|Year : 2017 | Volume
| Issue : 2 | Page : 73-79
Validating the implementation of the triage system in an emergency department in a University Hospital
Abdulaziz Bin Saeed1, Faisal Majid Al-Fayyadh2, Khalid Mohammad Alshomar2, Ziad Wael Zekry2, Nawaf Nizar Alamiri2, Abdulrahman Mansour Abaalkhail2, Abdullah Abdulmohsen Aldughaither2, Yasser Abdulkarem Alaska3
1 Ministry of Health, College of Medicine, King Saud University, Riyadh 11461, Saudi Arabia
2 College of Medicine, King Saud University, Riyadh, Saudi Arabia
3 Department of Emergency Medicine, King Saud University, Riyadh 11323, Saudi Arabia
|Date of Web Publication||24-Apr-2017|
Faisal Majid Al-Fayyadh
King Saud University, Riyadh
Introduction: The implementation of triage systems seeks to aid patient organisation in order to assure smooth patient flow. Inconsistency in the application of the triage system at a local university hospital has been a recent concern for the administrative faculty. Therefore, our aim was to validate the implementation of the Canadian Triage and Acuity Scale (CTAS), currently applied by nurses, in a university hospital.
Materials and Methods: Patient information was collected in the Emergency Department and translated from real case scenarios into paper-based scenarios. A total of 49 cases were distributed among 45 nurses to categorise. The nurses' categorisations were compared with a CTAS expert's categorisations.
Results: Of the 2,205 cases (49 cases each given to the 45 nurses), 49% (1,059 cases) were correctly categorised and 51% (1,146 cases) were miscategorised. Overtriage and undertriage percentages were 55.93% and 44.07%, respectively. The highest level of agreement between nurses and the expert categorisation was in category 1; the lowest was in category 5.
Conclusion: The nurses' overall results were below expectations. Statistically significant variables affecting correct categorisation included age, experience, education level and nationality of the nurses. Nurses above the age of 45 years with more years of experience, obtained top scores. Nurses with the highest level of education also scored significantly higher. Filipino nurses scored better than nurses of other nationalities. With the widespread utilisation of triage systems in the region, further studies that evaluate their implementation are needed.
Keywords: Emergency Department, nurse, triage, university hospital
|How to cite this article:|
Saeed AB, Al-Fayyadh FM, Alshomar KM, Zekry ZW, Alamiri NN, Abaalkhail AM, Aldughaither AA, Alaska YA. Validating the implementation of the triage system in an emergency department in a University Hospital. J Health Spec 2017;5:73-9
|How to cite this URL:|
Saeed AB, Al-Fayyadh FM, Alshomar KM, Zekry ZW, Alamiri NN, Abaalkhail AM, Aldughaither AA, Alaska YA. Validating the implementation of the triage system in an emergency department in a University Hospital. J Health Spec [serial online] 2017 [cited 2018 Apr 24];5:73-9. Available from: http://www.thejhs.org/text.asp?2017/5/2/73/205072
| Introduction|| |
Triage: A recipe for reliability
Due to the combined effects of rising demand for care and resource limitations, multiple approaches have been implemented to organize priority of treatment amongst patients in emergency departments (ED). Most commonly, hospitals have adapted the triage classification system, which is under continuous development. The triage system is a simple yet carefully structured process in which patients are categorized to subsequent groups according to the severity of their presenting condition. This system aims to maximize benefits by prioritizing patients with urgent cases from non-urgent cases, as well as for the efficient use of available health resources.
Variations of triage systems are applied in different hospitals to best suit the given resource availability, economic situation, and patient capacity of each ED. The more popular variants of the triage system with good reported reliability rates include the Canadian Triage and Acuity Scale (CTAS), the Australasian Triage Scale, and the Manchester Triage System (MTS).,, While these systems may differ in details (criteria of classification and guidelines), each focuses strictly on categorizing patients in terms of urgency, thus achieving the desired results from applying the triage system. The CTAS system, for example, prioritizes patients by assigning them to a group from 1–5 (1: most severe, 5: least severe) which allows ED physicians to provide medical attention based on the patient's triage category.
