Farrohknia et al.

Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine 2011, 19:42

http://www.sjtrem.com/content/19/1/42

REVIEW Open Access

Emergency Department Triage Scales and Their

Components: A Systematic Review of the
Scientific Evidence
Nasim Farrohknia1*, Maaret Castrén2, Anna Ehrenberg3, Lars Lind4, Sven Oredsson5, Håkan Jonsson6, Kjell Asplund7
and Katarina E Göransson8,9

Abstract
Emergency department (ED) triage is used to identify patients’ level of urgency and treat them based on their
triage level. The global advancement of triage scales in the past two decades has generated considerable research
on the validity and reliability of these scales. This systematic review aims to investigate the scientific evidence for
published ED triage scales. The following questions are addressed:
1. Does assessment of individual vital signs or chief complaints affect mortality during the hospital stay or within
30 days after arrival at the ED?
2. What is the level of agreement between clinicians’ triage decisions compared to each other or to a gold
standard for each scale (reliability)?
3. How valid is each triage scale in predicting hospitalization and hospital mortality?
A systematic search of the international literature published from 1966 through March 31, 2009 explored the British
Nursing Index, Business Source Premier, CINAHL, Cochrane Library, EMBASE, and PubMed. Inclusion was limited to
controlled studies of adult patients (≥15 years) visiting EDs for somatic reasons. Outcome variables were death in
ED or hospital and need for hospitalization (validity). Methodological quality and clinical relevance of each study
were rated as high, medium, or low. The results from the studies that met the inclusion criteria and quality
standards were synthesized applying the internationally developed GRADE system. Each conclusion was then
assessed as having strong, moderately strong, limited, or insufficient scientific evidence. If studies were not
available, this was also noted.
We found ED triage scales to be supported, at best, by limited and often insufficient evidence.
The ability of the individual vital signs included in the different scales to predict outcome is seldom, if at all,
studied in the ED setting. The scientific evidence to assess interrater agreement (reliability) was limited for one
triage scale and insufficient or lacking for all other scales. Two of the scales yielded limited scientific evidence, and
one scale yielded insufficient evidence, on which to assess the risk of early death or hospitalization in patients
assigned to the two lowest triage levels on a 5-level scale (validity).

Introduction there is no queue for care. Triage scales aim to optimize

Triage is a central task in an emergency department the waiting time of patients according to the severity of
(ED). In this context, triage is viewed as the rating of their medical condition, in order to treat as fast as
patients’ clinical urgency [1]. Rating is necessary to iden- necessary the most intense symptom(s) and to reduce
tify the order in which patients should be given care in the negative impact on the prognosis of a prolonged
an ED when demand is high. Triage is not needed if delay before treatment. ED triage is a relatively modern
phenomenon, introduced in the 1950s in the United
States [2]. Triage is a complex decision-making process,
* Correspondence: [email protected] and several triage scales have been designed as decision-
1
The Swedish Council for Health Technology Assessment and Dep of
Medical Sciences, Uppsala University Hospital, Uppsala, Sweden support systems [3] to guide the triage nurse to a
Full list of author information is available at the end of the article

© 2011 Farrohknia et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative
Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and
reproduction in any medium, provided the original work is properly cited.
Farrohknia et al. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine 2011, 19:42 Page 2 of 13
http://www.sjtrem.com/content/19/1/42

correct decision. Triage decisions may be based on both 2. In adult ED patients, what is the level of agree-
the patients’ vital signs (respiratory rate, oxygen satura- ment between clinicians’ triage decisions compared
tion in blood, heart rate, blood pressure, level of con- to each other or to a gold standard for each scale (i.
sciousness, and body temperature) and their chief e. the reliability of triage scales)?
complaints. Internationally, no consensus has been 3. In adult ED patients, how valid is each triage scale
reached on the functions that should be measured. in predicting hospitalization and hospital mortality?
Apart from emergency care, triage may be used in other
clinical activities, e.g. deciding on a certain investigation Methods
[4] or treatment [5]. A systematic search of the international literature pub-
Since the early 1990s, several countries have devel- lished from 1966 through March 31, 2009 explored the
oped and introduced ED triage [6-10]. Development of British Nursing Index, Business Source Premier,
triage scales in some countries has been influenced lar- CINAHL, Cochrane Library, EMBASE, and PubMed.
gely by the seminal work of FitzGerald [11], resulting Inclusion was limited to studies of adult patients (≥15
in most of the triage scales developed in the 1990s and years) visiting EDs for somatic reasons. Another criter-
2000s being designed as 5-level scales. Of these, the ion for inclusion was that the study design must contain
Australian Triage Scale (ATS), Canadian Emergency a control, i.e. randomized controlled trials (RCT), obser-
Department Triage and Acuity Scale (CTAS), Manche- vational studies with a control group based on pre-
ster Triage Scale (MTS), and Emergency Severity Index viously collected data, and before-after studies.
(ESI) have had the greatest influence on modern ED Descriptive studies without a control group and retro-
triage [12-15]. Other scales have not disseminated as spective studies were excluded.
widely around the globe, e.g. the Soterion Rapid Triage
Scale (SRTS) from the United States and the 4-level Inclusion criteria for vital signs and chief complaints used
Taiwan Triage System (TTS) [6,7,9,16,17]. Some coun- in triage scales
tries, e.g. Australia, have a national mandatory triage • Studies analyzing individual vital signs or chief
scale while many European countries lack such stan- complaints
dards [7,9]. • Outcome variable defined as death within 30 days
Patients may have a life-threatening condition, but after ED arrival or during the hospital stay
show normal vital signs. Hence, in triaging the patient it
is important to consider information given by patients
or accompanying persons regarding the patient’s chief Inclusion criteria for reliability and validity of triage scales
complaints or medical history, which can provide essen- • Studies based on real patients triaged at EDs
tial information about serious diseases. The chief com- (validity)
plaints describe the incident or symptoms that caused • Studies based on real patients triaged at EDs or fic-
the patient to seek care. titious patient scenarios (reliability)
In 2005, a joint task force of the American College of • Studies reporting reliability at separate triage levels
Emergency Physicians and the Emergency Nurses Asso- (reliability)
ciation published a review of the literature on ED triage • Studies reporting mortality and hospitalization per
scales. Based on expert consensus and available evi- triage level (validity)
dence, the task force supported adoption of a reliable 5- • Outcome variables defined as death in the ED or
level triage scale, stating that either the CTAS or the hospital, and need for hospitalization (validity)
ESI are good choices for ED triage [18]. In 2002, a
national survey conducted in Sweden identified the use
of 37 different triage scales across the country. Further, Exclusion criteria for studies on reliability of triage scales
some 30 EDs did not use any type of triage scale [19]. • Studies on interrater reproducibility are excluded
This systematic review aims to investigate the scienti- in cases where any rater in the study had access to
fic evidence underlying published ED triage scales. retrospective data only.

