Frequently Asked Questions

There is evidence that up to 30% of patients who have a rapid response call while in hospital have sepsis (1). The recently updated consensus definitions for sepsis use the quick Sepsis-related Organ Failure Assessment (qSOFA) tool to identify patients with suspected infection who are at a greater risk for a poor outcome (2). It uses fast respiratory rate, low blood pressure and altered level of consciousness - three parameters which are already scored in the Wellington early warning score. A recent study of 30,000 patients has shown that the British national early warning score (on which the Wellington score is based) is more accurate than qSOFA for predicting death and ICU transfer in non-ICU patients. The authors concluded that qSOFA scores should not replace early warning scores when risk-stratifying patients with suspected infection (3).

Early identification and treatment of sepsis is incorporated into guidance about the competencies clinicians need to safely recognise and respond to deterioration.

Changes in level of consciousness may be overt (unconscious) or subtle (personality change) and may reflect a variety of factors. AVPU is simple to use and has been shown to be better at identifying early deterioration in conscious level when occurring in critically ill ward patients (4).

A number of other systems have been used to assess and document changes in level of consciousness associated with specific conditions or interventions. For example, sedation scores have been validated to detect the impact of sedative drugs like opioids but have not been validated for detecting changes in level of consciousness from other causes (such as infection, hypotension or hypercapnia) (5).

Similarly, the Glasgow Coma Scale (GCS) was developed as a tool for assessing patients with neurological injury. As a relatively complex scoring system, it has been shown to have significant interrater variability (6). For patients with specific neurological injury, the individual components of the GCS are required. Within tertiary hospitals, such patients are usually managed in specific neurosurgical or neurology wards where clinicians are more familiar with the complexity of the GCS.

Pain has been proposed as a vital sign for a number of years, often by specialist pain teams (7,8). Although it is important to record this information, pain scores have not been validated as a component of early warning scores. It is important that pain be recorded on the vital signs chart to help interpret abnormal vital signs and ensure patients’ pain is effectively managed.

The Wellington EWS will detect both early and late signs of opioid toxicity by scoring abnormal respiratory rate, heart rate, systolic blood pressure, and subsequently altered conscious state or hypoxaemia. Such abnormalities will prompt escalation to those with the required skills in managing opioid toxicity.

Urine output measurement, although a potentially useful marker of end-organ perfusion, is both difficult to measure in certain circumstances and subject to a variety of confounders. Ambulant patients without a urinary catheter who are able to mobilise to the toilet will be difficult to assess, as will patients with chronic renal failure who may normally produce little or no urine.

Urine output may also be influenced by drugs that either increase or decrease its volume as well as normal post-operative states where there is an appropriate release of antidiuretic hormone to conserve volume in the face of (elective surgical) trauma. For these reasons, there is no EWS assigned to urine output.

Fluid balance is measured over a 24-hour time period. Vital sign charts may cover much longer periods of time depending upon the frequency of vital sign acquisition, which varies with degree of illness. For this reason fluid balance is not included on the vital sign chart but documented separately.

Neither of these are considered 'vital signs' and as such do not need to be recorded on a vital signs chart. Weight is often used for medication calculation and there is space on the national medication chart for recording this. If daily weight measurement is required for other reasons (such as in the assessment of fluid balance or nutritional status), then this should be on a daily weight chart to enable the trend to be more easily seen over time. Bowels can be recorded on a bowel chart if required, often paired with the Bristol school scale, or in the clinical record.

Oxygen is a drug and should be prescribed and titrated to a target oxygen saturation (usually measured with a pulse oximeter) (9). Any patient who develops a new need for supplemental oxygen to maintain normoxia is at higher risk of deterioration. This is recognised in both the score weighting (2) and the need for medical review within 60 minutes (orange band). Patients who receive oxygen at home or require it for other reasons (e.g. carbon monoxide poisoning, decompression sickness) should have their EWS modified if it is clinically appropriate.

Patients who are hypoxaemic despite receiving additional oxygen will score twice (once for their hypoxaemia and once for the supplemental oxygen). Such patients are at greater risk of adverse outcomes so require more senior review. Other methods of oxygen delivery such as high-flow devices or non-invasive ventilation may be required.

In situations where oxygen is routinely administered regardless of oxygen saturation (such as in a Post Anaesthetic Care Unit), a time-limited modification for supplemental oxygen may be required.
In Wellington, patients leaving PACU with oxygen will often have their supplemental oxygen score modified to 0 which expires 4 hours after returning to the ward. Patients who still require supplemental oxygen after this time should be medically reviewed for atelectasis or aspiration events.