Preview · awaiting clinical review
A sore throat. What is worth finding out?
Alex, 24
Alex has had a sore throat for three days and asks whether antibiotics would help.
A decision exercise with a ten-minute clock. You choose what to ask, examine, test or send; not everything is worth its minutes, and some results only come back after the consultation. Stridor and breathing difficulty have been excluded before the exercise starts.
Probability of a streptococcal cause, before you know anything else.
The consultation
Choose what to ask, examine, test or send. Each option costs minutes. Some results arrive only after the consultation. Make the call when you have enough.
Your call
Probability of a streptococcal cause, using only what you found out.
Your working
One thing to take away
Your worry
Your route
The checked route from the same start
Day three: the results come back
Back in clinical practice
NICE NG84 uses FeverPAIN or Centor to judge the likelihood of streptococcal infection: at FeverPAIN 0 to 1, no antibiotic; at 2 to 3, consider no antibiotic or a back-up prescription; at 4 to 5, consider an immediate or a back-up prescription. Red flags for a peritonsillar abscess (quinsy) need same-day assessment. Throat swabs and blood tests do not inform the decision in the room. This simplified model cannot decide a real prescription.
Safety net. Discuss symptom relief and expected duration with every option. Advise returning urgently with difficulty swallowing saliva, trismus, a muffled voice, drooling, one-sided neck swelling or any breathing difficulty, and if symptoms worsen rapidly or they become very unwell. A back-up prescription needs clear instructions on when to use it.
Sources and further reading
- NICE NG84: sore throat (acute), antimicrobial prescribing. The guidance this case follows. FeverPAIN bands (about 13 to 18%, 34 to 40% and 62 to 65% likelihood of isolating streptococcus), management at each band, and red flags.
- NICE Clinical Knowledge Summaries: sore throat, acute (UK access). Features of quinsy and glandular fever, and the role of throat swabs and bloods in primary care.
- Little et al., BMJ 2013: the PRISM trial of FeverPAIN. Further reading: the UK primary care trial of FeverPAIN-guided prescribing that underpins the score.
Every number in this case, and where it came fromHow the verdicts are worked out
What the verdicts mean
Sound: nothing materially better was available, whichever starting likelihood is true. Depends: the best choice for some starting likelihoods, not others. Also sound: no single alternative is materially better in every case. Criticised: one alternative, named, is materially better whatever the starting likelihood. Differences smaller than the case's materiality threshold are never criticised.