A new study shows that outpatient clinicians prescribe macrolide antibiotics to treat community acquired pneumonia (CAP) in most pediatric patients — even though established guidelines recommend amoxicillin in the majority of cases.

CAP is a potentially serious infection in children and a significant cause of respiratory morbidity and mortality in these young patients. In fact, community acquired infection is the leading cause of death worldwide in children younger than the age of 5.

To ensure appropriate coverage for streptococcus pneumonia, guidelines issued by the Pediatric Infectious Diseases Society of America 2011 recommend amoxicillin as the first-line agent in these cases. Despite these guidelines, antibiotic prescribing for CAP varies among pediatricians.

Macrolides are a class of antibiotics derived from a type of soil-borne bacteria known as Saccharopolyspora erythraea. Macrolides include azithromycin, clarithromycin, erythromycin, spiramycin and telithromycin.

In this retrospective cohort study, published this month in the journal Pediatrics, researchers reviewed the electronic health records of 31 pediatric primary care offices with the intent of identifying factors influencing clinicians' prescribing patterns. The results show that pediatricians prescribed macrolides more often than amoxicillin to treat CAP in outpatient settings, even though greater than 25 percent of S. pneumonia strains are resistant to this class of antibiotics.

The study included 10,414 pediatric patients ranging in age from 3 months to 18 years. The study also included the 196 physicians from 31 primary practices in urban, suburban and rural Pennsylvania and New Jersey who provided care to these patients from 2009 to 2013. The researchers collected patient and provider data, including demographics, practice settings, patient medical history and clinical presentation.

The scientists found that 42.5 percent of the children received macrolides and 16.8 percent received broad-spectrum antibiotics — only 40.1 received amoxicillin as recommended. The predicted probability that a child would received a macrolide ranged significantly across clinics, varying between 0.22 and 0.83.

The focus of the study was to identify variables that affect antibiotic prescribing patterns in outpatient pediatric settings. The researchers discovered that both clinical and nonclinical factors influenced prescription choice. Influential clinical factors include previous antibiotic use and age, while nonclinical factors influencing prescription choice included insurance status and urban versus rural practice settings.

With an adjusted odds ratio (aOR of 6.18), children aged 5 years and older were more likely to receive a macrolide prescription for community acquired pneumonia. Children with previous antibiotic exposure were also more likely to receive macrolides (aOR of 1.70).

Some factors decreased the odds of macrolide prescription. Abnormal exam results brought the aOR of receiving macrolides to 0.80, while the adjusted odds ratio of receiving macrolides in the presence of fever was only 0.44. These clinical factors helped prescribers differentiate pneumococcal from atypical pneumonia and appropriately influence choice of antibiotic.

The researchers also found that various nonclinical factors influenced the odds of macrolide use over amoxicillin. Physicians tended to prescribe macrolides more frequently than amoxicillin for patients with private insurance (aOR: 1.47). Clinician experience also influenced prescribing patterns, as practitioners with more than 10 years of provider experience were less likely to prescribe macrolides over amoxicillin.

Study limitations included potentially confounding variables that may have influenced treatment choice but were not well documented in the health record. Despite its limitations, the results of the study underscores the importance of understanding the clinical and nonclinical factors that influence inappropriate prescribing patterns in outpatient care for pediatric infections.