Centor score: 1 point each for tonsillar exudate, tender swollen anterior neck nodes, absence of cough and fever (0–4). The McIsaac (modified Centor) score adds 1 point for ages 3–14 and takes 1 away at 45 and over. Scores of 0–1 are low risk; 2 or more usually means a rapid strep test, not antibiotics on the score alone.
This calculator gives both scores, the risk band used by StatPearls and the share of patients who tested positive for group A strep (GAS) at each score in the largest validation study. It is written for clinicians and students.
How to use this calculator
- Tick each finding that is present. Fever means a history of temperature above 38 °C (100.4 °F).
- Enter the patient's age for the McIsaac adjustment.
- Read the McIsaac score, the unadjusted Centor score and the risk band.
- Use the band to decide on testing, then act on the rapid antigen test or culture result.
Centor criteria and McIsaac points
| Criterion | Points |
|---|---|
| Fever (history of temperature above 38 °C / 100.4 °F) | +1 |
| Tonsillar exudate | +1 |
| Swollen, tender anterior cervical lymph nodes | +1 |
| Absence of cough | +1 |
| Age 3–14 years (McIsaac only) | +1 |
| Age 15–44 years (McIsaac only) | 0 |
| Age 45 years or older (McIsaac only) | −1 |
Centor and colleagues (1981) built the four-item model with logistic regression on adults in an urban emergency room, testing it against throat culture. McIsaac and colleagues (1998) validated an age-adjusted version on 521 family practice patients aged 3 to 76 in Toronto. In that study, following the score would have cut initial antibiotic prescriptions by 48% compared with usual care.
Chance of strep at each score
The share of patients who tested positive for group A strep at each score in the original study and in the largest validation, alongside the StatPearls risk band.
| Score | Centor 1981 (adults, ED) | Centor, Fine 2012 (age 15+) | McIsaac, Fine 2012 (age 3+) | Risk band (StatPearls) |
|---|---|---|---|---|
| 0 (McIsaac ≤0) | 2.5% | 7% | 8% | Low risk (0–1): StatPearls says no further GAS testing is needed |
| 1 | 6.5% | 12% | 14% | Low risk (0–1): StatPearls says no further GAS testing is needed |
| 2 | 15% | 21% | 23% | Intermediate risk (2–3): rapid antigen test (RADT) |
| 3 | 32% | 38% | 37% | Intermediate risk (2–3): rapid antigen test (RADT) |
| 4 (McIsaac 4–5) | 56% | 57% | 55% | High risk (4+): rapid antigen test (RADT); treat if positive |
Fine, Nizet and Mandl (2012) checked both scores against 206,870 patients at a US retail health chain; overall, 23% of those aged 15 and over and 27% of those aged 3 and over tested positive. Their 95% confidence intervals overlapped the earlier published estimates. Even at the top score only a little over half had GAS, which is why a high score calls for a test rather than a prescription.
Worked examples
Five sample patients scored with the calculator's own functions, showing how age moves the McIsaac score.
| Patient | Age | Centor | McIsaac | GAS-positive (Fine, McIsaac) | Band |
|---|---|---|---|---|---|
| Child, all four signs | 8 | 4 | 5 | 55% | High risk (4+): rapid antigen test (RADT); treat if positive |
| Adult, all four signs | 30 | 4 | 4 | 55% | High risk (4+): rapid antigen test (RADT); treat if positive |
| Older adult, three signs | 50 | 3 | 2 | 23% | Intermediate risk (2–3): rapid antigen test (RADT) |
| Child, two signs | 12 | 2 | 3 | 37% | Intermediate risk (2–3): rapid antigen test (RADT) |
| Adult, one sign | 40 | 1 | 1 | 14% | Low risk (0–1): StatPearls says no further GAS testing is needed |
The 50-year-old with three signs drops from Centor 3 to McIsaac 2, while the 12-year-old with two signs rises from 2 to 3. That is the whole point of the age adjustment: strep is much more common in school-age children.
What the IDSA guideline says
The Infectious Diseases Society of America's 2012 guideline is the key US reference, and it treats clinical scores as a way to pick who needs testing, not as a diagnosis:
- Test to diagnose. Swab the throat for a rapid antigen test and/or culture, because clinical features alone do not reliably separate GAS from viral pharyngitis.
- Skip testing with clear viral features. Testing usually is not recommended with cough, runny nose, hoarseness or mouth ulcers.
- Back up negative rapid tests in children. A negative RADT in children and adolescents should be followed by a culture; this is not routinely needed in adults.
- Usually no testing under 3. Except selected children with risk factors such as an older sibling with GAS.
The guideline notes that clinical scores are useful for spotting patients at such low risk that a test is usually unnecessary. McIsaac's 2004 JAMA study compared strategies in 787 patients and found empirical treatment of adults with a score of 3 or 4 had a specificity of only 43.8%, with many unnecessary prescriptions. Treating only positive rapid tests or cultures reduced antibiotic use.
