Skip to 0 minutes and 9 secondsIs there anything that we can add to that? Can we improve on the Centor score? And what's the role of rapid near-patient testing? So we've looked at both these things. Firstly, thinking about can the Centor score be improved-- it was developed in an emergency department setting. And it was used to predict the presence of group A streptococcus. And it's not that selective. And just to show you-- just to think about whether other streptococcus might be important, we looked at the presence of various streptococcus in the throat. And this is a graph showing the association of a positive swab with symptoms. And you can see that group A is important, but group C and G is also associated with symptoms.

Skip to 0 minutes and 59 secondsAnd C and G is present in around 20% of swabs from people presenting with sore throats. So if you use a score which is focused on predicting group A strep, you're going to miss the C and Gs. So we went to, in a cohort study, develop predictors of A, C, and G. We came up with the acronym FeverPAIN. And the score consists of fever in the last 24 hours. Pus on the tonsils. Attending rapidly-- that means early in the onset of the illness. The presence of severely inflamed tonsils. And the absence of cough. So sharing some features with the Centor score.

Skip to 1 minute and 41 secondsJust to give you an idea of how this new score performs compared to the Centor-- this is the Centor score in a population of people presenting with acute sore throat. And you can see there, that those present with a score of naught or one. They have low levels of strep in the throat. And that represents around about 25% of the population. You can see scores of three or four-- those which are targeted by the NICE guidance, with levels of around 50% of streptococcus. And that's around about 40% of the population. And so quite a high level of the population with that high score.

Skip to 2 minutes and 23 secondsIf you compare that to FeverPAIN, you've got-- on the left hand side, there-- the naught and one score. Similarly, low levels of streptococcus. But a larger proportion-- around 40% of people with that low score. And then a smaller number with a high score-- only 17% with a score of four or more. So it looks as if the FeverPAIN score might be slightly more efficient-- if you like it-- for identifying people with a low risk of carrying streptococcus in the throat. We went on to test that in a clinical trial. And we showed that use of the FeverPAIN score did reduce antibiotic prescribing and improved symptom control. And the control group in this trial was delayed prescribing.

Skip to 3 minutes and 8 secondsSo what previously would have been regarded as, maybe, the gold standard was a delayed prescription. Then using the targeted score improved symptom control and further reduced antibiotic uptake. We also-- in that trial-- looked at near-patient testing. And just to show you the results here-- so there's the FeverPAIN pain score in the middle compared to the control, showing a reduction in the severity of sore throat, a reduction in the duration, and a reduction in the antibiotic use. The addition of the near-patient test for streptococcus, didn't add any value over use of the score alone.

Skip to 3 minutes and 52 secondsSo in conclusion, then, targeting antibiotics using a clinical score does improve symptoms and reduce antibiotic use for sore throat. But near-patient tests used according to the clinical score had similar benefits, but no clear advantage. So to summarise-- in acute sore throat, outcomes are similar using immediate versus delayed prescribing. Short-term re-consultation is higher with no prescribing strategy. And immediate prescribing encourages a belief in antibiotics and future re-consultation. Complications of acute sore throat are rare and they're hard to predict. But delayed antibiotics are, probably, as effective as immediate antibiotics to prevent complications. Remember the A, C, and G are all important. And C and G constitute around 20% of the population attending with acute sore throat.

Skip to 4 minutes and 53 secondsFeverPAIN was designed to predict the presence of A, C, and G. And if you used FeverPAIN rather than Centor, then that does result in better symptom control and lower antibiotic use than a delayed prescription. So what's the optimal strategy? Now, we think it's a targeted prescription using a clinical score-- a default position of delayed prescribing strategy. You can use the score to identify those with more severe symptoms who might need an immediate prescription. And then those with intermediate scores, then still use a delayed prescribing strategy.

Skip to 5 minutes and 32 secondsSo FeverPAIN is hard to remember, but you don't have to remember it because it's available on a website-- the link is here. If you click on that, it will take you to a scoring tool. And then you can cut and paste the results from that into your clinical notes. It also recommends a treatment strategy based on the score. So you don't need to remember what the components are, nor how to score it. You just need to remember how to find that website and you can drag that onto your desktop.

The FeverPAIN score

The FeverPAIN score has been developed to include not only B haemolytic streptococci group A but also group C and G which can also cause acute sore throat whereas the Centor score was developed only using group A streptococcal infections.

The FeverPAIN acronym stands for:

Fever last 24h

Pus on tonsils

Attend rapidly (3 or less days)

• Severely Inflamed tonsils

No cough or coryza (i.e. pharyngeal illness)

The evidence shows that using FeverPAIN as a clinical score is as least as effective as the Centor score in reducing antibiotic use and is as effective as using near patient rapid streptococcal antigen test which is more expensive.

Share this video:

This video is from the free online course:

TARGET Antibiotics – Prescribing in Primary Care

BSAC