the patient.
Therefore, there is a limit to the point in time when the pre-hospital severity classification is not clear (immediately after contact with the patient, after the primary evaluation, during the secondary evaluation, during the transfer, after the evaluation of the patient, and after the treatment. Although the pre-hospital severity classification should be performed once, it is considered that the status and treatment status of the patients changed during the transfer were not taken into account and the degree of agreement with the severity classification score in the hospital was low.
Cohen’s kappa coefficient of 0.000–0.200 was defined as slight agreement, 0.201–0.400 as fair agreement, 0.401–0.600 as moderate agreement, 0.601–0.800 as substantial agreement, and 0.801–1.000 as perfect agreement. 13 In this study, the agreement rate was 0.211 for all patients, 0.232 for the Paramedic group, 0.205 for the EMT-basic group, and 0.216 for the Nurse group, showing fair agreement across all groups. A simulation study 14 using virtual samples for comparison of pre-hospital and in-hospital levels showed higher accuracy than this. The agreement on the triage time and accuracy was also analyzed. Only the ER triage nurses were included in the analysis, and the START method for disaster situations or mass screening was selected as the pre-hospital triage method.
The reason for the low level of agreement was that the analysis did not take into consideration cases of patients triaged first in the emergency situation and then receiving treatment or cases that exacerbated during transfer to the hospital. Overtriage or Undertriage occurred because the patients were triaged for the second time in-hospital without considering the changed patient condition. In addition, the qualifications and career experience requirements of pre-hospital triage personnel were not controlled, and it was not possible to find out whether they had received the same training. Another limitation is that the triage workers at the in-hospital stage were all nurses with triage training but there was a wide variety in terms of their work experiences in the ER. The results of this study are more realistic because the triage workers on the field analyzed the records of actual patients’ symptoms and emergency site conditions. This study was the first to compare the pre-hospital triage and the in-hospital triage of the study. It could be helpful in classifying the field so quickly and simply, helping a pre-hospital triage system to develop reflecting the patient’s condition adequately, and selecting the right hospital.
Conclusion
There is a low agreement between the pre-hospital and in-hospital triage systems. Therefore, there is a need to develop a more systematic and unambiguous pre-hospital triage system based on the KTAS, which is one way to reduce overcrowding in the ERs in Korea, helping to operate an efficient emergency medical system.
Footnotes
Acknowledgements
S.S. conceived the idea for the study, designed the study, analyzed the data, and drafted the final manuscript. C.Y.K. participated in study design and data collection, contributed to, and approved the final manuscript as submitted. H.Y.L. participated in data collection, analysis and approved the final manuscript as submitted. J.H.L. participated in data collection, analysis and approved the final manuscript as submitted. O.H.K. assisted in study design, critically reviewed and revised the manuscript, and approved the final manuscript as submitted. H.Y. reviewed the manuscript and approved the final manuscript as submitted. K.H.L. assisted in study design, critically reviewed and revised the manuscript, and approved the final manuscript as submitted. All authors contributed significantly to drafting and editing the manuscript.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the Korea Ministry of Land, Infrastructure and Transport, the Korea Agency for Infrastructure Technology Advancement (project no. 16PTSI-C054118-08).
Availability of data and materials
This study does not allow access to websites because it contains links to third-party data stores or detailed contact information about third-party data sources.
Ethical approval
This study has been approved by the Research Ethics Committee of the Wonju Severance Christian Hospital, Yonsei University (approval number CR317016).