Objectives Six experts from different professions and clinical spe-

The following questions are addressed: cialties reviewed the studies, independently in groups of
2 or 3, for quality by using methods validated for inter-
1. In triage of adults at EDs, does assessment of indi- nal validity, precision, and applicability (external validity)
vidual vital signs or chief complaints affect mortality [20]. The methodological quality and clinical relevance
during the hospital stay or within 30 days after arri- of each study was graded as high, medium, or low.
val at the ED? Results from the studies that met the inclusion criteria
Farrohknia et al. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine 2011, 19:42 Page 3 of 13
http://www.sjtrem.com/content/19/1/42

and quality standards were synthesized by applying the All of the studies were found to have medium quality
internationally developed GRADE system [21]. and relevance. Only a few studies included all patients
In accordance with GRADE, the following factors were (albeit limited to “medical” patients”) that arrived at the
considered in appraising the overall strength of the evi- ED, regardless of diagnosis. Hence, studies of patients
dence: study quality, concordance/consistency, transfer- classified as surgical disciplines were generally lacking.
ability/relevance, precision of data, risk of publication Several studies described compiled scales or indexes for
bias, effect size, and dose-response. In synthesizing the appraising the severity level of the patient’s conditions,
data, studies having low quality and relevance were but provided no information on the importance of spe-
included when studies of medium quality and relevance cific vital signs or chief complaints. Hence, little or no
were not available. Based on the overall quality and rele- evidence can be found on the association between speci-
vance of the studies reviewed, each conclusion was rated fic vital signs or reasons for the ED visit and mortality
as having strong, moderately strong, limited, or insuffi- in the group of general patients presenting in EDs.
cient scientific evidence. If studies were not available, Respiratory rate
this was noted [21]. Only a single study, which described the predictive
importance of respiratory rate, fulfilled the inclusion cri-
Results teria [22]. The study aimed to assess whether the Rapid
Figures 1 and 2 illustrate the results of the primary Acute Physiology Score (RAPS) could be used to predict
search. mortality in nonsurgical patients on ED arrival. It also
aimed to study whether an advanced version of RAPS, i.
Vital signs and chief complaints e. the Rapid Emergency Medicine Score (REMS), could
Most of the studies that investigated associations yield better predictive information [22].
between different vital signs or chief complaints and RAPS was developed for prehospital care and involves
mortality after ED arrival were observational cohort stu- assessing respiratory rate, pulse, blood pressure, and the
dies based on selected, diagnosis-specific, patient groups. Glasgow Coma Scale (GCS). REMS is based on RAPS,

Abstracts identified
through database
seaching
4 185 Abstracts excluded
by relevance
4 096
Articles studied
in full text
89
Articles excluded
Articles identified through by relevance,
other sources study design and
10 non-sufficient
Articles included in systematic eligibility
review 95
4

Low quality Medium quality High quality

1 3 0

Figure 1 Results of literature search and selection process.

Farrohknia et al. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine 2011, 19:42 Page 4 of 13
http://www.sjtrem.com/content/19/1/42

Abstracts identified
through database
seaching
2 776 Abstracts excluded
by relevance
2 608
Articles studied
in full text
168
Articles excluded
Articles identified through by relevance,
other sources study design and
1 non-sufficient
Articles included in systematic eligibility
review 149
20

Low quality Medium quality High quality

11 9 0

Figure 2 Results of literature search and selection process regarding reliability (10 articles), and validity (10 articles) of triage scales.
One article studied both reliability and validity and was rated differently due to the studied endpoint, low quality regarding reliability and medium
quality regarding validity.

but also assesses oxygen saturation, body temperature, in a group of 11 751 patients receiving care for nonsur-
and age. In total, 11 751 patients were studied pro- gical disorders. With a decrease of one step on the
spectively after arrival at the ED of a university hospi- RAPS scale, 67% of the patients showed an increased
tal in Sweden. Respiratory rate was found to be a risk of mortality within 30 days.
significant predictor of mortality during the hospital Level of consciousness
stay. A decrease of one step on the RAPS scale was The Swedish study (described above) also investigated
found to nearly double the risk of mortality within 30 the association between acute mortality and the level of
days (Table 1). consciousness on arrival at the ED [22]. Another study
Oxygen saturation in blood used the same methods mentioned above, i.e. RAPS and
Two studies used RAPS and REMS to predict acute REMS [23], to analyze 5583 patients that had called the
mortality after ED arrival and specifically studied the emergency phone number and were classified as urgent.
predictive importance of saturation [22,23]. Oxygen The study showed that level of consciousness was one of
saturation was found to be one of the three variables, three variables (age and saturation being the other two)
along with age and level of consciousness, that best pre- that best predicted mortality during the hospital stay.
dicted mortality during hospitalization. Another study analyzed 986 stroke patients on ED arri-
Pulse val. Impaired level of consciousness appeared to be the
One study investigated the importance of assessing pulse best predictor of mortality during the hospital stay [24].
in the ED as a means to predict mortality during the Blood pressure and body temperature
hospital stay. The importance of blood pressure or body temperature
The study, which was conducted in Sweden [22], in assessing the risk of acute mortality after ED arrival
showed a significant association between the pulse on could not be supported by the included studies due to
arrival to the ED and mortality during the hospital stay the lack of scientific evidence.
Farrohknia et al. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine 2011, 19:42 Page 5 of 13
http://www.sjtrem.com/content/19/1/42

