To compare the competencies of primary care physicians (PCPs) with poor and good prescribing performance in frequently encountered indications.

Primary care centers are one of the mostly visited health facilities by the population for different health issues.

In this cross-sectional study, we analyzed 6 125 487 prescriptions generated by 1431 PCPs which were selected by systematic sampling in 2016 in Istanbul. We defined PCPs as poor prescriber (n = 227) or good prescriber (n = 210) in terms of their prescribing performance per WHO/INRUD criteria. We compared solo diagnosis prescriptions of these two groups in ‘percentage of prescriptions in compliance with clinical guidelines’ and also rational prescribing indicators.

Poor prescribers and good prescribers significantly differed in each of the prescribing indicators for their all solo diagnosis prescriptions. Hypertension had the highest difference of the average cost per encounter (Δ = 284.2%) between poor prescribers (US$43.99 ± 63.05) and good prescribers (US$11.45 ± 45.0), whereas headache had the highest difference between the groups in the percentage encounters with an antibiotic (14.9% vs. 1.5%). Detailed analysis of the prescribing performances showed significantly higher values of each WHO/INRUD indicators for all examined diagnoses. We found significantly higher percentages of guideline-compliant drugs in good prescribers compared to that in poor prescribers in hypertension (40.8% vs 34.8%), tonsillopharyngitis (57.9% vs 50.7%), and acute sinusitis (46.4% vs 43.6%).

This study shows that the prescribing performances of PCPs are not rational enough in terms of drug selection and prescription content. Furthermore, even the physicians who have good prescribing practice appear as not satisfactorily rational in compliance with current pharmacotherapy competencies.