User Rating 0.0 ★★★★★

Total Usage 0 times

Category Medicine & Tests

Tonsillar exudates or swelling

Swollen, tender anterior cervical nodes

Temperature > 38°C (100.4°F)

Absence of cough

Patient age group

Select criteria above and press Calculate

Is this tool helpful?

Your feedback helps us improve.

★ ★ ★ ★ ★

About

Empirical antibiotic prescribing for sore throat remains a leading driver of antimicrobial resistance. The original Centor criteria, published in 1981, assigned one point each for tonsillar exudates, swollen tender anterior cervical nodes, fever history, and absence of cough. McIsaac et al. (1998) added an age-adjustment factor that shifts the score by +1 for patients aged 3 - 14 and −1 for patients ≥ 45. The resulting Modified Centor Score (range −1 to 5) stratifies the probability of Group A β-hemolytic streptococcal (GABHS) pharyngitis from roughly 1% to 51%. Misclassification carries tangible risk: undertreating true GABHS raises the incidence of peritonsillar abscess and acute rheumatic fever, while overtreating viral pharyngitis accelerates resistance and exposes patients to unnecessary drug adverse effects.

This calculator implements the five McIsaac criteria exactly as validated in prospective cohorts totaling over 600,000 patient encounters. It outputs the numeric score, the estimated GABHS probability, and the guideline-concordant management recommendation per ACP/IDSA/AAP consensus. The tool assumes a standard-prevalence outpatient setting. It does not replace clinical judgment for immunocompromised patients, patients with recurrent infections, or regions with endemic rheumatic heart disease where thresholds differ.

Formulas

The Modified Centor Score (S) is the algebraic sum of five binary criteria:

S = E + N + T + C + A

where E = 1 if tonsillar exudates or swelling are present, else 0. N = 1 if swollen, tender anterior cervical lymph nodes are present, else 0. T = 1 if temperature > 38°C (100.4°F), else 0. C = 1 if cough is absent, else 0.

A = {

+1 if age 3 - 140 if age 15 - 44−1 if age ≥ 45

The total S ranges from −1 to 5. Each integer value maps to a validated probability of GABHS pharyngitis derived from the McIsaac 1998 cohort and subsequent meta-analyses. The probability lookup is not a continuous function but a discrete mapping based on pooled sensitivity/specificity data across multiple validation studies.

Reference Data

Modified Centor Score	Estimated GABHS Probability	Recommended Action
≤ 0	1 - 2.5%	No further testing or antibiotics
1	5 - 10%	Optional rapid antigen detection test (RADT)
2	11 - 17%	RADT or throat culture; treat if positive
3	28 - 35%	RADT or throat culture; treat if positive
4	51 - 53%	Consider empirical antibiotics or confirm with RADT
5	≥ 51%	Consider empirical antibiotics or confirm with RADT
Individual Criteria Breakdown
Tonsillar exudates or swelling		+1
Swollen, tender anterior cervical lymph nodes		+1
Temperature > 38°C (100.4°F)		+1
Absence of cough		+1
Age 3 - 14 years		+1
Age 15 - 44 years		0
Age ≥ 45 years		−1
First-Line Antibiotic Reference (If Indicated)
Penicillin V (adult)	500mg PO BID × 10d	First choice per IDSA
Amoxicillin (adult)	500mg PO BID × 10d	Preferred in children (taste)
Amoxicillin (pediatric)	50mg/kg/d PO QD × 10d	Max 1000mg/d
Azithromycin (PCN allergy)	500mg day 1, then 250mg × 4d	Second line only
Cephalexin (PCN allergy, non-anaphylactic)	500mg PO BID × 10d	Avoid if anaphylaxis history
Complications of Untreated GABHS
Peritonsillar abscess	Suppurative; occurs in 1 - 2% of untreated cases
Acute rheumatic fever	Non-suppurative; risk 0.3 - 3% in endemic areas
Post-streptococcal glomerulonephritis	Non-suppurative; antibiotics may not prevent
Scarlet fever	Erythrogenic toxin-mediated; resolves with treatment
Retropharyngeal abscess	Rare; more common in children < 5y

Frequently Asked Questions

GABHS pharyngitis prevalence drops markedly after age 44. In the McIsaac validation cohort, patients ≥ 45 with all four original Centor criteria still had a GABHS rate below 15%. The age adjustment improves specificity in this demographic and reduces unnecessary antibiotic prescriptions by approximately 12% compared to the unmodified score.

No. The Modified Centor Score is validated for ages 3 and above. Children under 3 rarely develop GABHS pharyngitis; their sore throats are overwhelmingly viral. Additionally, the clinical signs (exudate pattern, node distribution) differ in this age group and the scoring criteria lose predictive validity.

At a threshold of ≥ 3, sensitivity for GABHS is approximately 50-70% and specificity is 58-75%, depending on the population prevalence. This means roughly 30-50% of true GABHS cases score below 3. That is why guidelines recommend RADT confirmation rather than empirical treatment for scores of 2-3, and why a score of 1 still warrants consideration for testing in high-risk populations.

The original Centor and McIsaac studies used patient-reported or clinician-measured oral temperature with a cutoff of 38.0°C (100.4°F). Tympanic readings correlate closely with oral. If using rectal temperature, subtract approximately 0.5°C before comparing to the threshold. Axillary readings should add 0.5°C. The criterion also accepts a reliable history of fever even if the patient is afebrile at presentation.

No. Even at a maximum score of 5, the estimated probability of GABHS is approximately 51-53%. Nearly half of patients with all criteria present still have a non-streptococcal etiology (EBV, Fusobacterium necrophorum, adenovirus). The score is a pre-test probability estimator, not a diagnostic test. Confirmatory RADT or culture remains the gold standard.

In endemic regions (parts of Sub-Saharan Africa, Pacific Islands, Aboriginal Australian communities), some guidelines recommend lower thresholds for empirical treatment - often starting antibiotics at a score of 2 or even 1 - because the cost of missed GABHS (rheumatic heart disease) is disproportionately severe. The calculator outputs standard-prevalence recommendations; clinicians in endemic settings should adjust accordingly.

Sore throat is the presenting complaint for every patient being scored, so it has zero discriminative value. Cough, however, strongly suggests a viral upper respiratory infection (rhinovirus, coronavirus, influenza). Its absence raises the posterior probability of a bacterial etiology. In the original Centor logistic regression model, absence of cough had an odds ratio of approximately 1.8 for GABHS.