Patients are stratified into several subcategories according to certain criteria in which patient presentation, vital signs, and overall patient appearance are taken into consideration. Emergency staff follow certain criteria of assessment according to the selected variant of the triage system, in which they categorize patients according to their training on the triage system and experience, which surely differs among healthcare professionals.
Variants of the triage system have been under strict scrutiny since their recent introduction to EDs. To determine a basis for statistical analysis, researchers tend to look at the level of agreement amongst healthcare providers as an indicator for validity and reliability of the implemented triage system. Measuring the inter-rater agreement has enabled health facilities to observe how well the applied triage system is implemented.
An attempt to assess the reliability and validity of the MTS in ED was made in the Netherlands in 2008. Researchers simulated 50 patient cases on paper and presented them to the emergency nurses. The nurses' evaluations were then compared to those of two MTS triage experts to measure inter-rater reliability. Results showed that inter-rater reliability was “substantial” (kappa score = 0.62) and patients were more frequently undertriaged (a lower rating than required) than overtriaged (a higher rating than required).
Another study in France in 2011 compared the concurrency of ED categorizations between nurses and physicians. Overall nurse-physician agreement was moderate and there was higher sensitivity and specificity among physicians than among nurses (94.9% vs. 89.5%, and 43.1% vs. 30.9%, respectively). The authors deduced that the lack of agreement might be the cause of significant implications concerning patient safety. Furthermore, to determine inter-rater reliability, a 2010 Canadian study  compared triage scores given to patients by base hospital nurses and emergency nurses and found a 0.50 level of agreement using the kappa statistic.
Triage systems in Saudi Arabia
In 2011, the Saudi General Directorate of Nursing, as part of the Ministry of Health, released the “Manual of Nursing Policies and Procedures.” The manual extensively analyzes the most common cases that usually present to emergency departments in Saudi Arabia. A significant number of hospitals in Saudi Arabia have chosen to adapt a formerly established triage system to aid in accurately evaluating patients that is suitable for their health settings. Hospitals invest in training programs for internationally recognized triage scales, train their nurses accordingly, and incorporate the chosen scale in their emergency department. Majority of EDs in Saudi Arabia do not have a formal triage system, thus it is unlikely that these hospital institute training programs for triage scales.,
A review of local studies discussing the topic shows that published papers mainly aim towards improving the existing triage system and trying to meet standard guidelines. King Faisal Specialist Hospital and Research Centre in 2010 tested the application of CTAS in a tertiary care hospital outside Canada. The researchers compared certain ED quality indicators with pre-established CTAS triage objectives such as time to triage assessment, triage duration, and total length of stay. Results showed that in the majority of cases, the CTAS objectives were met.
Triage in our institution
In the institution where the study was conducted, the ED abides by the rules and guidelines of the CTAS triage system. Their ED consists of four areas: triage area, resuscitation area, acute care area, and rapid assessment zone (RAZ). The triage area is where the categorization process takes place. After this initial assessment, patients proceed to one of the remaining three areas and are eventually examined by an ED physician. The triage area is usually occupied by nurses and is led by a chief nurse. Each of the other three areas requires a team of ED physicians and nurses led by a consultant/registrar. All nurses employed are required to have experience with the CTAS triage system.
The resuscitation area is where patients who fall under CTAS categories 1 and 2 are found. These patients usually arrive at the hospital in an ambulance and usually are in dire need of medical attention and continuous follow-up by emergency medicine physicians. Here the patients are resuscitated if needed, and are kept in a stable condition until they are referred to the wards to be seen by a specialist.
The acute care area is where CTAS category 3 patients are guided. These cases are mostly minor trauma cases. Here, an ED physician evaluates the patient to grant him/her a bed to receive the necessary treatment. ED physicians will then consult a specialist, and the patient is either admitted to the wards or is released if the treatment provided in the acute care area is sufficient.
