Comparison of clinical prediction rules for management of pharyngitis in settings with limited resources

Objective

To compare the effectiveness of several clinical prediction rules for culture-positive streptococcal pharyngitis in a single group of patients in a setting in which clinicians routinely treat all cases of pharyngitis presumptively, without laboratory data.

Study design

A MEDLINE search identified clinical prediction rules for streptococcal pharyngitis in children. Each rule was applied analytically to data from 410 children in Cairo, Egypt with clinical pharyngitis, in whom throat cultures were performed. The diagnostic effectiveness of these rules for predicting a positive culture were assessed and compared.

Results

Seven prediction rules were identified. Of these 7 rules, 4 were developed in North American children, 1 was recommended by the World Health Organization (WHO), and 2 were developed in Egypt. In the Cairo children, the WHO rule was the least sensitive, at 12%. The 6 other rules had sensitivities ranging from 81% to 99% and specificities ranging from 4% to 40%; 2 rules seemed to be effective, with diagnostic odds ratios of 5.2 and 6.1.

Conclusions

The prediction rules demonstrated variable diagnostic effectiveness in the Egyptian children. Without laboratory testing, 2 clinical rules detected > 90% of cases of pharyngitis with positive culture for group A streptococcus and reduced overtreatment of culture-negative cases by ∼40%. Selected clinical prediction rules have useful characteristics in settings of limited resources and need further validation.

Abbreviations:  AUC, Area under the curve , GAS, Group A streptococcus , NPV, Negative predictive value , PPV, Positive predictive value , ROC, Receiver operating characteristic , WHO, World Health Organization