Manchester triage system in paediatric emergency care

The Manchester triage system is used in emergency departments to determine the clinical priority of patients on the basis of their presenting features.1 It is not a diagnostic based system but ascribes a time by which ideally each patient should be seen by a clinician. It therefore functions as a risk management tool and can also be used to monitor overall activity in the emergency department. It is widely used in the United Kingdom and Europe. In the linked study (doi:10.1136/bmj.a1501), Van Veen and colleagues assess the validity of this system in 17 600 children visiting an emergency department in the Netherlands in 2006-7.2

Using this system, experienced clinicians, such as emergency department nurses, assess patients on the basis of the available history and, by means of focused questioning, choose one of 52 flow pathways. The sequential questions delineate the risk and hence the time in which the patient should be seen. The discriminating questions asked form an integral part of the framework of triage and make it a repeatable and consistent system. Inter-rater studies of clinicians conducting triage show good levels of agreement, outperforming ad hoc triage.3 4 5

Of the 52 pathways in the Manchester triage system, six are specifically for children, and the remaining pathways also consider children. The discriminating questions, which determine the acceptable risk of the patient’s presenting condition, were derived by consensus of the Manchester Triage Group and have been revised 10 years after the initial launch of the system in 1996.1 In this latest version, the placement of patients in the emergency department (for example, the resuscitation room or the minors area) is suggested for the triage categories of each pathway.

One of the key requirements of the system is that an experienced clinician undertakes triage—one who is experienced with the process, has good interpersonal skills, and can communicate well with patients. The clinician must also have an understanding of the patient’s condition and be able to collate the presenting features with the clinical course of the patient’s condition, which is particularly important if the presenting features do not fit into one of the pathways. This assessment of risk for each patient depends on the confidence of the person performing any triage.

Elderly people and young people make up much of the workload of the average emergency department—25% of those attending emergency departments in the UK are children.6 Older and younger patients tend to be overtriaged; this may affect how all patients pass through the department. The undertriaged patients may wait longer and their diagnosis and treatment may be delayed, particularly if the availability of clinicians and other resources is limited.4

It has not been possible to determine how effective triage is in terms of morbidity and mortality owing to the many confounding factors involved. Only a few studies have compared the validity of various emergency department triage systems in terms of use of resources, rates of admission to hospital, and length of stay in the emergency department.1 7

Van Veen and colleagues assessed the validity of the Manchester triage system by comparing the triage category ascribed to patients with a reference standard, comprising five levels of urgency. They found that it had moderate validity in children; the triage category agreed with the reference standard in 34% of children, whereas 54% were overtriaged and 12% undertriaged.

The authors suggested that additional information may improve the sensitivity and specificity of triage in terms of these validity outcomes—for example, by using a physiological based system such as the paediatric early warning score. Unsurprisingly, studies have shown that high scores (associated with children being ill) are linked to admission to hospital. However, this score was derived from inpatients and may not be representative of the general paediatric population; a low score does not exclude admission to hospital because children may present with conditions that need urgent attention but that do not alter their immediate physiological parameters.8 9

Is there a way to determine how good triage is? The validity outcomes used by researchers may not be meaningful because the purpose of the Manchester triage system is to prioritise patients on the basis of their presenting features to a rigid time frame, not the reference standards used. Two triage systems for emergency departments have shown high levels of agreement with each other in predicting use of resources and the immediate course of patients, but both were limited in how well they predicted these two outcomes.10

The derived consensus of the pathways and their discriminators need to be superseded by prospective collection of data so that levels of attributable risk to each of the presenting conditions can be derived.