Table 1 Does assessment of certain vital signs and chief complaints in emergency department triage of adults have an
impact on 30-day or in-hospital mortality?
Author Study design Patient characteristics Primary Outcome Frequency RR Missing data (%) Study
Year, Sample Female/age Male/ outcome (relative risk), OR (odds quality and
reference age Inclusion criteria Type ratio) P-value, 95% CI relevance
Country of emergency department (confidence interval) Comments
Goodacre Observational Emergency medical Mortality Age, Glascow Coma Scale Rapid Acute Physiology Moderate
S et al Cohort admissions, life threatening in (GCS) and oxygen saturation Score (RAPS - blood
2006 [23] Retrospective category A emergency calls hospital independent predictors of pressure, pulse, GCS, RR, Acceptable
United database during mortality in multivariate saturation and temp) in external
Kingdom review N = 5 583 the stay analysis, blood pressure is not only 3 624 (64.9%). Missing validity
Female: 2 350 (42.3%) useful in 35.1% Good/
Male: 3 233 (57.7%) acceptable
Mean age 63.4 years Glascow Coma Scale (GCS): Rapid Emergency Medicine internal
OR 2.10 (95% CI 1.86-2.38) p Score (REMS - Blood validity
Inclusion criteria: Any case < 0.001 pressure, pulse, GCS, RR) in
where caller report chest pain, only 2 215 (39,7%). Missing Age, GCS and
unconsciousness, not Age: OR 1.74 (95% CI 1.52- in 60.3%. saturation
breathing and patient 1.98) p < 0.001 independent
admitted to hospital or died in New Score (GCS, predictors of
emergency department (ED) Saturation: OR 1.36 (95% CI saturation, age) in 2 743 mortality.
1.13-1.64) p = 0.001 (49.1%). Missing in 50.9% Blood
Setting: variables recorded on pressure is
ambulance arrival not a useful
predictor
Olsson T Observational Nonsurgical emergency Mortality In-hospital mortality 2.4%, Moderate
et al cohort department (ED) patients in mortality within 48 hours Good internal
2004 [22] Prospective hospital, 1.0%. validity
Sweden n = 11 751 within 48
Female: 51.6% hours Predictors for mortality:
Male: 48.4% Saturation OR: 1.70 (95% CI:
Mean age 61.9 (SD ± 20.7) 1.36-2.11) p < 0.0001
Respiratory frequency OR:
Inclusion criteria: Patients 1.93 (95% CI: 1.37-2.72)
consecutively admitted to the p < 0.0002
emergency department (ED) Pulse frequency OR 1.67
over 12 months. (95% CI 1.36-2.07) p < 0.0002
Exclusion criteria: Patients Coma OR: 1.68 (95% CI:
with cardiac arrest that could 1.38-2.06) p < 0.0001
not be resuscitated, patients Age OR: 1.34 (95% CI:
with more than one parameter 1.10-1.63) p < 0.004
missing.
Setting: 1 200 bed University
hospital ED in Sweden
Han JH et Observational Suspected acute coronary Mortality 2.7% in-hospital mortality for Missing data for ECG, Low
al 2007 cohort syndrome (ACS) in- patients age ≥75 years, symptoms or gender in 1
[25] Retrospective hospital/ higher 30 day mortality 810 (15.2%) Convenience
USA database n = 10 126 within 30 (Adjusted OR: 2.6, 95% CI: sample-
Singapore review Female: 5 635 days 1.6-4.3) selection bias
Comparison Male: 4 491 Confounders,
patients ≥/≤ Mean age = ? such as co-
75 years 11.4% ≥75 years morbidity not
described
Inclusion criteria: ≥ age 18,
suspected ACS verified by Acceptable
electrocardiogram (ECG), intern validity
cardiac biomarkers, dyspnoea,
light-headedness, dizziness
and weakness.
Exklusion criteria: Inter-
hospital transfer, if missing
data concerning gender, age
or clinical presentation
Setting: 8 emergency
departments (ED) (USA), 1 ED
(Singapore)
Farrohknia et al. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine 2011, 19:42 Page 6 of 13
http://www.sjtrem.com/content/19/1/42

Table 1 Does assessment of certain vital signs and chief complaints in emergency department triage of adults have an
impact on 30-day or in-hospital mortality? (Continued)
Arboix A Observational Stroke Mortality Overall mortality 16.3%. Not stated Moderate
et al cohort in-hospital Age OR: 1.05 (95% CI:
1996 n = 986 1.03-1.07), previous or
[24] Female: 468 concomitant Pathologic
Spain Male: 518 conditions OR: 1.83 (95% CI:
Mean age = ? 1.19-2.82)
Deteriorated level of
Inclusion criteria: First-ever Consciousness OR: 11.70
stroke, admitted to hospital. (95% CI: 7.70-17.77)
Vomiting OR: 2.18 (95% CI:
Setting: Department of 1.20-3.94)
neurology, university hospital Cranial nerve palsy OR: 2.61
(95% CI: 1.34-5.09)
Seizures OR: 5.18 (95% CI:
1.70-15.77) and
Limb weakness OR: 3.79
(95% CI: 1.96-7.32) were
independent prognostic
factors of in-hospital mortality

Chief complaints Interrater agreement of triage scales (reliability)

Studies describing the association between different All 11 articles that were found to answer the question con-
chief complaints and acute mortality were found to be cerning reliability of triage scales and met the defined
lacking. inclusion criteria were observational studies. They
Age addressed reliability of the ATS [26], CTAS (including
Three of the studies described above showed that the eTriage) [19,27-30], MTS [31], SRTS [6], and two locally
higher the patient’s age, the greater the risk of death produced scales without names [8,32] (Table 3). Based on
within 30 days of hospital care following ED arrival the quality review, 9 articles [6,8,19,26-31] were found to
[22-24]. The results showed an increase in mortality of be of low and 1 [32] of medium quality. One article was
5% per year. Furthermore, one study showed that older excluded due to deficient quality resulting from high inter-
patients (above 75 years of age) with symptoms of cor- nal dropout [16]. Deficient external validity was the major
onary heart disease had a greater risk of death within 30 reason for the low- and medium-quality ratings of the stu-
days after arrival at the ED compared to younger dies. Selection of patients and triage nurses were both
patients with the same symptoms [25] (Table 1). found to be irrelevant or insufficiently described. Hence,
Based on the studies described above, Table 2 sum- 10 articles remained as a basis for the conclusions.
marizes assessments and comments regarding the level The scientific evidence was found to be insufficient to
of scientific evidence. assess the reliability of ATS, CTAS, MTS, SRTS and the

