Healthcare Claims Data: An Underutilized Tool for Pediatric Outpatient Antimicrobial Stewardship.

Abstract

Healthcare claims are underutilized to identify factors associated with high outpatient antibiotic use. We evaluated ambulatory encounter claims of Medicaid-insured children in 34 Ohio counties in 2014. Rates of total antibiotic and azithromycin prescriptions dispensed were determined by county of patient residence. Standardized treatment rates by county were estimated for uncomplicated upper respiratory tract encounters (acute otitis media, pharyngitis, sinusitis, presumed viral infection) after adjusting for patient age and encounter provider type. Uncomplicated encounters included healthy children at initial presentation of illness. Adjusted odds of treatment were calculated for patient age, provider type, and county characteristics (rural vs metropolitan; poverty rate). Retail pharmacies dispensed 255291 antibiotics to this cohort in 2014. More than 25% were to children <3 years. County rates of total antibiotic and azithromycin prescriptions dispensed were 530.4-1548.3 and 57.3-378.7 per 1000 person-years, respectively. Of 246866 uncomplicated upper respiratory tract encounters, antibiotics were dispensed (within 3 days) in 46.1%. Presumed viral infection accounted for 18.5% of antibiotics. Standardized treatment rates by county ranged widely from 35.9% (95% confidence interval [CI], 33.3%-38.5%) to 63.2% (95% CI, 61.5%-64.9%). Compared to encounters with pediatricians, adjusted odds ratio of treatment was 2.02 (95% CI, 1.96-2.07) for family physicians and 1.74 (95% CI, 1.68-1.79) for nurse practitioners. Residence in rural or high-poverty counties increased odds of treatment. Healthcare claims were useful to identify populations and providers with high antibiotic use. Claims data could be considered to track and report antibiotic prescribing frequency, especially where electronic medical records are not available.