Introduction
Objectives
Methods
Inclusion criteria for vital signs and chief complaints used in triage scales
-
Studies analyzing individual vital signs or chief complaints
-
Outcome variable defined as death within 30 days after ED arrival or during the hospital stay
Inclusion criteria for reliability and validity of triage scales
-
Studies based on real patients triaged at EDs (validity)
-
Studies based on real patients triaged at EDs or fictitious patient scenarios (reliability)
-
Studies reporting reliability at separate triage levels (reliability)
-
Studies reporting mortality and hospitalization per triage level (validity)
-
Outcome variables defined as death in the ED or hospital, and need for hospitalization (validity)
Exclusion criteria for studies on reliability of triage scales
-
Studies on interrater reproducibility are excluded in cases where any rater in the study had access to retrospective data only.
Results
Vital signs and chief complaints
Respiratory rate
Author Year, reference Country | Study design | Patient characteristics Sample Female/age Male/age Inclusion criteria Type of emergency department | Primary outcome | Outcome Frequency RR (relative risk), OR (odds ratio) P-value, 95% CI (confidence interval) | Missing data (%) | Study quality and relevance Comments |
---|---|---|---|---|---|---|
Goodacre S et al 2006 [23] United Kingdom | Observational Cohort Retrospective database review | Emergency medical admissions, life threatening category A emergency calls N = 5 583 Female: 2 350 (42.3%) Male: 3 233 (57.7%) Mean age 63.4 years Inclusion criteria: Any case where caller report chest pain, unconsciousness, not breathing and patient admitted to hospital or died in emergency department (ED) Setting: variables recorded on ambulance arrival | Mortality in hospital during the stay | Age, Glascow Coma Scale (GCS) and oxygen saturation independent predictors of mortality in multivariate analysis, blood pressure is not useful Glascow Coma Scale (GCS): OR 2.10 (95% CI 1.86-2.38) p < 0.001 Age: OR 1.74 (95% CI 1.52-1.98) p < 0.001 Saturation: OR 1.36 (95% CI 1.13-1.64) p = 0.001 | Rapid Acute Physiology Score (RAPS - blood pressure, pulse, GCS, RR, saturation and temp) in only 3 624 (64.9%). Missing in 35.1% Rapid Emergency Medicine Score (REMS - Blood pressure, pulse, GCS, RR) in only 2 215 (39,7%). Missing in 60.3%. New Score (GCS, saturation, age) in 2 743 (49.1%). Missing in 50.9% | Moderate Acceptable external validity Good/acceptable internal validity Age, GCS and saturation independent predictors of mortality. Blood pressure is not a useful predictor |
Olsson T et al 2004 [22] Sweden | Observational cohort Prospective | Nonsurgical emergency department (ED) patients n = 11 751 Female: 51.6% Male: 48.4% Mean age 61.9 (SD ± 20.7) Inclusion criteria: Patients consecutively admitted to the emergency department (ED) over 12 months. Exclusion criteria: Patients with cardiac arrest that could not be resuscitated, patients with more than one parameter missing. Setting: 1 200 bed University hospital ED in Sweden | Mortality in hospital, within 48 hours | In-hospital mortality 2.4%, mortality within 48 hours 1.0%. Predictors for mortality: Saturation OR: 1.70 (95% CI: 1.36-2.11) p < 0.0001 Respiratory frequency OR: 1.93 (95% CI: 1.37-2.72) p < 0.0002 Pulse frequency OR 1.67 (95% CI 1.36-2.07) p < 0.0002 Coma OR: 1.68 (95% CI: 1.38-2.06) p < 0.0001 Age OR: 1.34 (95% CI: 1.10-1.63) p < 0.004 | Moderate Good internal validity | |