Table 2 Appraisal of scientific evidence according to GRADE - Association between vital signs/chief complaints and
acute mortality after arrival at the emergency department.
Effect measure (endpoint) No. Patients (no. Effect (OR, Scientific Comments
Studies) Reference odds ratio*) evidence
Respiratory rate predicts 30-day mortality 11 751 1.9 Insufficient Only one study (-1)
1 study [22] ⊕○○○
Oxygen saturation predicts 48-hour mortality or 17 334 1.4 Limited
in-hospital mortality 2 studies [22,23] 1.7 ⊕⊕○○
Pulse predicts 30-day mortality 11 751 1.7 Insufficient Only one study (-1)
1 study [22] ⊕○○○
Level of consciousness predicts 48-hour 18 320 2.1 Limited
mortality or in-hospital mortality 3 studies [22-24] 1.7 ⊕⊕○○
11.7
Age predicts 30-day mortality 28 446 1.7 Moderate Upgrading due to effect size and
4 studies [22-25] 1.3 ⊕⊕⊕○ dose-response effect (+1)
2.6
1.1
All studies are observational.
* OR indicates each step of change in RAPS (Rapid Acute Physiology Score) or REMS (Rapid Emergency Medicine Score).
Farrohknia et al. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine 2011, 19:42 Page 7 of 13
http://www.sjtrem.com/content/19/1/42

Table 3 Reliability of triage scales

Author Year, Triage Patient characteristics: Age Results: -values, Drop Study quality and relevance
reference system Gender Triageur: Amount, percentage agreement out (%)
Country profession (PA)/triage level
Considine J ATS 10 scenarios Triage level: 0% Low
et al 31 RNs 1: 59.7% PA
2000, [26] 2: 58% PA External validity is uncertain, internal validity is
Australia 3: 79% PA good while sample size is of uncertain
4: 54.8% PA adequacy
5: 38.7% PA
Dong S et al ETriage 569 patients 0.40 (unweighted ) 1% Low
2006, [28] (CTAS) 49.4 years Triage level:
Canada 51 % male 1: 62.5% PA External validity can not be assessed, internal
Unknown amount of RNs 2: 49.5% PA validity is excellent while sample size is of
3: 59.7% PA uncertain adequacy
4: 68.5% PA
5: 43.5% PA
Dong S et al CTAS/ 693 patients 0.202 (unweighted ) 4% Low
2005, [29] eTriage 48 years Triage level:
Canada 49 % male 1: 50% PA External validity can not be assessed, internal
73 RNs 2: 9% PA validity is excellent while sample size is of
3: 53.5% PA uncertain adequacy
4: 73.3% PA
5: 7.2% PA
Manos D et al CTAS 42 scenarios 0.77 overall (weighted ) 0.2% Low
2002, [30] 5 BLS BLS: 0.76 (weighted )
Canada 5 ALS ALS: 0.73 (weighted ) External validity can not be assessed, internal
5 RNs RNs: 0.80 (weighted ) validity is acceptable while sample size is of
5 Drs Drs: 0.82 (weighted ) uncertain adequacy

Triage level:
1: 78% PA
2: 49% PA
3: 37% PA
4: 41% PA
5: 49% PA
Beveridge R CTAS 50 scenarios 0.80 overall (weighted ) 15% Low
et al 10 RNs 0.84 RNs (weighted )
1999, [27] 10 Drs 0.83 Drs (weighted ) External validity can not be assessed, internal
Canada validity is acceptable while sample size is of
Weighted  / triage level uncertain adequacy
(RNs):
Triage level:
1: 0.73
2: 0.52
3: 0.57
4: 0.55
5: 0.66
Göransson K CTAS 18 scenarios 0.46 (unweighted ) 0.8% Low
et al 423 RNs Triage level:
2005, [19] 1: 85.4% PA External validity can not be assessed, internal
Sweden 2: 39.5% PA validity is acceptable while sample size is of
3: 34.9% PA uncertain adequacy
4: 32.1% PA
5: 65.1% PA
van der Wulp I MTS 50 scenarios 0.48 (unweighted ) 7.5-35.7% Low
et al 55 RNs Triage level:
2008, [31] 2: 9.8% PA External validity is uncertain, internal validity is
The 3: 35.5% PA good while sample size is of uncertain
Netherlands 4: 22% PA adequacy
Maningas P SRTS 423 patients 0.87 (weighted ) Low
et al 29.7 years Triage level:
2006, [6] 44% male 1: 85.7% PA External validity can not be assessed, internal
USA 16 RN pairs 2: 86.7% PA validity is good while sample size is of uncertain
3: 86.8% PA adequacy
4: 93.9% PA
5: 74.2% PA
Farrohknia et al. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine 2011, 19:42 Page 8 of 13
http://www.sjtrem.com/content/19/1/42