The RAZ is where patients in CTAS categories 4 and 5 are usually directed. Patients here complain of relatively mild symptoms and are medically stable. The waiting times for patients to be examined by a physician are noticeably longer when compared to other areas of the emergency department. When patients enter the RAZ, they are seen by a physician, who, if needed, orders necessary lab investigations and accordingly provides the needed medical treatment. If the treatment involves monitoring the patient, medical staff assistance, or the consult of a specialist, one of the beds in the fast track area is granted to the patient.
After choosing to implement CTAS in this ED, audits of this decision have not yet been documented. One study in Quebec, Canada in 2012 questioned the validity of CTAS implementation in emergency departments. Researchers claimed that the inter-rater reliability in the CTAS system might be significantly lower than expected.,,, With an estimated 218 patients visiting per day, this has been a concern of both the administrative faculty and practicing physicians at the ED that are also in need of such an evaluative study.
| Materials and Methods|| |
This was a quantitative observational cross-sectional study. Participating emergency nurses categorised the incoming patients according to the CTAS system, and the information regarding these patients was collected without interfering with the daily routines in the ED.
Researchers collected the actual cases presenting to the ED and then summarised them into paper-based cases. These scenarios were then given to an emergency medicine consultant who is an expert in the CTAS system to be correctly categorised according to CTAS standards; this expert classification was considered as a gold standard. The paper-based scenarios were later distributed in the form of questionnaires to emergency nurses to be triaged. Investigators distributed the paper-based scenarios and collected them upon completion. The nurses were asked to categorise the cases in the paper-based scenarios to provide a mean of comparison between the nurses and the CTAS expert.
Inclusion and exclusion criteria
All patients entering the ED during the period of data collection who agreed to participate by verbal consent were included in the study. Those patients who did not provide consent were excluded from this study. Inclusion criteria regarding the nurses being evaluated for their categorisation included all those who worked at the ED at the time of the study as only 69 nurses were on duty.
Patients who refused to provide a full medical history and patients/nurses who refused to participate were excluded from data collection. Nurses who were off duty were also excluded from the study.
Sampling technique (nurses)
During the data collection period, there were 69 nurses on duty. Age was divided and grouped into decades. Training methods were classified into two groups: (1) The more common traditional paper-based training and (2) other types of training including lectures, workshops and computer-assisted training. Nursing experience was measured in years, and experience level was stratified into four categories as follows: Years of experience as a nurse, years of working experience in Saudi Arabia, years of experience in an ED and years of experience in our institution. The nurses were also categorised as those with <10 years or >10 years of experience.
Sampling technique (patients)
Due to the scarcity of local studies that aim to evaluate the implementation of triage systems from real cases that present to EDs, an expert panel of emergency physicians and CTAS experts were consulted. Fifty cases were randomly chosen to represent the most common and serious scenarios a nurse may face during his/her daily practice in the hospital where the study was conducted. All cases presenting to the ED during the time of data collection were considered. To ensure randomisation, cluster sampling was employed for collecting patient data; each day was divided into 12 2-h observation sessions (clusters). Three observation sessions per day were randomly chosen using random number generators. All patients entering the ED during that period were enrolled.
Investigators formulated paper-based scenarios based on raw data from real patients. Patient variables included sociodemographic data, medical data (chief complaint, vital signs, level of consciousness, previous admission to the hospital and chronic illnesses) and triage data (category assigned by nurse, route of arrival, time of arrival, time to triage, duration of the triage process by the nurse, total length of stay, further hospital admission and patient behaviour); all patient variables were documented based on a predesigned abstraction form and were approved by an expert panel of ED consultants with expertise in the CTAS system. The paper-based scenarios contained the following information: Age, sex, route of arrival, chief complaint, past medical history, patient behaviour and vital signs. An example of a finalised case scenario is shown in [Figure 1].