Han JH et al 2007 [25] USA Singapore | Observational cohort Retrospective database review Comparison patients ≥/≤ 75 years | Suspected acute coronary syndrome (ACS) n = 10 126 Female: 5 635 Male: 4 491 Mean age = ? 11.4% ≥75 years Inclusion criteria: ≥ age 18, suspected ACS verified by electrocardiogram (ECG), 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) | Mortality in-hospital/within 30 days | 2.7% in-hospital mortality for patients age ≥75 years, higher 30 day mortality (Adjusted OR: 2.6, 95% CI: 1.6-4.3) | Missing data for ECG, symptoms or gender in 1 810 (15.2%) | Low Convenience sample-selection bias Confounders, such as co-morbidity not described Acceptable intern validity |
Arboix A et al 1996 [24] Spain | Observational cohort | Stroke n = 986 Female: 468 Male: 518 Mean age = ? Inclusion criteria: First-ever stroke, admitted to hospital. Setting: Department of neurology, university hospital | Mortality in-hospital | Overall mortality 16.3%. Age OR: 1.05 (95% CI: 1.03-1.07), previous or concomitant Pathologic conditions OR: 1.83 (95% CI: 1.19-2.82) Deteriorated level of Consciousness OR: 11.70 (95% CI: 7.70-17.77) Vomiting OR: 2.18 (95% CI: 1.20-3.94) 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 | Not stated | Moderate |
Oxygen saturation in blood
Pulse
Level of consciousness
Blood pressure and body temperature
Chief complaints
Age
Effect measure (endpoint) | No. Patients (no. Studies) Reference | Effect (OR, odds ratio*) | Scientific evidence | Comments |
---|---|---|---|---|
Respiratory rate predicts 30-day mortality | 11 751 1 study [22] | 1.9 | Insufficient ⊕○○○ | Only one study (-1) |
Oxygen saturation predicts 48-hour mortality or in-hospital mortality | 17 334 | 1.4 1.7 | Limited ⊕⊕○○ | |
Pulse predicts 30-day mortality | 11 751 1 study [22] | 1.7 | Insufficient ⊕○○○ | Only one study (-1) |
Level of consciousness predicts 48-hour mortality or in-hospital mortality | 18 320 | 2.1 1.7 11.7 | Limited ⊕⊕○○ | |
Age predicts 30-day mortality | 28 446 | 1.7 1.3 2.6 1.1 | Moderate ⊕⊕⊕○ | Upgrading due to effect size and dose-response effect (+1) |
Interrater agreement of triage scales (reliability)
Author Year, reference Country | Triage system | Patient characteristics: Age Gender Triageur: Amount, profession | Results: κ-values, percentage agreement (PA)/triage level | Drop out (%) | Study quality and relevance |
---|---|---|---|---|---|
Considine J et al 2000, [26] Australia | ATS | 10 scenarios 31 RNs | Triage level: 1: 59.7% PA 2: 58% PA 3: 79% PA 4: 54.8% PA 5: 38.7% PA | 0% | Low External validity is uncertain, internal validity is good while sample size is of uncertain adequacy |
Dong S et al 2006, [28] Canada | ETriage (CTAS) | 569 patients 49.4 years 51 % male Unknown amount of RNs | 0.40 (unweighted κ) Triage level: 1: 62.5% PA 2: 49.5% PA 3: 59.7% PA 4: 68.5% PA 5: 43.5% PA | 1% | Low External validity can not be assessed, internal validity is excellent while sample size is of uncertain adequacy |
Dong S et al 2005, [29] Canada | CTAS/eTriage | 693 patients 48 years 49 % male 73 RNs | 0.202 (unweighted κ) Triage level: 1: 50% PA 2: 9% PA 3: 53.5% PA 4: 73.3% PA 5: 7.2% PA | 4% | Low External validity can not be assessed, internal validity is excellent while sample size is of uncertain adequacy |