Swiss scale (Table 4). However, limited scientific evi- Discussion

dence was found in assessing the reproducibility of the Our systematic review shows that when adjudicated by
Brillman scale (North America) as having moderate standard criteria for study quality and scientific evi-
interrater agreement. dence, the triage scales used in EDs are supported, at
best, by limited evidence. Often, the evidence is weaker,
Validity of triage scales regarding acute mortality and not above insufficient by the GRADE criteria. The ability
hospital admission rates of the individual vital signs included in the different
Mortality scales to predict outcome has seldom, or never, been
None of the studies reported on hospital admission rates studied in the ED setting. The scientific evidence for
adjusted for age and gender or mortality (Table 5). Since assessing interrater agreement (reproducibility) was lim-
previous studies have shown that age is one of the major ited for one triage scale (Brillman) whereas it was insuf-
predictors of hospital mortality [33,34] the scientific evi- ficient or lacking for all other scales. Two of the scales
dence was found to be insufficient to asses the validity (CTAS and ATS) offered limited scientific evidence, and
of the triage scales ATS, CTAS, and Medical Emergency the scientific evidence for one scale (METTS) was insuf-
Triage and Treatment System (METTS) (Table 6). How- ficient to assess the risk of early death or hospitalization
ever, safety as measured by hospital mortality in patients in patients assigned to the two lowest triage levels in 5-
graded as low risk (triage levels 4-5/green-blue) by the level scales; the studies showed the risk of death to be
triage systems may be regarded as one aspect of validity. low, but a need for inpatient care was not excluded
When assessing the above-mentioned triage scales’ level (about 5% hospital admission rate on average). Studies
of validity as regards mortality at the lowest triage levels on validity of the triage scales across all levels, i.e. their
only (levels 4-5/green-blue), the quality and relevance of ability to distinguish the urgency in patients assigned
the studies were found to be moderate. Hence, scientific the five different levels, were generally of low quality.
evidence is limited. Consequently, evidence was insufficient to assess the
Hospital admission rates in patients triaged as non-acute validity of the scales.
Nine studies reported on admission rates for the ESI, As none of the studies reported on mortality rates
ATS, and SRTS triage scales (Table 7). The studies adjusted for differences in age and gender between the
showed a range between 0.0% and 17.0% at level 5, the triage levels, we could not evaluate the validity of the
lowest triage level [6,16,35-41]. A range was also triage scales across all triage levels as regards the risk of
observed in the age panorama (mean ages between 30 early death. To estimate the safety of the scales, we stu-
and 47 years) and in hospital admission rates at triage died early death among patients assigned to the lowest
level 4 (3%-33%): 18% to 33% for ATS, 6% to 10% for triage levels (green and blue/4-5). Two triage scales
ESI, and 3% for SRTS. (ATS and CTAS) offered limited scientific evidence for
Seven of these studies were found to be of moderate assessing safety. In both scales, the patients assigned to
and two of low quality and relevance, and the scientific the two lowest triage levels had a very low risk of dying
evidence for validity of admission rates for patients in within 24 hours after triage. Hence, in this respect, the
the lowest triage levels (levels 4-5/green-blue) was found scales are safe to use. Scientific evidence for METTS,
to be limited (Table 8). the newly developed Swedish triage scale, was found to

Table 4 Appraisal of scientific evidence (according to GRADE) - Reliability of triage scales

Effect measure Triage No. Patients/cases (no. Agreement (Kappa/ Scientific Comments
(endpoint) scale Studies) percent) evidence
Reliability ATS 10 cases 38.7%-79% Insufficient Reduction for study quality and imprecise
(1 study) [26] ⊕○○○ data (-1)
CTAS 1372 patients/cases 0.20-0.84 Insufficient Reduction for study quality and heterogeneity
(5 studies) [19,27-30] (-value) ⊕○○○ of results (-1)
MTS 50 cases 0.48 (-value) Insufficient Reduction for study quality and imprecise
(1 study) [31] ⊕○○○ data (-1)
SRTS 423 patients 0.87 (-value) Insufficient Reduction for study quality and uncertainty of
(1 study) [6] ⊕○○○ transferability (-1)
Rutschmann 22 cases 0.28-0.40 Insufficient Reduction for study quality (-1)
(1 study) [8] (-value) ⊕○○○
Brillman 5123 patients 0.45 (-value) Limited
(1 study) [32] ⊕⊕○○
All studies are observational.
Farrohknia et al. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine 2011, 19:42 Page 9 of 13
http://www.sjtrem.com/content/19/1/42

Table 5 Studies on how the assessment of the urgency of need to see a physician according to different triage
systems could predict hospital mortality
Author Year, Triage Patient Outcome Results (Mortality Remarks Study quality
reference Country system characteristics: Age frequency per triage level) and relevance
Gender
Dong SL et al ECTAS 29 346 patients Mortality in Triage level: - Low number of fatalities Moderate
2007, [43] 47 years ED 1: 22% (70 cases)
Canada 48% female 2: 0.22%
3: 0.031%
4: 0.018%
5: 0%
OR 664 (357-1233),
1 vs 2-5
Dent A et al ATS 42 778 patients In-hospital Triage level: Moderate
1999, [35] Age & sex not given mortality 1: 16%
2: 5%
3: 2%
4: 1%
5: 0.1%
p < 0.0001
Widgren BR et al METTS 8 695 patients In-hospital Triage level: - Only patients admitted to Moderate
2008, [10] 65 years mortality 1: 14% hospital evaluated
Sweden 45% female 2: 6%
3: 3%
4: 3%
5: 0.5%
p < 0.001
Doherty SR et al ATS 84 802 patients 24 hours Triage level: - Consecutive patients Moderate
2003, [36] Age & sex not given mortality 1: 12%
2: 2.1%
3: 1.0%
4. 0.3%
5: 0.03%
p < 0.001
Mortality figures (%) are shown for each triage level for patients admitted to a hospital emergency department.
CTAS = Canadian Emergency Department Triage and Acuity Scale; ATS = Australian Triage Scale; METTS = Medical Emergency Triage and Treatment System

be insufficient to assess safety. Since the study recorded admitted as inpatients. The variations were wide not
the risk of dying during the in-hospital stay, mortality only between different triage scales, but also between
was higher than in the studies on ATS and CTAS. studies using the same scales. This indicates differences
In using the need of hospitalization as a measure of between the studies in (a) patient populations in the ED,
safety, the situation was found to be more complex. (b) access to hospital beds, (c) hospital admission poli-
Again, none of the studies reported on hospital admis- cies and traditions, and/or (d) inaccurate triage decisions
sion rates adjusted for age and gender, so we could not (i.e. patients were rated as less urgent than their actual
evaluate the validity of the triage scales across all triage urgency).
levels. However, on average, about 5% (in some studies No definitive conclusions could be drawn regarding
up to 17%) of patients in the lowest (4-5/green-blue) which of the scales was the safest as measured by the
triage levels in ATS, ESI, and SRTS were reported to be need of hospitalization. Hence, we suggest that none of

Table 6 Appraisal of scientific evidence (according to GRADE) - Validity of 5-level triage scales measured by acute
mortality
Effect measure Triage No. Patients (no. Mortality at triage level 5 Scientific Comments
(endpoint) scale Studies) (percent) evidence
Patient mortality CTAS 29 346 0% Limited Only one study, but large
(1 study) [43] ⊕⊕○○ population
ATS 127 079 0.03%-0.1% Limited
(2 studies) [35,36] ⊕⊕○○
METTS 8695 0.5% Insufficient Reduction for study quality (-1)
(1 study) [10] ⊕○○○
All the studies are observational
Farrohknia et al. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine 2011, 19:42 Page 10 of 13
http://www.sjtrem.com/content/19/1/42

