|| Checking for direct PDF access through Ovid
The objective was to study prescription practices of primary care physicians in prescribing antibiotics for community-acquired respiratory tract infections. Design was time series analysis and cross-sectional survey. The setting was 30 community primary health care centres. A case report form was completed for 3478 patient consultations treated by 198 office-based primary care physicians. Main outcome measures were: classification of diagnoses of respiratory tract infections made by each physician; number of antibiotic prescriptions related to these diagnoses; each physician's mean weekly number of antibiotic prescriptions during 6 months before and after the survey. Patients’ risk (odds ratio: OR) to receive an antibiotic prescription from the high and medium prescribers was 5.81 (95% confidence interval [CI] 4.85–6.96) and 2.41 (95% CI 2.04–2.86), compared to low prescribers. High and medium prescribers made more diagnoses of otitis media (OR 2.07, 95% CI 1.70–2.53 and 1.85, 95% CI 1.51–2.26, respectively) and fewer diagnoses of unspecified upper respiratory tract infection (OR 0.32, 95% CI 0.26–0.38 and 0.57, 95% CI 0.48–0.68, respectively) than low prescribers. The rank of the prescription rate of high, medium and low prescriber groups remained the same for all diagnoses except pneumonia. In addition, the annual rank between high, medium and low prescriber groups remained stable; high group prescribed more antibiotics during the year than medium group, which prescribed more than low prescriber group.