Manos D et al 2002, [30] Canada | CTAS | 42 scenarios 5 BLS 5 ALS 5 RNs 5 Drs | 0.77 overall (weighted κ) BLS: 0.76 (weighted κ) ALS: 0.73 (weighted κ) RNs: 0.80 (weighted κ) Drs: 0.82 (weighted κ) Triage level: 1: 78% PA 2: 49% PA 3: 37% PA 4: 41% PA 5: 49% PA | 0.2% | Low External validity can not be assessed, internal validity is acceptable while sample size is of uncertain adequacy |
Beveridge R et al 1999, [27] Canada | CTAS | 50 scenarios 10 RNs 10 Drs | 0.80 overall (weighted κ) 0.84 RNs (weighted κ) 0.83 Drs (weighted κ) Weighted κ / triage level (RNs): Triage level: 1: 0.73 2: 0.52 3: 0.57 4: 0.55 5: 0.66 | 15% | Low External validity can not be assessed, internal validity is acceptable while sample size is of uncertain adequacy |
Göransson K et al 2005, [19] Sweden | CTAS | 18 scenarios 423 RNs | 0.46 (unweighted κ) Triage level: 1: 85.4% PA 2: 39.5% PA 3: 34.9% PA 4: 32.1% PA 5: 65.1% PA | 0.8% | Low External validity can not be assessed, internal validity is acceptable while sample size is of uncertain adequacy |
van der Wulp I et al 2008, [31] The Netherlands | MTS | 50 scenarios 55 RNs | 0.48 (unweighted κ) Triage level: 2: 9.8% PA 3: 35.5% PA 4: 22% PA | 7.5-35.7% | Low External validity is uncertain, internal validity is good while sample size is of uncertain adequacy |
Maningas P et al 2006, [6] USA | SRTS | 423 patients 29.7 years 44% male 16 RN pairs | 0.87 (weighted κ) Triage level: 1: 85.7% PA 2: 86.7% PA 3: 86.8% PA 4: 93.9% PA 5: 74.2% PA | Low External validity can not be assessed, internal validity is good while sample size is of uncertain adequacy | |
Rutschmann OT et al 2006, [8] Switzerland | 4-tier system | 22 patient scenarios 45 RNs 8 Drs | RNs: 0.40 (weighted κ) Drs: 0.28 (weighted κ) Triage level: 1: 61% PA 2: 49.6% PA 3: 74.2% PA 4: 75.5% PA | 4% 0% | Low External validity is uncertain, internal validity is excellent while sample size is of uncertain adequacy |
Brillman J et al 1996, [32] USA | 4-tier system | 5 123 patients 64% < 35 years 54% male Unknown amount of RNs and Drs | 0.45 (unknown type of κ) Triage level: 1: 0.13% PA 2: 5.2% PA 3: 37.9% PA 4: 24.6% PA | 10% | Moderate External validity is clear, internal validity is good while sample size is of uncertain adequacy |
Effect measure (endpoint) | Triage scale | No. Patients/cases (no. Studies) | Agreement (Kappa/ percent) | Scientific evidence | Comments |
---|---|---|---|---|---|
Reliability | ATS | 10 cases (1 study) [26] | 38.7%-79% | Insufficient ⊕○○○ | Reduction for study quality and imprecise data (-1) |
CTAS | 1372 patients/cases | 0.20-0.84 (κ-value) | Insufficient ⊕○○○ | Reduction for study quality and heterogeneity of results (-1) | |
MTS | 50 cases (1 study) [31] | 0.48 (κ-value) | Insufficient ⊕○○○ | Reduction for study quality and imprecise data (-1) | |
SRTS | 423 patients (1 study) [6] | 0.87 (κ-value) | Insufficient ⊕○○○ | Reduction for study quality and uncertainty of transferability (-1) | |
Rutschmann | 22 cases (1 study) [8] | 0.28-0.40 (κ-value) | Insufficient ⊕○○○ | Reduction for study quality (-1) | |
Brillman | 5123 patients (1 study) [32] | 0.45 (κ-value) | Limited ⊕⊕○○ |
Validity of triage scales regarding acute mortality and hospital admission rates
Mortality
Author Year, reference Country | Triage system | Patient characteristics: Age Gender | Outcome | Results (Mortality frequency per triage level) | Remarks | Study quality and relevance |
---|---|---|---|---|---|---|
Dong SL et al 2007, [43] Canada | ECTAS | 29 346 patients 47 years 48% female | Mortality in ED | Triage level: 1: 22% 2: 0.22% 3: 0.031% 4: 0.018% 5: 0% OR 664 (357-1233), 1 vs 2-5 | - Low number of fatalities (70 cases) | Moderate |