The paper-based scenarios were categorised by the expert according to the CTAS system. Data from fifty patients were collected, reviewed and summarised into paper-based scenarios. Due to a close similarity between the two cases, one was excluded, resulting in a total of 49 paper-based scenarios. The scenarios were then distributed to the nurses to be categorised. Consent, demographic data forms and scenarios were distributed during the nurses' break time. All forms were collected after completion. Results were evaluated and then compared to those of the CTAS expert categorisations. Nurse-based variables (demographic data, experience level and triage training) were also documented and considered.
The data were analysed using IBM SPSS Statistics 20 statistical software (IBM Inc., Armonk, NY, USA) and Epi Info statistical software (Center for Disease Control, Atlanta, GA, USA). The pass/fail cut-off point was defined as the median number of correctly scored cases (49 totally) achieved by the nurses. Scoring half of the cases correctly was considered as passing a passing point. Each nurse's total score and the overall pass-fail rate were evaluated for correlations with variables by means of Chi-square and odds ratio testing. P< 0.05 was considered statistically significant.
This study was approved by the International Review Board of the institution. Consent was obtained from the participants for inclusion in this study, and included all aspects of the study and all the rights of the research candidates, including their right to drop out of the study upon their request. Candidates were assured of full confidentiality by writing their information on papers that would only be seen by the investigators and evaluators. No motivational payment was provided to thr candidates.
| Results|| |
During the data collection period, a total of 69 nurses were on duty, and thus enrolled in our study; 62 of whom met the eligibility criteria. The response rate from eligible participants was 72.6% (45 nurses).
Median nurse age was 35 years, of them, 75.6% were married. Almost all nurses were females (95.6%). Indian/Pakistani nurses constituted 64.4%, while 35.6% were Filipinos. All nurses surveyed were registered nurses licensed to work in Saudi Arabia. Although the level of education was different, two-thirds had a 2-year diploma, while one-third had a bachelor's degree. Triage training is a requirement for all nurses working in the ED. Most of the nurses included in our study went through a training programme (95.6%). The duration of training varied from 4 weeks or less for one-third of the nurses to more than 4 weeks for the remaining two-thirds [Figure 2]. Results showed that the majority of nurses working in the ED were trained using paper-based training (83.3%) [Figure 3]. Experience level was determined by the number of years a nurse spent practicing his/her job. This was further stratified into four subgroups according to where each nurse practiced his/her job [Figure 4]. Full characteristics of the nurses are shown in [Table 1].
|Figure 2: Duration of training of Emergency Department nurses in relation to median score.|
Click here to view
The median score was 24, this was set as the cut-off score. Nurses who obtained a score above the cut-off were considered to have a passing score. The level of agreement between nurses' answers and the CTAS expert was calculated for each of the five CTAS categories. The highest level of agreement between inter-raters was in category 1 (62.2%), while the lowest was in category 5 (42.7%) [Figure 5]. Age was subcategorised into 10-year intervals (nurses aged between: 25-34, 35-44 and >45 years of age). Results showed that nurses above the age of 45 were more accurate in evaluating the cases given to them than their fellow nurses at 90.9% with P = 0.009. Results varied when it came to nationality as well. Almost 81.3% of all Filipino nurses included in our study scored above the median with P = 0.001. While on the other hand, only 31% of Indian and Pakistani nurses scored above the median. As the experience level varied among nurses, so did the results. Nurses with 10 or more years of experience scored significantly higher than their counterparts (78.6% and 35.5%, respectively, P = 0.007). In terms of education level, nurses with bachelor's degree scored better than those with a 2-year nursing diploma (73.3% vs. 36.7%, respectively). This variable proved to be statistically significant with P = 0.02. Results of the full Chi-square analysis are presented in [Table 2].
