Evaluation of antibiotic treatment initiation and duration practices in primary care. Cross-sectional study in two French multi-professional health centers

Antibiotic resistance poses a significant human and economic burden. In France, which ranks among the highest consumers of antibiotics in Europe, 93% of prescriptions are issued in primary care, primarily for respiratory tract infections. It is crucial to limit both the indications and the duration of antibiotic prescriptions, with recently updated recommendations in France aimed at achieving this goal. Our main objective was to evaluate whether general practitioners' antibiotic initiation and prescription durations for respiratory infections align with these recommendations. In this prospective cross-sectional study conducted over six weeks in two multi-professional health centers, all consultations for respiratory infections (in both adults and children) documented in patients' medical records were reviewed. Overall, 46.8% (N = 334/714) of consultations resulted in an antibiotic prescription (15.8% for nasopharyngitis and 83.5% for acute cough and bronchitis). Compliance with recommended antibiotic durations was observed in 66.7% (N = 476/714) [95% CI: 63.1%-70.0%] of consultations, with adherence rates exceeding 80% for nasopharyngitis and pharyngitis but falling below 20% for community-acquired pneumonia and acute cough and bronchitis. In total, 1,194 excess days of antibiotic therapy were identified, with an average excess of 1.7 days per prescription [95% CI: 1.4-1.9]. There remains significant room for improvement in both reducing the initiation of antibiotic treatments and shortening their prescribed durations. Managing acute coughs and bronchitis continues to be one of the key challenges in primary care. For nasopharyngitis, the high frequency of this condition translates into potentially large prescribing volumes on a collective scale. Efforts to promote the new paradigm of "shorter is better" for antibiotic prescription durations need to be intensified.