Dent A et al 1999, [35] | ATS | 42 778 patients Age & sex not given | In-hospital mortality | Triage level: 1: 16% 2: 5% 3: 2% 4: 1% 5: 0.1% p < 0.0001 | Moderate | |
Widgren BR et al 2008, [10] Sweden | METTS | 8 695 patients 65 years 45% female | In-hospital mortality | Triage level: 1: 14% 2: 6% 3: 3% 4: 3% 5: 0.5% p < 0.001 | - Only patients admitted to hospital evaluated | Moderate |
Doherty SR et al 2003, [36] | ATS | 84 802 patients Age & sex not given | 24 hours mortality | Triage level: 1: 12% 2: 2.1% 3: 1.0% 4. 0.3% 5: 0.03% p < 0.001 | - Consecutive patients | Moderate |
Effect measure (endpoint) | Triage scale | No. Patients (no. Studies) | Mortality at triage level 5 (percent) | Scientific evidence | Comments |
---|---|---|---|---|---|
Patient mortality | CTAS | 29 346 (1 study) [43] | 0% | Limited ⊕⊕○○ | Only one study, but large population |
ATS | 127 079 | 0.03%-0.1% | Limited ⊕⊕○○ | ||
METTS | 8695 (1 study) [10] | 0.5% | Insufficient ⊕○○○ | Reduction for study quality (-1) |
Hospital admission rates in patients triaged as non-acute
Author Year, reference Country | Triage system | Patient characteristics: Age Gender | Outcome | Results (Hospital admission frequency per triage level) | Comments | Study quality and relevance: |
---|---|---|---|---|---|---|
Van Gerven R et al 2001, [39] The Netherlands | ATS | 3 650 patients, Age & sex not given | Hospital admission | Triage level: 1: 85% 2: 71% 3: 48% 4: 18% 5: 17% p < 0.0001 | Moderate | |
Chi CH et al 2006, [16] Taiwan | ESI2 | 3 172 patients 47 years 47% female | Hospital admission | Triage level: 1: 96% 2: 47% 3: 31% 4: 7% 5: 7% p < 0.0001 | - ESI scored in retrospect - Unclear inclusion criteria | Moderate |
Wuerz RC et al 2000, [40] USA | ESI | 493 patients 40 years 52% female | Hospital admission | Triage level: 1: 92% 2: 61% 3: 36% 4: 10% 5: 0 % p < 0.0001 | - Unclear inclusion criteria | Low |
Dent A et al 1999, [35] | ATS | 42 778 patients Age & sex not given | Hospital admission | Triage level: 1: 83% 2: 69% 3: 49% 4: 33% 5: 9% p < 0.0001 | Moderate | |
Eitel DR et al 2003, [37] USA | ESI2 | 1 042 patients 7 different EDs 43 years 47% female | Hospital admission | Triage level: 1: 83% 2: 67% 3: 42% 4: 8% 5: 4% p < 0.001 | - Not consecutive patients | Moderate |
Tanabe P et al 2004, [38] USA | ESI3 | 403 patients 45 years 49% female | Hospital admission | Triage level: 1: 80% 2: 73% 3: 51% 4: 6% 5: 5% p < 0.001 | - Not consecutive patients - Retrospective triage | Low |
Wuerz RC et al 2001b, [41] USA | ESI | 8 251 patients Age & sex not given | Hospital admission | Triage level: 1: 92% 2: 65% 3: 35% 4: 6% 5: 2% p < 0.001 | - consecutive patients | Moderate |
Doherty S et al 2003, [36] | ATS | 84 802 patients Age & sex not given | Hospital admission | Triage level: 1: 79% 2: 60% 3: 41% 4: 18% 5: 3.1% p < 0.001 | - consecutive patients | Moderate |
Maningas PA et al 2006, [6] | SRTS | 33 850 patients Age 30, 56% female | Hospital admission | Triage level: 1: 43% 2: 30% 3: 13% 4: 3.0% 5: 1.4% p < 0.0001 | - consecutive patients | Moderate |
Effect measure (endpoint) | Triage scale | No. patients (no. studies) | Hospitalization rate at triage level 5 (percent) | Scientific evidence | Comments |
---|---|---|---|---|---|
Patient safety related to hospital admission | ATS | 131 230 | 3.1%-17% | Limited ⊕⊕○○ | |
ESI | 13 361 | 0%-7% | Limited ⊕⊕○○ | ||
SRTS | 33 850 (1 study) [6] | 1.4% | Limited ⊕⊕○○ | Only one study, but many patients |