Objective: To compare the degree to which the Emergency Severity Index (ESI) and the Manchester Triage System (MTS) predict admission and mortality.
Methods: A retrospective observational study of four emergency department (ED) databases was conducted. Patients who presented to the ED between 1 January and 18 July 2006 and were triaged with the ESI or MTS were included in the study.
Results: 37 974 patients triaged with the ESI and 34 258 patients triaged with the MTS were included. The likelihood of admission decreased significantly with urgency categories in both populations, and was greater for patients triaged with the ESI than with the MTS. Mortality rates were low in both populations. Most patients who died were triaged in the most urgent triage categories of both systems.
Conclusion: Both the ESI and MTS predicted admission well. The ESI was a better predictor of admission than the MTS. Mortality is associated with urgency categories of both triage systems.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
In recent years triage systems have frequently been used in emergency departments (EDs). In the Netherlands, the Dutch Institute for Healthcare Improvement developed a guideline in 2004 that advised all EDs in the Netherlands to implement the Manchester Triage System (MTS). Creators of the guideline preferred the MTS because it is not diagnosis-based and therefore is particularly applicable for use by nurses.1 After publication of the guideline, EDs started implementing the MTS and, at the end of 2006, 87% of hospitals that used a triage system had implemented the MTS while 13% used the Emergency Severity Index (ESI).2
The MTS was developed by the Manchester Triage Group and consists of 52 flowcharts, each presenting a complaint. Urgency of the patient’s problem is assessed by discriminators presented in each flowchart. Every patient is classified in one of five urgency categories: red (needs to see a doctor immediately); orange (can wait 10 min); yellow (can wait 1 h), green (can wait 2 h); and blue (can wait 4 h).3
In contrast to the MTS, the ESI consists of one flowchart and patient urgency is assessed with four conceptual key questions. Answers to these four questions will lead to one of five urgency categories: 1 (requires immediate live-saving intervention); 2 (a high-risk situation); 3 (patient needs two or more resources); 4 (patient needs one resource); and 5 (no resources are needed). Resources are defined as laboratory tests, radiology, intravenous fluids, specialty consultation, a simple or complex procedure and intravenous/intramuscular/nebulised medications.4
A major question for triage systems is validity. Previously conducted studies were mainly concentrated on measuring relationships between urgency and certain outcome variables such as admission, length of stay and mortality. These relationships were studied in an attempt to find evidence for the construct validity of triage systems. It is not possible to assess criterion validity as a gold standard for triage urgency is absent.5
Associations between urgency and admission have been reported in eight studies, of which six studied the ESI,6–11 one the MTS12 and one the Australasian Triage Scale.13 Associations with length of stay in the ED and resource use were only found in the ESI.6 8 9 11 14 15 Associations between urgency categories of triage systems and hospital length of stay, mortality and survival have been studied less frequently. Hospital length of stay was not associated with the ESI, but 6-month survival was.15 16 The ESI and Australasian Triage Scale showed associations with mortality.10 17 One study measured the predictive ability of two triage systems—the ESI and the Canadian Triage and Acuity Scale. Admitted patients and patients who died were allocated to a significantly higher urgency category than patients who were not admitted or did not die. There were no significant differences between the ability of the two triage systems to predicting admission and in-hospital deaths.18
In our opinion, the above-mentioned studies are limited by the fact that a large number of studies have focused on specific subgroups of patients (eg, elderly patients, self-referred patients, children or patients aged >14 years). Also, none of these studies adjusted for variables that could influence the outcomes. A study was therefore undertaken to compare the associations between urgency categories of the ESI and the MTS to predict hospital admission and mortality in EDs in the Netherlands.
Study setting and population
A retrospective observational study of ED databases was conducted in four hospitals in the Netherlands. The Onze Lieve Vrouwe Gasthuis (Amsterdam) and Sint Elisabeth Hospital (Tilburg) implemented the ESI in 2003 and 2004, respectively, and the Haaglanden Medical Center (The Hague) and Meander Medical Center (Amersfoort) implemented the MTS in 2004 and 2005, respectively. The EDs of the participating hospitals have an annual number of patients ranging from 30 000 to 47 000. All nurses who triage with the ESI were trained in the use of the system; nurses at the Onze Lieve Vrouwe Gasthuis were trained and tested by one of the founders of the system19 and, at the Sint Elisabeth Hospital, nurses followed a 1-day course of training by an expert. Nurses who triage with the MTS were trained by in-company training sessions following the guidelines.1 In 2007 a reliability study was conducted in two hospitals that triage with the MTS, including the Meander Medical Center. Inter-rater agreement in this study was substantial (quadratically weighted kappa 0.62).20
All patients who presented at the ED and were triaged with the MTS or ESI between 1 January and 18 July 2006 were included in the study.
