Using National Early Warning Scores for COVID-19 hospital patients

What we did

We investigated how NEWS2 scores related to hospital deaths in COVID-19 patients in four hospitals in the West of England:

• Great Western Hospitals NHS Foundation Trust
• North Bristol Trust
• Royal United Hospitals Bath NHS Foundation Trust
• University Hospitals Bristol and Weston NHS Foundation Trust

For all adults with COVID-19 in the study hospitals between March and June 2020, we calculated the first and maximum NEWS2 scores and looked at who died while in hospital. We also looked at the six individual NEWS2 components such as how fast someone was breathing.

What we found and what this means

We looked at data from 1,263 patients. Of these, 26 per cent died in hospital, 7 per cent went into intensive care and were later discharged from hospital, and 67 per cent left hospital without having to go to intensive care.

Most patients had low NEWS2 scores when they were first admitted to hospital:

• 50 per cent had a score of 0-2
• 27 per cent had a score of 3-4
• 14 per cent had a score of 5-6
• 9 per cent had a score of 7+

However, many patients had a high maximum score during their stay:

• 14 per cent had a score of 0-2
• 22 per cent had a score of 3-4
• 17 per cent had a score of 5-6
• 47 per cent had a score of 7+

Patients with higher first and maximum scores were more likely to go into intensive care, have a longer stay or die in hospital. Having higher NEWS2 scores was associated with an increased risk of dying.

These results show that NEWS2 is fairly good at predicting who is likely to die from COVID-19 in hospital. The respiratory elements are the most useful measures. Our findings add weight to the need to monitor COVID patients’ oxygen levels, both in hospital and the community.

Our findings support the Royal College of Physician’s recommendations to use NEWS2, alongside clinical judgement, when assessing COVID-19 patients.

What next?

These findings will help hospital staff monitor COVID-19 patients and potentially recognise when a patient is getting worse or is at risk of dying. They may also help clinicians when deciding whether to send a patient to hospital.