Management by score
| McIsaac score | StatPearls band | Usual next step |
|---|---|---|
| −1 to 1 | Low | No GAS test; supportive care. StatPearls adds that in telehealth a 0–1 score does not need an in-person visit. |
| 2 to 3 | Intermediate | Rapid antigen test; in children, culture if negative (IDSA) |
| 4 to 5 | High | Rapid antigen test; antibiotics if positive. Empirical treatment only if testing is unavailable (StatPearls). |
For confirmed GAS, StatPearls and CDC name penicillin or amoxicillin as first-line antibiotics.
Common mistakes
- Treating the score as a diagnosis. In Fine 2012, 45% of patients at the top McIsaac score tested negative.
- Forgetting the age adjustment. Using the plain Centor score in a child underestimates risk; in someone 45 or older it overestimates it.
- Counting a fever you have not confirmed. The criterion is a history of fever above 38 °C, not just feeling hot.
- Scoring under-3s. The scores were not built for this age group and IDSA generally advises against testing.
More clinical tools: anion gap, BUN/creatinine ratio, A/G ratio and Winter's formula.
For healthcare professionals and students only. The Centor and McIsaac scores estimate the chance of strep throat to guide testing; they do not diagnose it, and this tool offers no medical advice. Anyone with a sore throat who is struggling to swallow or breathe needs urgent care. Patients should see their own clinician.
Frequently asked questions
What are the Centor criteria?
Four findings, one point each: tonsillar exudate, swollen tender anterior cervical lymph nodes, absence of cough and a history of fever. They come from Centor and colleagues' 1981 study of adults with sore throat in an emergency department. The score runs from 0 to 4.
What are the modified Centor criteria (McIsaac score)?
The McIsaac score adds age to the Centor criteria: +1 point for ages 3–14, no change for 15–44 and −1 for 45 and over. Fine and colleagues (2012) describe it this way, and it runs from −1 to 5. McIsaac and colleagues published it in CMAJ in 1998 to cover children as well as adults, because strep is more common in school-age children and less common after 45.
What is the probability of strep at each score?
It depends on the population. In Centor's 1981 emergency-room adults, 2.5% at 0, 6.5% at 1, 15% at 2, 32% at 3, 56% at 4 criteria had a positive culture. In Fine et al. (2012), 206,870 US retail-clinic patients, the figures were 7%, 12%, 21%, 38%, 57% for Centor 0–4 (age 15+) and 8%, 14%, 23%, 37%, 55% for McIsaac 0 to 4+ (age 3+). Local strep prevalence shifts these numbers, so treat them as estimates, not a patient's exact risk.
What Centor score needs a strep test?
StatPearls sorts scores into low risk (0–1), where no further GAS testing is needed, and intermediate (2–3) or high risk (4 or more), where a rapid antigen detection test (RADT) should be done. The IDSA recommends confirming strep with a RADT and/or culture instead of relying on clinical features.
Should you give antibiotics based on the Centor score alone?
Generally no. The IDSA 2012 guideline says clinical features alone do not reliably tell GAS from viral pharyngitis, so it recommends testing. McIsaac's 2004 JAMA study found that treating adults with a score of 3 or 4 without testing led to a high rate of unnecessary antibiotics. StatPearls allows empirical treatment of a high-risk score only when a test is not available.
Does the Centor score apply to children under 3?
No. The IDSA says GAS testing is generally not indicated under age 3 because rheumatic fever and classic strep throat are rare at that age, with exceptions such as a child with an older sibling who has strep.
Does a cough rule out strep throat?
Not completely, but the IDSA says testing is usually not needed when features strongly suggest a virus, such as cough, runny nose, hoarseness or mouth ulcers. Absence of cough is one of the four Centor points.
Should a negative rapid strep test be followed by a culture?
The IDSA recommends a backup throat culture after a negative RADT in children and adolescents, but not routinely in adults. CDC guidance also says to follow a negative RADT in children with a throat culture.
What antibiotic is used for confirmed strep throat?
StatPearls and CDC name penicillin or amoxicillin as first-line treatment. CDC notes no clinical isolate of group A strep resistant to penicillin has ever been reported.
Sources & method
- Centor et al. (1981), Med Decis Making — The diagnosis of strep throat in adults in the emergency room (original four criteria and probabilities)
- McIsaac et al. (1998), CMAJ — A clinical score to reduce unnecessary antibiotic use in patients with sore throat
- McIsaac et al. (2004), JAMA — Empirical validation of guidelines for the management of pharyngitis in children and adults
- Fine, Nizet & Mandl (2012), Arch Intern Med — Large-scale validation of the Centor and McIsaac scores (206,870 patients)
- Shulman et al. (2012), Clin Infect Dis — IDSA clinical practice guideline for group A streptococcal pharyngitis, 2012 update
- StatPearls (NCBI Bookshelf) — Streptococcal pharyngitis (risk bands, RADT, treatment)
- CDC — Clinical guidance for group A streptococcal pharyngitis
Results are estimates for general information. Found an error? It helps everyone — see our methodology.