Skip to main content

Low compliance with a validated system for emergency department triage

Dorthea Christensen1, Nanna Martin Jensen2, Rikke Maaløe1, Søren Steemann Rudolph1, Bo Belhage1 & Hans Perrild2,

1. jun. 2011
14 min.

Faktaboks

Fakta

Patients arrive at the Emergency Department (ED) with varying severity of illness. To optimize patient care and resources at the ED, triage systems have been developed. Triage can be based on clinical assessment or validated methods, e.g. the Manchester Triage System and the Emergency Severity Index [1-4]. A recent Danish study showed that no Danish ED currently uses a validated triage system and, furthermore, that triage is primarily based on clinical assessment [5].

Bispebjerg University Hospital has been using triage systems for medical emergencies since 2006 and for trauma patients since 2000. The currently used triage system was introduced in October 2007. It comprises a physiological scoring system, named the Bispebjerg Early Warning Score (BEWS) and a number of "primary criteria". If patients are identified as critically ill or severely injured, a multidisciplinary team is activated. These teams are termed Emergency Call and Trauma Call, respectively. In another study, we have shown that the BEWS is valid for ED triage and can reliably identify critically ill patients [6].

The aim of this study was to examine whether the BEWS triage system is used systematically and correctly in a mixed ED population.

MATERIAL AND METHODS

Setting: Bispebjerg University Hospital is a 600-bed urban teaching hospital located in the Northwestern District of Copenhagen. The hospital serves a population of 400,000 citizens with approximately 38,000 ED visits per year. In the hospital catchment area, a mobile emergency care unit (MECU) manned by anaesthesiologists performs pre-hospital triage for patients requiring special emergency treatment. Patients arrive at our ED by self-transportation, by ambulance or by the MECU.

The Emergency Call (EC) and Trauma Call (TC) are activated approximately 340 and 80 times per year, respectively.

Upon arrival, ED patients are met by a triage nurse. The triage nurses have received additional training and have at least two years of ED experience. The triage nurse allocates the patients to one of three waiting categories (red, blue or white) based on the perceived severity of their injuries or illnesses. This is done in accordance with regional guidelines [7] and it is based on the patient’s symptoms/diagnoses and the nurse’s clinical judgement. The most severely ill and injured patients are allocated to the "red" category. These patients presumably need immediate treatment. According to the regional guidelines, treatment must be initiated within one hour.

Patients in the "red" category ("red" patients) immediately undergo further evaluation to assess whether an EC or TC is warranted. This is a two-step process (Figure 1). The first step is to determine whether an EC or TC should be activated based on a "primary criterion". ‘‘Primary criteria’’ are signs, symptoms and mechanisms presumed to be immediately life-threatening, e.g. upper airway obstruction or high-energy trauma.

If no "primary criteria" are present, the evaluation continues to step two, where the BEWS is calculated. This calculation is based on five vital signs each assigned a score of 0-3 points. The total score gives the BEWS. A BEWS ≥ 5 will activate an EC or TC. For patients with a BEWS < 5, the EC or TC can be activated if the nurse or physician has concerns for the patient’s clinical condition. In "red" patients where no EC or TC is activated, standard ED care is given.

Most cases of EC and TC are activated after the patients have arrived at the ED as described above. In a few cases, an EC or TC is activated prior to the arrival of the patient, e.g. when a call is requested by the MECU or when the triage nurse is concerned for the patient due to information telephoned in by the ambulance.

We performed a retrospective analytical cross-sectional study. A sample of 300 randomly selected "red" patients treated over a six-month period in 2009 were included. This sample size corresponded to one ninths of the total population of »red« patients seen in the ED during this six-month period.

Demographic data, presenting problem, vital signs on arrival or within the first 15 minutes of arrival, a BEWS and the presence of "primary criteria" were obtained from nursing admission charts.

Based on information documented in the ED medical charts and ambulance records, a retrospective evaluation was made to determine whether the ED triage had been performed according to the defined criteria for "red" patients. Patients with insufficient data were excluded. The charts and records were searched for the presence of »primary criteria« and a BEWS was calculated. Patients were then subdivided into two groups: BEWS ≥ 5 and BEWS < 5.

If an EC or TC was activated in the absence of a "primary criterion" or a BEWS ≥ 5, "concern for the patient" was listed as a criterion according to our triage system.

We considered an "activated EC/TC" a call that was activated by the ED triage nurses and a "potential EC/TC" as a call not activated although a criterion was present.

Statistical method

Data are primarily presented descriptively. Age is reported as median and range.

