To assess variation in antibiotic prescribing practices among village doctors in a rural region of Shandong province, China.
Almost all outpatient encounters at village clinics result in a prescription being issued. Prescriptions were collected over a 2.5-year period from 8 primary care village clinics staffed by 24 doctors located around a town in rural Shandong province. A target of 60 prescriptions per clinic per month was sampled from an average total of around 300. Prescriptions were analysed at both aggregate and individual-prescriber levels, with a focus on diagnoses of likely viral acute upper respiratory tract infections (AURIs), defined as International Classification of Diseases, 10th Revision codes J00 and J06.9.
Proportions of prescriptions for AURIs containing (1) at least one antibiotic, (2) multiple antibiotics, (3) at least one parenteral antibiotic; classes and agents of antibiotics prescribed.
In total, 14 471 prescriptions from 23 prescribers were ultimately included, of which 5833 (40.3%) contained at least 1 antibiotic. Nearly two-thirds 62.5% (3237/5177) of likely viral AURI prescriptions contained an antibiotic, accounting for 55.5% (3237/5833) of all antibiotic-containing prescriptions. For AURIs, there was wide variation at the individual level in antibiotic prescribing rates (33.1%–88.0%), as well multiple antibiotic prescribing rates (1.3%–60.2%) and parenteral antibiotic prescribing rates (3.2%–62.1%). Each village doctor prescribed between 11 and 21 unique agents for AURIs, including many broad-spectrum antibiotics. Doctors in the highest quartile for antibiotic prescribing rates for AURI also had higher antibiotic prescribing rates than doctors in the lowest quartile for potentially bacterial upper respiratory tract infections (pharyngitis, tonsillitis, laryngopharyngitis; 89.1% vs 72.4%, p=0.002).
All village doctors overused antibiotics for respiratory tract infections. Variations in individual prescriber practices are significant even in a small homogenous setting and should be accounted for when developing targets and interventions to improve antibiotic use.