Table 7 Studies on how the assessment of the urgency of need to see a physician according to different triage
systems could predict hospitalization
Author Year, Triage Patient Outcome Results (Hospital admission Comments Study quality and
reference Country system characteristics: Age frequency per triage level) relevance:
Gender
Van Gerven R et al ATS 3 650 patients, Hospital Triage level: Moderate
2001, [39] Age & sex not given admission 1: 85%
The Netherlands 2: 71%
3: 48%
4: 18%
5: 17%
p < 0.0001
Chi CH et al ESI2 3 172 patients Hospital Triage level: - ESI scored in Moderate
2006, [16] 47 years admission 1: 96% retrospect
Taiwan 47% female 2: 47% - Unclear
3: 31% inclusion criteria
4: 7%
5: 7%
p < 0.0001
Wuerz RC et al ESI 493 patients Hospital Triage level: - Unclear Low
2000, [40] 40 years admission 1: 92% inclusion criteria
USA 52% female 2: 61%
3: 36%
4: 10%
5: 0 %
p < 0.0001
Dent A et al ATS 42 778 patients Hospital Triage level: Moderate
1999, [35] Age & sex not given admission 1: 83%
2: 69%
3: 49%
4: 33%
5: 9%
p < 0.0001
Eitel DR et al ESI2 1 042 patients Hospital Triage level: - Not Moderate
2003, [37] 7 different EDs admission 1: 83% consecutive
USA 43 years 2: 67% patients
47% female 3: 42%
4: 8%
5: 4%
p < 0.001
Tanabe P et al ESI3 403 patients Hospital Triage level: - Not Low
2004, [38] 45 years admission 1: 80% consecutive
USA 49% female 2: 73% patients
3: 51% - Retrospective
4: 6% triage
5: 5%
p < 0.001
Wuerz RC et al ESI 8 251 patients Hospital Triage level: - consecutive Moderate
2001b, [41] Age & sex not given admission 1: 92% patients
USA 2: 65%
3: 35%
4: 6%
5: 2%
p < 0.001
Doherty S et al ATS 84 802 patients Hospital Triage level: - consecutive Moderate
2003, [36] Age & sex not given admission 1: 79% patients
2: 60%
3: 41%
4: 18%
5: 3.1%
p < 0.001
Maningas PA et al SRTS 33 850 patients Hospital Triage level: - consecutive Moderate
2006, [6] Age 30, 56% female admission 1: 43% patients
2: 30%
3: 13%
4: 3.0%
5: 1.4%
p < 0.0001
Hospitalization figures (%) are shown for each triage level for patients admitted to a hospital emergency department.
ATS = Australian Triage Scale; ESI = Emergency Severity Index; SRTS = Soterion Rapid Triage Scale.
Farrohknia et al. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine 2011, 19:42 Page 11 of 13
http://www.sjtrem.com/content/19/1/42

Table 8 Appraisal of scientific evidence (according to GRADE) - Safety of 5-level triage scales as measured by
hospitalisation rates in patients at triage level 5.
Effect measure (endpoint) Triage No. patients (no. Hospitalization rate at triage level Scientific Comments
scale studies) 5 (percent) evidence
Patient safety related to hospital ATS 131 230 3.1%-17% Limited
admission (3 studies) [35,36,39] ⊕⊕○○
ESI 13 361 0%-7% Limited
(5 studies) ⊕⊕○○
[16,37,38,40,41]
SRTS 33 850 1.4% Limited Only one study, but many
(1 study) [6] ⊕⊕○○ patients
All studies are observational.

the scales be used in referral of patients in the lowest ability of individual vital signs to predict outcome. Our
triage levels (4-5/green-blue), e.g. to primary care, with- literature search revealed that many more studies had
out further medical examination in the ED. been performed in intensive care units, or soon after
New diagnostic tests typically need to meet rigid cri- hospital admission.
teria before they can be accepted for widespread use. Regarding specific vital signs, limited scientific evi-
These criteria include documentation on precision. For dence supports the use of oxygen saturation and con-
non-laboratory tests, interrater agreement (reliability) is sciousness level as predictors of mortality early after
a key precision issue. Our review shows that most triage triage. However, scientific evidence was found to be
scales present insufficient scientific evidence for asses- insufficient as regards respiration and pulse, blood pres-
sing interrater agreement. The study designs used to sure, and body temperature. Hence, it remains unclear
estimate interrater agreement have often been subopti- whether the selected vital signs are the best ones to use
mal. Most of the studies are based on fictitious cases in distinguishing different risk groups. Moderate scienti-
rather than on authentic patients in real-life settings. fic evidence indicated age as a predictor of mortality
The value of the studies as regards interrater agreement early after triage, yet most triage scales do not take age
is also compromised by the fact that the mean age of into account.
patients assessed has either been low (as low as 30 MTS and eCTAS include the chief complaint leading
years) or unreported. The generalizability to real-life ED to the ED visit, but we did not find any studies that ana-
patients must therefore be questioned. lyzed which of the chief complaints are important pre-
All 5-level triage scales present insufficient evidence dictors of mortality early after triage. It appears likely
on interrater variability. The few studies that have been that in the construction of triage scales, much of the
published (most of low quality) have reported widely information was deduced from studies performed in set-
divergent interrater agreement, with kappa values ran- tings other than EDs.
ging from 0.2 (slight agreement) to 0.9 (almost perfect).
Only a single study [32] presented limited scientific evi- Strengths and limitations
dence. This was a 4-grade scale reporting a kappa value The strength of this review of the scientific literature on
of 0.45, a value usually considered to be in the moderate triage in the ED lies in its systematic approach. Our
agreement range [42]. It is evident that inter-observer search for relevant literature has been meticulous; the
agreement in triage scales must be documented in quality of the included studies has been evaluated in a
greater detail, and, if low, actions must be taken to uniform manner; and the level of evidence has been sum-
reduce variability. marized using the GRADE methodology developed under
The literature shows variations in the vital signs and the auspices of the World Health Organization [21].
chief complaints applied in triage scales. It is unclear Our review is limited to ED triage in adult patients in
whether the selected vital signs are the best at distin- somatic care. However, EDs are only part of a conti-
guishing different risk groups. Further, evidence sup- nuum of services for acutely ill and injured patients.
porting the selected thresholds for continuous variables Studies are also needed in other aspects along the conti-
is deficient. The inclusion criteria for this systematic lit- nuum of care, e.g. prehospital, psychiatric, and pediatric
erature review place considerable emphasis on relevance. triage. Other limitations are ascribed to the volume and
Triage scales are intended to be used in EDs irrespective quality of the scientific literature available. Since all stu-
of specific symptoms or disease. Hence, only studies of dies were observational, none of the evidence came
unselected patient populations in ED settings were from randomized controlled trials, the “gold standard”
included, greatly limiting the number of studies on the for evaluating new methods. As none of the studies met
Farrohknia et al. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine 2011, 19:42 Page 12 of 13
http://www.sjtrem.com/content/19/1/42

