Fillers bmjupdates

WHO's clinical decision rule misses most cases of streptococcal sore throat in children

BMJ 2005; 331 doi: (Published 24 November 2005) Cite this as: BMJ 2005;331:0-e

Research question Is the World Health Organization's clinical decision rule useful for diagnosing streptococcal sore throat among children in poorer countries?

Answer No. The rule is not sensitive enough to be useful.

Why did the authors do the study? WHO's clinical decision rule is meant to help doctors identify and treat streptococcal pharyngitis in children without the aid of laboratory tests such as culture of throat swabs. According to the rule, doctors should suspect streptococcal pharyngitis in, and treat with penicillin, all children who have a sore throat with pharyngeal exudate and large tender cervical lymph nodes. Penicillin reduces the risk of rheumatic fever and rheumatic heart disease, and the rule was designed for low and middle income countries with a high incidence of these diseases. The clinical decision rule performed poorly in a previous evaluationfrom Egypt, and the present authors wanted to confirm and extend the evaluation to other, similarcountries.

What did they do? They tested the rule's sensitivity and specificity in cohorts of children from Rio de Janeiro (Brazil), Zagreb (Croatia), and Cairo (Egypt). The three cohorts included 1810 children aged 2-12 years presenting to paediatric outpatient clinics with a sore throat. Doctors assessed them clinically and then took throat swabs for culture. Children with throat swabs that grew group A β haemolytic streptococci had streptococcal pharyngitis diagnosed.The authors calculated the diagnostic accuracy of the clinical decision rule, compared with the definitive results from throat swabs. They used sensitivity and specificity as their main measuresof accuracy, because these measures are unaffected by the prevalence of streptococcal pharyngitis, which varies from country to country.

What did they find? A quarter of the children from Brazil (47/191), 28% (387/1419) of those from Egypt, and 42% (84/200) from Croatia had streptococcal pharyngitis diagnosed by throat swab. The clinical decision rule was an insensitive test that missed most cases of streptococcal pharyngitis in all three cohorts (sensitivity 4.1-8.5%), and in children aged < 5 years and ≥ 5 years. The rule was relatively specific (93.8-97.4%), but the authors say that a high sensitivity is more important in countries where rheumatic heart disease is still a problem and undertreatment of streptococcal sore throat could make it worse.

What does it mean? WHO's clinical decision rule looks like a poor diagnostic testfor streptococcal sore throat in the countries that are meant to be using it. If the rule was used in isolation, more than nine out of 10 children with streptococcal sore throat would go untreated. This rule is less sensitive than other similar diagnostic tools and should probably not be used to guide individual treatment or policy in countries with a relatively high incidence of rheumatic disease after streptococcal infection.

Rimoin et al. Evaluation of the WHO clinical decision rule for streptococcal pharyngitis. Archives of Disease in Childhood 2005;90: 1066-70.

This summarises a paper that has been selected by bmjupdates. To register for bmjupdates (freeemail alerts about high quality new papers in your favourite subjects) go to

View Abstract