How this instrument works
The modified Centor score, sometimes called the McIsaac score, adds up four signs a clinician checks during a sore-throat visit — tonsillar exudate, tender anterior cervical nodes, fever or a history of fever, and the absence of a cough — each worth one point if present. An age adjustment then shifts the total: +1 for a patient aged 3 to 14, no change for 15 to 44, and −1 for 45 and older, since strep pharyngitis grows less common as adults get older.
Robert Centor's original 1981 paper, built from adults seen in a single emergency room, did not add points at all. It used logistic regression on the same four signs to output a direct probability of a positive throat culture: roughly 2.5% with none of the signs present, 6.5% with one, 15% with two, 32% with three, and 56% with all four. The simple one-point-per-sign system in use today, together with the age adjustment, comes from later work by Warren McIsaac and colleagues, published in CMAJ in 1998 and validated again in JAMA in 2004 — a practical adaptation built for everyday charting, not the arithmetic Centor himself ran.
Every input here is something a clinician has already observed or asked about — a throat exam, a temperature check, a cough history — not a lab result the tool can pull on its own. The sum is a starting point for deciding whether a rapid strep test or throat culture is worth ordering and whether antibiotics make sense while waiting on results; it does not replace either test, and it is not a stand-alone diagnosis.
- Toggle Tonsillar exudate present to Yes if the throat exam shows it, No if not.
- Toggle Tender anterior cervical nodes to Yes if the neck lymph nodes are swollen and tender.
- Toggle Fever or history of fever to Yes if either applies.
- Toggle Absence of cough to Yes only when the patient has no cough.
- Choose the Age adjustment band — 3 to 14, 15 to 44, or 45 and older — then read the total score.
Worked example — three vectors across the scoring range
All four core signs present — exudate, tender nodes, fever, and no cough — in a patient aged 15 to 44 (age adjustment 0): 1 + 1 + 1 + 1 + 0 = 4, the top of the core-criteria range and a total associated with a substantially higher probability of a positive strep culture in McIsaac's validation data.
None of the four signs present, but the patient is a child aged 3 to 14 (age adjustment +1): 0 + 0 + 0 + 0 + 1 = 1 — a low score driven entirely by the pediatric age band, since strep runs more common in that age group even before any exam finding is counted.
Two of four signs present — tonsillar exudate and tender nodes, but no fever and a cough still present — in a patient 45 or older (age adjustment −1): 1 + 1 + 0 + 0 − 1 = 1, showing how identical exam findings can land at very different totals once age is folded in.
Questions
What does a modified Centor score of 4 mean?
It means all four core signs — tonsillar exudate, tender anterior cervical nodes, fever, and no cough — were present, with no age penalty applied (age 15 to 44). In McIsaac's validation work, a total this high tracked with a meaningfully higher share of positive strep cultures than lower totals, which is why many clinical guidelines treat a score of 4 as a strong prompt to test or treat rather than a reason to reassure and send someone home.
Did Centor's 1981 study actually use one point per sign?
No — that's a common misreading of the history. Centor's original paper ran a logistic regression on the four signs and reported a probability of a positive culture directly: about 2.5% with zero signs, rising to 56% with all four. The tidy one-point-per-sign total, plus the age band, was introduced later by Warren McIsaac in a 1998 CMAJ paper and confirmed again in a 2004 JAMA study — that later adaptation is what clinicians use today, not Centor's original arithmetic.
Why does a patient's age change the score?
Because the same four exam signs carry different weight at different ages. Strep pharyngitis is markedly more common in school-age children and less common in older adults, who more often have a viral sore throat or another cause entirely. McIsaac's age adjustment — +1 for ages 3 to 14, no change for 15 to 44, −1 for 45 and up — folds that base-rate difference into the same point total, rather than leaving age out of the arithmetic.
Can this score diagnose strep throat on its own?
No. Every input is an exam finding or a history question a clinician has already gathered — the tool only adds them up. It's decision support for whether a rapid antigen test or throat culture is worth ordering and whether starting antibiotics before results return makes sense; it is not a substitute for either test and belongs alongside the full clinical picture, not in place of it.
Does a low score rule out strep infection completely?
Not completely, but it lowers the odds substantially. A patient with none of the four signs and no age penalty carries a low estimated probability of a positive culture in the original derivation data, and many guidelines treat scores at the bottom of the range as grounds to skip testing and antibiotics. A small share of low-scoring patients will still culture positive, which is why the score informs a decision rather than settling it outright.
What are the four criteria the score actually adds up?
Tonsillar exudate (visible white or yellow patches on the tonsils), tender anterior cervical lymphadenopathy (swollen, tender neck nodes near the jaw), fever or a reported history of fever, and the absence of a cough — a cough points toward a viral cause, so its absence counts in favor of strep. Each is worth one point when present, before the age adjustment is added on top.
References
- Centor et al. 1981, Med Decis Making — original derivation (PubMed)
- McIsaac et al. 1998, CMAJ — clinical score adaptation (PubMed)
- McIsaac et al. 2004, JAMA — validation study (PubMed)
Read this first: This instrument computes a screening figure from population formulas — it is not a diagnosis, and it cannot see the whole picture a clinician can. Use it to inform a conversation, not to replace one.