Group A beta-hemolytic streptococci (GABHS) are found in 15 to 36 percent of children presenting with sore throat. Numerous guidelines recommend treating GABHS, only after testing, using penicillin as first-line treatment, with amoxicillin, erythromycin, and first-generation cephalosporins as second-line alternatives. Linder and colleagues analyzed data, focusing on antibiotic prescribing rates based on a chief complaint of sore throat and changes in types of antibiotics prescribed as related to GABHS testing.
The authors used national ambulatory care statistics from 768,553 patients seen in community, office-based physician practices, and hospital outpatient and emergency departments from 1995 to 2003. The study population included 4,158 children three to 17 years of age presenting with a chief complaint of sore throat and no other symptoms that could warrant an antibiotic prescription.
Most patients with a complaint of sore throat were diagnosed with acute pharyngitis (34 percent), streptococcus (17 percent), or upper respiratory tract infection (17 percent). More than one half of visits were to pediatricians, whereas the rest were to other primary care physicians and emergency departments. Antibiotics were prescribed in 53 percent of these visits. Of those antibiotics, 27 percent were not specifically recommended for GABHS. Nonrecommended antibiotics included other cephalosporins, extended-spectrum macrolides, and amoxicillin/clavulanate (Augmentin). Antibiotic prescriptions decreased during the study, declining from 66 percent in 1995 to 44 percent in 2002, then increasing to 54 percent in 2003.
There was a significant decrease in recommended prescriptions, whereas the number of nonrecommended antibiotic prescriptions remained stable. A GABHS test was performed in 53 percent of visits, a percentage that did not change over the study and resulted in an antibiotic prescription in 48 percent of cases. When no GABHS test was performed, antibiotics were prescribed 51 percent of the time. Overall, GAHBS testing was not associated with antibiotic prescribing, but when analyzed according to diagnostic code, there was less antibiotic prescribing in patients who had the GABHS test than in those who did not. For example, when the visit code was for acute pharyngitis, tonsillitis, or streptococcal sore throat, antibiotics were prescribed in 57 percent of visits where the GABHS test was performed and in 73 percent of visits where it was not. This represents a 16 percent absolute reduction in the prescribing rate for this diagnostic code.
This study found that the prescribing rate for GABHS was higher than the expected prevalence of this infection, and that in association with some diagnostic categories, antibiotics were prescribed less often when GABHS testing was performed. Inappropriate, broad-spectrum antibiotic use showed a trend toward increase even as the use of recommended antibiotics declined. The study did not determine whether GABHS testing was indicated or not or whether the results of the test were positive or negative. The authors conclude that correctly targeting a population for GABHS testing by age and symptoms, and treating only those with a positive result using narrow-spectrum antibiotics, are important steps towards judicious antibiotic use.