the standards for high quality, we included studies of 2. Gilboy N, Travers D, Wuerz R: Re-evaluating triage in the new millennium:
A comprehensive look at the need for standardization and quality. J
low and moderate quality in our review in accordance Emerg Nurs 1999, 25:468-73.
with the creed in evidence based medicine to use the 3. Bullard MJ, Unger B, Spence J, Grafstein E: Revisions to the Canadian
best available scientific evidence. Low study quality Emergency Department Triage and Acuity Scale (CTAS) adult guidelines.
CJEM 2008, 10:136-51.
affected the GRADE rating and was a reason why scien- 4. Jolliffe VM, Harris DW, Morris R, Wallacet P, Whittaker SJ: Can we use
tific evidence was rated as insufficient or limited for so video images to triage pigmented lesions? Br J Dermatol 2001,
many aspects of so many scales. 145:904-10.
5. Alcazar JL, Royo P, Jurado M, Minguez JA, Garcia-Manero M, Laparte C, et al:
Triage for surgical management of ovarian tumors in asymptomatic
Conclusions women: assessment of an ultrasound-based scoring system. Ultrasound
This systematic literature review reveals shortcomings in Obstet Gynecol 2008, 32:220-5.
6. Maningas PA, Hime DA, Parker DE, McMurry TA: The Soterion Rapid Triage
the scientific evidence on which presently available System: evaluation of inter-rater reliability and validity. J Emerg Med
triage scales are based. Stronger scientific evidence is 2006, 30:461-9.
needed to determine which of the vital signs and chief 7. Parenti N, Ferrara L, Bacchi Reggiani ML, Sangiorgi D, Lenzi T: Reliability
and validity of two four-level emergency triage systems. Eur J Emerg Med
complaints have the greatest prognostic value in triage. 2009, 16:115-20.
Interrater agreement (reliability), validity, and safety of 8. Rutschmann OT, Kossovsky M, Geissbuhler A, Perneger TV, Vermeulen B,
triage scales need to be investigated further, and head- Simon J, et al: Interactive triage simulator revealed important variability
in both process and outcome of emergency triage. J Clin Epidemiol 2006,
to-head comparisons are needed to determine whether 59:615-21.
any of the scales have advantages over others. 9. Taboulet P, Moreira V, Haas L, Porcher R, Braganca A, Fontaine JP, et al:
Triage with the French Emergency Nurses Classification in Hospital
scale: reliability and validity. Eur J Emerg Med 2009, 16:61-7.
Limitations 10. Widgren B, Jourak M: Medical Emergency Triage and Treatment System
This review was confined to ED triage scales for adult (METTS): A new protocol in primary triage and secondary priority
ED patients with non-psychiatric illnesses or injuries. In decision in emergency medicine. J Emerg Med , In press.
11. FitzGerald G: Emergency department triage. University of Brisbane:
the absence of an internationally agreed outcome mea- Brisbane; 1989.
sure for ED triage scale validity, the proxy variables hos- 12. Australasian College for Emergency Medicine: Guidelines on the
pital admission and mortality were used in the current implementation of the Australasian triage scale in emergency
departments.[http://www.acem.org.au/media/policies_and_guidelines/
study. These proxy variables have limitations with G24_Implementation__ATS.pdf].
regards to ED triage scale validity as the variables may 13. Gilboy N, Tanabe P, Travers D, Eitel D, Wuerz R: The Emergency Severity
be affected by events occurring after the triage assess- Index. Emergency Nurses Association Des Plaines; 2003.
14. Manchester Triage Group: Emergency Triage. Blackwell Publishing Inc:
ment. Further, comparison between ED triage scales Harayana;, Second 2006.
need to be done with caution as there may be contex- 15. Murray M, Bullard M, Grafstein E: Revisions to the Canadian Emergency
tual differences influencing the result. Department Triage and Acuity Scale implementation guidelines. CJEM
2004, 6:421-7.
16. Chi CH, Huang CM: Comparison of the Emergency Severity Index (ESI)
and the Taiwan Triage System in predicting resource utilization. J
Author details
1 Formos Med Assoc 2006, 105:617-25.
The Swedish Council for Health Technology Assessment and Dep of
17. Gottschalk SB, Wood D, DeVries S, Wallis LA, Bruijns S: The Cape Triage
Medical Sciences, Uppsala University Hospital, Uppsala, Sweden. 2Dept of
Score: a new triage system South Africa. Proposal from the Cape Triage
Clinical Science and Education and Section of Emergency Medicine,
Group. Emerg Med J 2006, 23:149-53.
Södersjukhuset (Stockholm South General Hospital) Stockholm, Sweden.
3 18. Fernandes CM, Tanabe P, Gilboy N, Johnson LA, McNair RS, Rosenau AM,
School of Health and Social Studies, Dalarna University, Falun, Sweden.
4 et al: Five-level triage: a report from the ACEP/ENA Five-level Triage Task
Dept of Medicine, Uppsala University Hospital, Uppsala, Sweden. 5Dept of
Force. J Emerg Nurs 2005, 31:39-50, quiz 118.
Emergency Medicine, Helsingborg Hospital, Helsingborg, Sweden. 6Dept of
19. Goransson K, Ehrenberg A, Marklund B, Ehnfors M: Accuracy and
Orthopedics, Uppsala University Hospital, Uppsala, Sweden. 7Dept of Public
concordance of nurses in emergency department triage. Scand J Caring
Health and Clinical Medicine, University Hospital, Umeå, Sweden. 8Dept of
Sci 2005, 19:432-8.
Emergency Medicine, Karolinska University Hospital, Solna, Sweden. 9Dept of
20. [http://www.sbu.se/upload/Publikationer/Content0/1/Bilagor_triage.pdf].
Medicine, Karolinska Institutet, Solna, Sweden.
21. Grading quality of evidence and strength of recommendations. BMJ
[http://www.bmj.com/cgi/content/full/328/7454/1490].
Authors’ contributions
22. Olsson T, Terent A, Lind L: Rapid Emergency Medicine score: a new
All authors contributed to study concept and design, and acquisition,
prognostic tool for in-hospital mortality in nonsurgical emergency
analysis, and interpretation of the data. Finally all authors read and approved
department patients. J Intern Med 2004, 255:579-87.
the submitted manuscript.
23. Goodacre S, Turner J, Nicholl J: Prediction of mortality among emergency
medical admissions. Emerg Med J 2006, 23:372-5.
Competing interests
24. Arboix A, Garcia-Eroles L, Massons J, Oliveres M: Predictive factors of in-
The authors declare that they have no competing interests.
hospital mortality in 986 consecutive patients with first-ever stroke.
Cerebrovasc Dis 1996, 6:161-5.
Received: 11 April 2011 Accepted: 30 June 2011
25. Han JH, Lindsell CJ, Hornung RW, Lewis T, Storrow AB, Hoekstra JW, et al:
Published: 30 June 2011
The elder patient with suspected acute coronary syndromes in the
emergency department. Acad Emerg Med 2007, 14:732-9.
References 26. Considine J, Ung L, Thomas S: Triage nurses’ decisions using the National
1. Gerber Zimmerman P, McNair R: From Triage exxence and process. In Triage Scale for Australian emergency departments. Accid Emerg Nurs
Traige nursing secrets. Edited by: Gerber Zimmerman P, Herr R. Missouri: 2000, 8:201-9.
Mosby Inc; 2006:.
Farrohknia et al. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine 2011, 19:42 Page 13 of 13
http://www.sjtrem.com/content/19/1/42

