Centor Score Calculator

The Centor criteria estimate the pretest probability that a sore throat is caused by group A streptococcus rather than a virus, which matters because the great majority of pharyngitis is viral and does not benefit from antibiotics. Three of the four items point toward bacterial infection: tonsillar exudate, tender anterior cervical lymphadenopathy, and fever above 38 degrees C. The fourth, absence of cough, works in the opposite direction, ruling out the viral upper respiratory pattern in which postnasal drip and lower airway irritation produce cough alongside the sore throat. The McIsaac modification adds an age term that encodes epidemiology directly into the score: plus 1 point for ages 3 to 14, where streptococcus causes perhaps 20 to 30 percent of pharyngitis, no adjustment for ages 15 to 44, and minus 1 point from age 45, where the organism accounts for well under 5 percent of cases. Approximate strep probabilities run from 1 to 2.5 percent at a score of 0 up to roughly 51 to 53 percent at 4 or 5. Guidelines therefore recommend no testing and no antibiotics at 0 to 1, rapid antigen detection testing at 2 to 3 with treatment only if positive, and testing or empiric treatment at 4 or above. Modern rapid antigen tests are about 86 to 95 percent sensitive and 95 to 99 percent specific, and a negative test in a child is usually confirmed by throat culture. The score is not valid below age 3 and does not identify Fusobacterium necrophorum, an important cause of severe pharyngitis in adolescents and young adults.

Robert Centor and colleagues derived the original four criteria from adults presenting with sore throat to an emergency room in Virginia and published the result in Medical Decision Making in 1981. The clinical problem they were addressing was over-prescription: physicians were treating most sore throats with penicillin despite knowing that only a minority were streptococcal, and there was no structured way to decide who warranted a throat culture, which at the time took a day or more to return. Warren McIsaac and colleagues in Toronto revisited the rule in a general practice population that included children, and published the age-adjusted version in the Canadian Medical Association Journal in 1998. Adding the age term corrected the original score for the fact that it had been derived entirely in adults, and improved calibration across the age spectrum. As rapid antigen detection tests became widely available, the score shifted role again, from deciding who received antibiotics to deciding who was worth testing at all. Guidance from the Infectious Diseases Society of America and equivalent bodies now treats it as the front end of a testing algorithm, motivated as much by antimicrobial stewardship as by diagnosis, since the absolute risk of acute rheumatic fever in high-income countries has fallen sharply. The FeverPAIN score, developed by Paul Little and colleagues in the United Kingdom and published in 2013, offers an alternative built on a modern primary care cohort.