The ED databases of the four participating hospitals were used to collect information about triage category, gender, age, admission and mortality of all patients. Hospitalised patients included patients who were admitted to a hospital ward or who were transferred to a ward of another hospital. Mortality was defined as all patients who died in the ED. Patients who were pronounced dead on arrival were excluded.
The influence of the ESI and MTS on the prediction of admission was assessed by logistic regression analyses. Univariate logistic regression analyses were performed on age, gender, hospital and urgency for each triage system. All variables which significantly (p<0.05) predicted admission were selected for multivariate analyses.
Because mortality rarely occurred in the ED, associations were assessed with the Fisher exact test. This test measures whether the probability of dying is equal in all triage categories. Data were analysed using SPSS for Windows Version 14.
A total of 38 330 patients presented at the ED of hospitals triaging with the ESI and 46 537 patients presented to the hospitals using the MTS. Of these, 356 patients (0.9%) were excluded because of missing ESI triage categories and 12 279 (26.4%) because of missing MTS urgency categories. A total of 37 974 patients triaged with the ESI and 34 258 patients triaged with the MTS were therefore included in the study.
The mean (SD) age of patients triaged with the ESI and MTS was 38.7 (23.6) years and 42.4 (23.5) years, respectively. In both patient populations more than half of the patients who presented to the ED were male (56.0% ESI and 51.6% MTS). Most of the patients triaged with the ESI were put in the third (33.0%) or fourth (32.6%) category (fig 1); relatively few patients were triaged in categories 1 (0.5%) and 2 (11.1%). Patients triaged with the MTS were mostly allocated to the yellow (37.7%) or green (44.5%) categories; the red and blue categories rarely occurred (0.6% and 0.8%, respectively).
In both triage systems, triaged patients were admitted more often than patients with missing triage scores (p<0.05). There was no significant difference in the mortality rates between the two groups (p = 0.59). Patients with missing triage scores in hospitals triaging with the MTS were significantly younger in both populations (p<0.01) and were more often male (p<0.01). In one hospital that triaged with the MTS, patients with missing triage scores were significantly (p<0.01) more often referred by the general practitioner or ambulance services. In the other hospitals, this information was not available.
The frequency of admission was almost equal in both populations. Of patients triaged with the ESI, 21.6% were admitted compared with 21.0% triaged with the MTS. In both systems, the majority of admitted patients were triaged in the middle category (table 1). Patients in the more urgent categories were more likely to be admitted (fig 2). Remarkably, 158 (1.9%) patients triaged with the ESI were admitted, while it was estimated that no resources were needed for these patients. In patients triaged with the MTS, 10 (0.1%) were admitted while it was estimated that they could wait up to 4 h in the waiting room.
In univariate logistic regression analyses, age and hospital were significant predictors for admission (p<0.01). Gender was only a significant predictor in the ESI population, with women having a greater likelihood of admission (p<0.01). The results of the multivariate logistic regression analyses are presented in table 2. The likelihood of admission decreased substantially with urgency in both populations. Also, for each category in both triage systems, no overlap was seen in 95% confidence intervals. The likelihood of admission for patients triaged with the ESI was greater than with the MTS. In the fifth triage category, the likelihood of admission for patients triaged with the ESI was almost three times lower than the reference category. The same effect was seen in the MTS, but this result appeared to be non-significant, possibly because only nine patients were admitted. In both triage systems, substantial differences were found between hospitals in the degree to which they predicted admission.
Twenty-eight patients triaged with the ESI and 29 patients triaged with the MTS died in the ED. Of the patients who died, 23 (82.1%) were triaged in ESI category 1 and 22 (75.9%) were triaged in MTS category red (table 3). A Fisher exact test revealed significant associations (p<0.01) between the urgency categories of both triage systems and mortality. Patients in the more urgent categories were at greater risk of dying. However, because of the small mortality rates, it is not possible to calculate the strength of the associations.