27. Beveridge R, Ducharme J, Janes L, Beaulieu S, Walter S: Reliability of the

Canadian emergency department triage and acuity scale: interrater
agreement. Ann Emerg Med 1999, 34:155-9.
28. Dong SL, Bullard MJ, Meurer DP, Blitz S, Ohinmaa A, Holroyd BR, et al:
Reliability of computerized emergency triage. Acad Emerg Med 2006,
13:269-75.
29. Dong SL, Bullard MJ, Meurer DP, Colman I, Blitz S, Holroyd BR, et al:
Emergency triage: comparing a novel computer triage program with
standard triage. Acad Emerg Med 2005, 12:502-7.
30. Manos D, Petrie DA, Beveridge RC, Walter S, Ducharme J: Inter-observer
agreement using the Canadian Emergency Department Triage and
Acuity Scale. CJEM 2002, 4:16-22.
31. van der Wulp I, van Baar ME, Schrijvers AJ: 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.
32. Brillman JC, Doezema D, Tandberg D, Sklar DP, Davis KD, Simms S, et al:
Triage: limitations in predicting need for emergent care and hospital
admission. Ann Emerg Med. 1996, 27:493-500.
33. Safwenberg U, Terent A, Lind L: The Emergency Department presenting
complaint as predictor of in-hospital fatality. Eur J Emerg Med 2007,
14:324-31.
34. Safwenberg U, Terent A, Lind L: Differences in long-term mortality for
different emergency department presenting complaints. Acad Emerg Med
2008, 15:9-16.
35. Dent A, Rofe G, Sansom g: Which triage category patients die in hospital
after being admitted through emergency departments? A study in one
teaching hospital. Emerg Med 1999, 11:68-71.
36. Doherty SR, Hore CT, Curran SW: Inpatient mortality as related to triage
category in three New South Wales regional base hospitals. Emerg Med
(Fremantle) 2003, 15:334-40.
37. Eitel DR, Travers DA, Rosenau AM, Gilboy N, Wuerz RC: The emergency
severity index triage algorithm version 2 is reliable and valid. Acad
Emerg Med 2003, 10:1070-80.
38. Tanabe P, Gimbel R, Yarnold PR, Kyriacou DN, Adams JG: Reliability and
validity of scores on The Emergency Severity Index version 3. Acad
Emerg Med 2004, 11:59-65.
39. Van Gerven R, Delooz H, Sermeus W: Systematic triage in the emergency
department using the Australian National Triage Scale: a pilot project.
Eur J Emerg Med 2001, 8:3-7.
40. Wuerz RC, Milne LW, Eitel DR, Travers D, Gilboy N: Reliability and validity
of a new five-level triage instrument. Acad Emerg Med 2000, 7:236-42.
41. Wuerz RC, Travers D, Gilboy N, Eitel DR, Rosenau A, Yazhari R:
Implementation and refinement of the emergency severity index. Acad
Emerg Med 2001, 8:170-6.
42. Landis JR, Koch GG: An application of hierarchical kappa-type statistics in
the assessment of majority agreement among multiple observers.
Biometrics 1977, 33:363-74.
43. Dong SL, Bullard MJ, Meurer DP, Blitz S, Akhmetshin E, Ohinmaa A, et al:
Predictive validity of a computerized emergency triage tool. Acad Emerg
Med 2007, 14:16-21.

doi:10.1186/1757-7241-19-42
Cite this article as: Farrohknia et al.: Emergency Department Triage
Scales and Their Components: A Systematic Review of the Scientific
Evidence. Scandinavian Journal of Trauma, Resuscitation and Emergency
Medicine 2011 19:42.

Submit your next manuscript to BioMed Central

and take full advantage of:

• Convenient online submission

• Thorough peer review
• No space constraints or color figure charges
• Immediate publication on acceptance
• Inclusion in PubMed, CAS, Scopus and Google Scholar
• Research which is freely available for redistribution

Submit your manuscript at

www.biomedcentral.com/submit