Skip to content

COVID-19 Red Flags for GP Telephone Triage

As we are supporting practices with telephone triage of patients and call support we thought this was an excellent article and a great piece of advice and training for our own teams that are currently supporting the NHS.

This blog was originally published by GP Online, you can find the original blog here.

It seems amazing to think that six months ago none of us had heard of coronavirus or the term COVID-19. Life has changed beyond all recognition both personally and professionally and, like many others, I now find myself spending most of the day on the phone and seeing very few patients face to face.

It’s very easy to say ‘just triage everyone’ but what criteria should we be using to decide who can stay at home to self-isolate, who should have an ambulance now and who needs to be seen face to face? What are the red flags that we should be looking for, on the phone or in person?

My practice has been doing telephone triage for all GP appointments since 2013 and my conversion of calls to face-to-face appointments is usually about half. I tend to see most patients with an acute respiratory illness if they want to be seen – clearly that is going to have to change.

The situation is so new that evidence-based guidelines are in short supply, so this is written from the perspective of an experienced GP, rather than based on guidance from NICE or any other similar body.

Three groups for triage

Essentially, in our practice, we are triaging into three groups:

  • Well enough to stay home and self-isolate
  • Sick enough for you to call an ambulance now
  • In between – probably don’t need admission but need to be seen. The management of this third group will depend on local pathways and most will be seen in ‘hot hubs’ which CCGs and primary care networks are currently scrambling to set up.

It would seem sensible to ask about symptoms such as shortness of breath (SOB), chest pain and ability to do normal activities of daily living (ADLs) when triaging on the phone.

However those who are used to triage will know that all but the most fit and sensible patients generally say that they are short of breath if they have a cough, so this is difficult to assess. Often all that you want to do is eyeball the patient and in that case the video apps available on GP systems are very useful.

Use of the Roth score

The Roth score has, if I may use the term in the current situation, ‘gone viral’ around doctors’ social media.

You ask the patient to take a deep breath and count to 30 in their native language and count the number of seconds before they have to take another breath. If it is less than 8 seconds, the saturations are likely to be less than 95% (sensitivity 78%/specificity 71%), with the sensitivity rising to 91% for a counting time of less than 5 seconds.

Unfortunately the Oxford-based COVID-19 Evidence Service in its recent review1 concludes that the Roth score, if used widely, may lead to an excess number of people being called in for face-to-face examination, because of its lack of specificity.

Individual doctors will have to decide if they want to use a higher Roth score (i.e. the patient can count for longer) as part of the notes justifying a decision not to see the patient.

Identifying red flags through questions

Otherwise, we are left with our usual clinical judgment and might want to consider the following, also suggested by the Oxford Evidence Service:1

  • Ask the patient to describe their breathing in their own words. Use your consultation skills and ask open questions. The longer they talk, the longer you have to assess SOB and ability to complete sentences. Being unable to complete a sentence at rest is a significant red flag.
  • Consider using the same question as the 111 symptom checker:
    • Are you so breathless that you are unable to speak more than a few words?
    • Are you breathing harder or faster than usual when doing nothing at all?
    • Are you so ill that you’ve stopped doing all of your usual daily activities?
  • Focus on change. Some patients will have underlying cough or SOB due to asthma or COPD, so asking how things have changed in the last day or so gives you more of an idea about whether they are deteriorating. A significant change for the worse is a definite red flag.
  • Consider what signs you can ask about on the phone. Blue lips, an audible wheeze, or needing to lean forwards and support yourself to breathe are red flags you can ask a patient about. They probably can’t reliably count their respiration rate, but many will be able to take their pulse; make sure you know what is normal or not, using NICE guidelines.2,3

If you decide that the patient needs to be seen, know where in your area they should go and, if they are coming to you, make sure you have appropriate protective equipment. Red flags are more obvious face to face and would include chest signs, tachycardia, tachypnoea and a raised CRB-65 score.4

Red flag symptoms and signs

  • Being unable to complete a sentence at rest
  • Significant change for the worse
  • (Reported by patient) Blue lips, audible wheeze, needing to lean forward and support themselves to breathe
  • (If possible for patient) Pulse rate
  • (If seen face to face) Chest signs, tachycardia, tachypnoea and a raised CRB score

CRB-65 score

Score one point for each of the following prognostic features:

  • Confusion
  • Raised respiratory rate
  • Low blood pressure
  • Age of 65 years or more

0 = low risk of death, 1-2 = intermediate risk, 3-4 = high mortality risk

  • Dr Toni Hazell is a GP in London


  1. Oxford COVID-19 Evidence Service. Are there any evidence-based ways of assessing dyspnea (breathlessness) by telephone or video? March 23, 2020.
  2. NICE. Sepsis: recognition, diagnosis and early management. NG51. July 2016.
  3. NICE. Fever in under 5s: assessment and initial management. NG143. November 2019.
  4. NICE CKS. Chest infections – adult. September 2019.