The results of this study showed that urgency, as defined in both triage systems, was a very good predictor of admission and proved to be closely related to ED mortality. A limiting factor for triage research is the fact that there is no gold standard for measuring triage urgency. We therefore studied the predictive ability of urgency categories of the ESI and MTS for hospital admission and ED mortality as an indicator of construct validity. By studying this relationship, it was expected that the urgency categories of the ESI and MTS would be associated with these patient outcomes. However, this relation will not be perfect as some patient categories—such as patients with severe pain—will be triaged in the more urgent categories, with or without subsequent hospitalisation. Also, the patient’s condition may deteriorate in the period between triage assessment and medical treatment, which can lead to unexpected patient outcomes.
Our data showed that admission decreased with urgency. The confidence intervals of the urgency categories predicting admission did not overlap, adding support to the discriminative power of the two systems. The ESI seems to be a better predictor of admission than the MTS, possibly because triage nurses using the ESI estimate the number of resources the patient needs. The number of resources is therefore likely to be associated with hospital admission.
To our knowledge, this is the first study to measure the predictive capacity of two triage systems by means of logistic regression analyses. In both triage systems, significant differences were found between hospitals in the degree to which they predict admission. It is possible that different admission policies may have accounted for these differences. It is therefore important that future studies should be conducted in several hospitals so that adjustment can be made for these influences. Previous studies have not adjusted for these factors and have mainly reported the percentage of admitted patients per triage category. Roukema et al12 found an increased rate of hospital admissions with increased urgency in the MTS. Patients triaged in the blue category were less likely to be admitted (0.9%) than those triaged in the red category (53.5%). Although the percentage of patients admitted per triage category of the MTS was higher in our study than that reported by Roukema et al, our study showed the same effect. Differences in study populations could explain these differences, because age was a significant predictor of urgency. The percentage of patients admitted per triage category in the ESI was also comparable to other studies. Tanabe et al7 and Wuerz et al9 16 reported that 80–92% of patients in the first category were admitted and 0–5% in the last category. Again, the differences could be explained by differences in the study populations and study setting.
The probability of dying in the ED is very small, but was significantly associated with urgency categories in both triage systems. Most patients who died in the ED were triaged in the highest urgency category. None of the patients who died were triaged in the two least urgent categories. A comparable effect has been reported by Wuerz et al,16 who also followed up for a period of 6 months after the ED visit. In that study, 32% of patients triaged in the first category died within 6 months of their visit to the ED. No patients died in the ED. The difference in design could explain the difference in mortality rate in our study.
Overall, compared with our study, several other studies have reported more patients in the second triage category of the ESI and fewer patients in the fifth category.6 9 11 15 The above-mentioned explanations, such as differences between hospitals, could also explain differences in reported case mix of the ESI.
Limitations of the study
Our study has some limitations and therefore the results should be interpreted carefully. First is the retrospective character of the study. Because information was registered for other purposes than this study, it is possible that not all information was entered correctly. This could have influenced our results. To minimise the limitations of retrospective studies, we have selected a study period in which no organisational changes in the ED occurred. A second limitation is the large number of missing triage scores from hospitals that triage with the MTS. It is possible that patients who arrived by ambulance were frequently not triaged because they needed care immediately. Furthermore, during a busy period in the ED, the triage nurse may stop registering triage scores or stop triaging and start helping colleagues. In one hospital, patients with missing triage scores were more often referred. This could indicate that triage nurses do not triage these patients because they had already been seen by a professional. The missing data could have influenced the case mix and therefore affected the associations between the MTS and admission.
Both the ESI and MTS strongly predicted admission. The ESI was found to be a better predictor of admission than the MTS. Mortality was associated with urgency categories in both triage systems. However, further study in a different design is needed to measure the strength of the association. Future studies measuring the association between triage systems and other parameters need to adjust for influencing factors.
The authors thank Sint Elisabeth Hospital Tilburg, The Netherlands; Onze Lieve Vrouwe Gasthuis Amsterdam, The Netherlands; Meander Medical Center Amersfoort, The Netherlands; and the Haaglanden Medical Center, The Hague, The Netherlands.
Competing interests: None.
Ethics approval: The protocol was approved by the Medical Ethical Committee of the University Medical Center Utrecht.