|Figure 5: Level of agreement; nurses and Canadian Triage and Acuity Scale expert for each Canadian Triage and Acuity Scale category.|
Click here to view
| Discussion|| |
Knowledge of the CTAS standards is a core component and was found to be an important factor for correct classification of ED patients. Other factors that influences the accuracy of triage judgment includes hours of triage training and education, level of hospital and triage mode of delivery. In this study, we assessed factors that influence nurses' classification of patients according to the CTAS guidelines. The age of the evaluating nurse was statistically significant, as nurses above 45 years of age were the most accurate amongst all nurses. This is likely reflective of nurse experience as results showed that more nurses who had 10 or more years of experience in our institution scored above the median than those with less experience. This could be due to the nature of the cases presenting to the ED on a daily basis that were reflected in the case vignettes used in our study. Experienced nurses naturally have been exposed to emergency cases that present to the institution's ED, thus, have a better sense of how urgent the case is with regards to CTAS guidelines. These findings are in contrast to those of Dallaire et al., who concluded that experienced nurses, even without recent certification, obtained a low triage score.
The education level also influenced nurses' case classification. Nurses with a bachelor's degree had nearly twice as many who scored above the median compared to those with a 2-year nursing diploma. This difference illustrates the great need for well-educated nurses EDs. To our knowledge, this is the first study showing a correlation of education level with accurate triage categorization.
Surprisingly, there was not a significant effect of triage training. This could be the result of a serious flaw in the training provided. Most received paper-based training, which, based on our results, should be reconsidered. Various training method for CTAS have previously been studied. Rankin J et al., reports that providing a structured web-learning course enhanced emergency nurses' triage accuracy. Their online training program proved successful in transferring triage learning to practice. However, evaluating nurses after fresh triage training may lead to an overestimation of the reproducibility of triage categorization later on. Furthermore, introducing various methods of training in our institution may aid in improving nurse categorization of patients.,
The percentages of overtriaged and undertriaged cases in our study were 55.93% and 44.07%, respectively. Upon comparing our results to Göransson's study, which reported values of 67.2% and 32.8% respectively, the overtriage rate was lower but the undertriage rate was higher. The evident tendency to undertriage patients in our study must be taken into consideration. Both undertriaged and overtriaged patients need to be evaluated, as suggested by Raynaud et al., in 2010, since these have effects either positively or negatively on the mortality of patients.
This study has some limitations that should be addressed. Firstly, the data represents a single-site study, which may limit the ability to generalise our results to other healthcare facilities. Conducting the study in an ED meant that the fast-paced, tense-working environment had its effects on data collection, in which allocating a designated time and place for questionnaire completion was not facilitated.
The number of CTAS category 1 patients who presented at the ED during the data collection period was limited. This restricted our ability to properly evaluate nurses regarding this category. Finally, the paper-based cases prevented the nurses from being able to see the patient, which may have affected the nurses' ability to correctly categorise the patients.
| Conclusion|| |
The overall scores of the nurses in triage categorisation were below expectations as the majority of participating nurses failed to achieve a passing score. This has important implications for EDs where the sole purpose of providing urgent patient care is based on nursing triage categorisation. Nurses' age, nationality, education level and experience all significantly affected proper categorisation. With the rapid development and expansion currently taking place in Saudi Arabia, the need for efficient EDs has increased. Resources from the Ministry of Health should aim towards improving and conducting studies on the implementation of the triage system in EDs. The scarcity of previous local literature about the triage systems is an opportunity for prospective researchers to pursue.
For future reference, there are some recommendations we suggest that may improve the overall outcome of potential studies. Allocating a set time and place during every day of the week during data collection would aid in form completion in a more effective manner. Distributing the nurses on duty into groups and distributing the data collection forms at the same time to each group would probably be a more efficient method to encourage data collection. The variability of applying different triage systems in the region may suggest an urge to include physician-based triage systems in future studies. This could possibly include a comparison of physician and nurse categorisation. Moreover, future studies should aim towards determining the most suitable method of triage training for nurses in a given institution.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Robertson-Steel I. Evolution of triage systems. Emerg Med J 2006;23:154-5.
Fitzgerald G, Jelinek G, Scott D, et al
. Emergency department triage revisited. Emerg Med J 2010;27:86-92.
Murray MJ. The Canadian Triage and Acuity Scale: A Canadian perspective on emergency department triage. Emerg Med (Fremantle) 2003;15:6-10.