RESULTS

Triage is performed by the ED triage nurses as described in Figure 1. In 89% (267/300) of the cases, the patients had been allocated to a standard ED reception. An EC had been activated in 24 patients, and a TC in nine patients. Demographic data on the study population are presented in Table 1.

Men and women were equally represented in our study population. Their median age was 53 years. The most common presenting problems were chest pain (22%), respiratory problems (17%) and neurological problems (16%). A total of 58% of the patients had been transported to the ED by ambulance or by the MECU. In these patients, monitoring and some level of treatment had been initiated before their arrival at the ED.

The criteria for activation of EC and TC are shown in Figure 2.

In 16 patients, the ED triage nurses identified a "primary criterion" and activated an EC or TC. These 16 cases counted eight cases of unconsciousness, three cases of ongoing convulsions, one case of airway obstruction, one case of respiratory arrest, one case of potentially life-threatening intoxication, one motorcyclist injured at > 50 km/hour and one case of fall trauma > 4 m.

Pulse, blood pressure and level of consciousness were documented in 90-95% of the "red" patients, while temperature and respiratory rate were documented in only 65% and 18% of cases, respectively. All five vital signs needed to calculate the BEWS had only been documented in 16% (47/300) of the cases.

A BEWS had both been calculated and documented in the nursing charts in 7% (22/300) of the cases. Among these, 17 patients had a BEWS < 5 and five patients had a BEWS ≥ 5. An EC or TC, however, had only been activated in two of the five patients with a BEWS ≥ 5.

The results of the retrospective triage are presented in Figure 3.

Step 1: A "primary criterion’’ was present in 24 cases. In three of these patients, a TC was correctly activated. An EC was activated in 13 cases. In the remaining eight cases of potential EC, four patients were unconscious, three patients had ongoing convulsions and one patient had a potentially life-threatening intoxication.

Step 2: A BEWS was calculated retrospectively in 54% (148/276) of the cases in which a "primary criterion" was not present. Twenty-two patients had a BEWS ≥ 5 and 126 patients had a BEWS < 5. An EC was activated in six of the 22 patients with a BEWS ≥ 5 leaving 16 potential ECs. A BEWS had been calculated by the triage nurses in only two of the six activated ECs.

The 16 potential ECs were 12 patients with respiratory problems/insufficiency, three patients with infection/sepsis and one patient with a possible cerebral apoplexy.

In 126 patients, the BEWS was below five. In six of these patients, an EC or TC had been activated without a documented reason, presumably because of "concern for the patient". The remaining 120 patients had been allocated to a standard ED reception.

In 42% (128/300) of the patients, it was impossible to calculate a retrospective BEWS because of insufficient documentation of vital signs. In five of these patients, an EC or TC had been activated presumably because of ‘‘concern for the patient". For the remaining 123 patients, a retrospective evaluation on whether the triage had been performed correctly according to the triage system could not be made.

In summary, the retrospective triage showed 24 potential ECs – eight patients with a ‘‘primary criterion" and 16 patients with a BEWS ≥ 5. All TCs had been correctly activated according to the triage system.

DISCUSSION

The principle finding of this study was that compliance with the triage system was low. The triage system had not been used systematically and the documentation of vital signs was insufficient.

In this study, we found that many critically ill patients, especially patients with respiratory problems and a decreased level of consciousness, were not properly allocated to an EC even though they fulfilled a triage system criterion. The reasons for this could be that the triage nurses misinterpreted and/or disregarded the severity of their symptoms and sent for the on-call doctors from the specialist department (e.g. the neurological department) instead. Impaired consciousness is, however, often a non-specific symptom that warrants activation of an EC to ensure thorough clinical assessment. In another study [8], we showed that a multidisciplinary team ensures early identification of critically ill patients, rapid initiation of relevant diagnostic procedures and treatment and rapid admission to the relevant departments.

The overall documentation of vital signs was poor, the BEWS was rarely calculated even when all or some vital signs were documented and almost half of the ECs and TCs were activated based on "concern for the patient". In our opinion, this shows that ED triage is primarily based on intuition and clinical judgment by the triage nurses.

Several studies have confirmed poor documentation of vital signs to be a wide-spread problem [9, 10]. In line with such studies, we found that the respiratory rate is particularly poorly documented even though it has been shown to be a sensitive marker for critical illness and high mortality [11-14]. We are pleased to see that some vital signs like blood pressure, pulse rate and level of consciousness have been documented in the majority of our patients. The predictive value of a single vital sign in identifying critically ill patients is, however, low [15]. The BEWS uses five vital signs in a weighed fashion. We have shown that this system has a high predictive value and that it is a sensitive tool for the detection of critically ill patients [6].

There are several reasons for the lack of compliance with the triage systems: Lack of sufficient training with the system, workload, an over-reliance on one’s own experience and clinical judgment among the triage nurses and a lack of acceptance of the system.

The triage nurses might see the triage system as too complicated and time-consuming compared with triage that is based on clinical judgment and intuition. However, in order to ensure a uniform, high-quality ED triage, we believe that triage should be performed according to a validated system based on objective criteria. The results published in this article support the need to use objective criteria to avoid under-triage. On the other hand, we acknowledge the importance of intuition and clinical judgment by experienced ED nurses. In our opinion, triage nurses should therefore continue to have the opportunity to activate an EC or TC based on "concern for the patient" in cases in which a "primary criterion" or a BEWS ≥ 5 is not present.

Implementation of a new system for ED triage requires substantial changes in attitudes and routines and is therefore an ongoing process that takes time, resources and which requires constant attention by the ED management. As our results show, the triage system had not been fully implemented after two years. Implementation of the system has continued to play an important role in the ongoing process of securing quality ED triage. In parallel with this study, we have therefore introduced education in the concept of EC and TC for new staff and simulation-based training and further education of the ED nurses that focus on explaining how and why to use the triage system.

As a result of these initiatives, an audit performed in May 2010 showed a rise in the proportion of "red" patients for whom all five vital signs were documented upon arrival at the ED.

In conclusion, the implementation of a validated early warning score based triage system has proved to be a long process. Despite two years of education and training, we have shown that compliance with the triage system is low. The triage system is not being used systematically and overall documentation of vital signs remains poor. As a result, EC and TC are not activated in all patients who fulfill a valid criterion.

Correspondence: Dorthea Christensen, Akacievej 18, st. th., 2830 Virum. Denmark. E-mail: dortheafc@gmail.com

Accepted: 4 May 2011

Conflicts of interest: none

Referencer

REFERENCES

  1. Cooke MW, Jinks S. Does Manchester triage system detect critically ill? J Accident Emerg Med 1999;16:179-81.

  2. Wuerz RC, Milne LW, Eitel DR et al. Reliability and validity of a new five-level triage instrument. Acad Emerg Med 2000;7:236-42.

  3. Eitel DR, Travers DA, Rosenau AM et al. The emergency severity index triage algorithm version 2 is reliable and valid. Acad Emerg Med 2003;10:1070-80.

  4. Tanabe P, Gimbel R, Yarnold PR et al. Reliability and validity of scores on the emergency severity index version 3. Acad Emerg Med 2004;11:59-65.

  5. Brabrand M, Folkestad L, Hallas P. Visitation og triage af akut indlagte medicinske patienter. Ugeskr Læger 2010;22:1666-8.

  6. Christensen D, Jensen NM, Maaløe R et al. Nurse-administered early warning score system can be used for emergency department triage. Dan Med Bul 2011;58(6):A4221.

  7. Region Hovedstaden. Visitation, journalføring og journalaudit i somatiske skadestuer. København: Region Hovedstaden, 2008.

  8. Christensen D, Maaløe R, Jensen NM et al. Multidisciplinary team reception ensures good quality of care in the emergency room. Dan Med Bul 2011;58(6):A4227.

  9. Paterson R, MacLeod DC, Thetford et al. Prediction of in-hospital mortality and length of stay using an early warning scoring system: clinical audit. Clin Med 2006;6:281-4.

  10. Rees JE, Mann C. Use of the patient at risk scores in the emergency department: a preliminary study. Emerg Med J 2004;21:698-9.

  11. Goldhill DR. The critically ill: following your mews. Oxford J Med 2001;94:507-10.

  12. Goldhill DR, McNarry AF, Mandersloot G et al. A physiologically-based early warning score for ward patients: the association between scores and outcomes. Anesthesia 2005;60:547-53.

  13. Goldhill DR, White SA, Sumner A. Physiological and values and procedures in the 24 h before ICU admission from the ward. Anesthesia 1999;54: 529-34.

  14. Parkhe M, Myles PS, Leach DS et al. Outcome of emergency department patients with delayed admission to an intensive care unit. Emerg Med 2002;14:50-7.

  15. Rudolph SF, Skousen MB, Isbye DI. The use of a physiological scoring system in the emergency department. Scand J Trauma Resusc Emerg Med 2007;15:160-3.