Ebrahim M, Heydari A, Mazlom R, et al
. The reliability of the Australasian Triage Scale: A meta-analysis. World J Emerg Med. 2015;6:94-9.
Parenti N, Reggiani MLB, Iannone P, et al
. A systematic review on the validity and reliability of an emergency department triage scale, the Manchester Triage System. Int J Nurs Studies. 2014;51:1062-9.
Qureshi NA. Triage systems: A review of the literature with reference to Saudi Arabia. East Mediterr Health J. 2010;16:690-8.
Farrohknia N, Castren M, Ehrenberg A, et al
. Emergency department triage scales and their components: A systematic review of the scientific evidence. Scand J Trauma Resusc Emerg Med 2011;19:42-10.1186/1757-7241-19-42.
Van der Wulp I, Van Baar ME, Schrijvers, AJP. Reliability and validity of the Manchester Triage System in a general emergency department patient population in the Netherlands: Results of a simulation study. Emerg Med J 2008;25:431-4.
Durand AC, Gentile S, Gerbeaux P, et al
. Be careful with triage in emergency departments: Interobserver agreement on 1,578 patients in France. BMC Emerg Med 2011;11:19.
Dallaire C, Poitras J, Aubin K, Lavoie A, Moore L, Audet G. Interrater agreement of Canadian Emergency Department Triage and Acuity Scale scores assigned by base hospital and emergency department nurses. CJEM 2010;12:45-9.
Twomey M, Wallis LA, Myers JE. Limitations in validating emergency department triage scales. Emergency Medicine Journal 2007;24:477-9.
Aljohani M. Consistency and accuracy of emergency department triage among clinicians using standard 5-point urgency scale in one Saudi Arabian emergency department. (Unpublished master's thesis). Melbourne University, Melbourne, 2006.
Elkum NB, Barrett CA, Al-Omran, H. Canadian Emergency Department Triage and Acuity Scale: Implementation in a tertiary care center in Saudi Arabia. BMC Emerg Med 2011;11:3.
Dallaire C, Poitras J, Aubin K, Lavoie A, Moore L. Emergency department triage: do experienced nurses agree on triage scores? J Emerg Med 2012;42:736-40.
Mirhaghi A, Heydari A, Mazlom R, et al
. The reliability of the Canadian triage and acuity scale: meta-analysis. N
Am J Med Sci 2015;7:299-305.
Gravel J, Gouin S, Manzano S. Interrater agreement between nurses for the pediatric Canadian Triage and Acuity Scale in a tertiary care center. Acad Emerg Med 2008;15:1262-7.
Considine J, Botti M, Thomas S. Do knowledge and experience have specific roles in triage decision-making? Acad Emerg Med 2007;14:722-6.
Chen SS, Chen JC, Ng CJ, et al
. Factors that influence the accuracy of triage nurses; judgement in emergency departments. BMJ Emerg Med 2010;27:451-455
Rankin JA, Then KL, Atack L. Can emergency nurses' triage skills be improved by online learning? Results of an experiment. J Emerg Nurs 2013;39:20-6.
Hadley N. Triage: Meeting the needs of today in a busy ED. Adv Emerg Nurs J 2005;27:217-22.
Loke SS, Liaw SJ, Tiong LK, et al
. Evaluation of nurse-physician inter-observer agreement on triage categorization in the Emergency Department of a Taiwan medical center. Chang Gung Med J 2002;25:446-51.
Göransson K, Ehrenberg A, Marklund B, Ehnfors M. Accuracy and concordance of nurses in emergency department triage. Scand J Caring Sci 2005;19:432-8.
Raynaud L, Borne M, Coste S, et al
. Triage protocol: Both undertriage and overtriage need to be evaluated. J Trauma Inj Inf Crit Care 2010;69:998.
Hupert N, Hollinhsworth E, Xiong W. Is overtriage associated with increased mortality? Insights from a simulation model of mass casualty trauma care. Disaster Med Pub Health Prep 2007;1:514-24.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5]
[Table 1], [Table 2]