Reduced effectiveness among β-lactam antibiotics: a population-based cohort study in primary care in Italy

Background There are few data comparing the relative effectiveness of the individual β-lactams. Objectives To quantify the reduced effectiveness, defined as switching to a different antibiotic being prescribed for the same indication, among new users of β-lactam antibiotics in primary care. Methods A retrospective cohort study was conducted using the Health Search Database, an Italian primary care data source. Patients newly prescribed with β-lactams for a specific indication between 1 January 2013 and 31 December 2017 were identified. A switch to a different antibiotic for the same indication occurring during a 30day follow-up was the study outcome. Cox regression was adopted to assess the risk of switching between the different β-lactams. Results Among 178 256 patients newly treated with β-lactam antibiotics, 1172 (0.65%) switched to a different antibiotic. Amoxicillin/clavulanate (co-amoxiclav: n = 104 891) and amoxicillin (n = 21 699) were the most frequently prescribed β-lactams. The other antibiotics showed significantly higher risk of switching when compared with co-amoxiclav for lower respiratory tract [e.g. ceftriaxone, hazard ratio (HR): 1.6, 95% CI: 1.2–2.0], dental [e.g. amoxicillin, HR: 4.2, 95% CI: 2.9–5.9], and middle ear infections [e.g. amoxicillin, HR: 1.8, 95% CI: 1.1–2.7]. The same results were gathered when parenteral formulations were excluded. Conclusions The prevalence of reduced effectiveness of newly prescribed β-lactam antibiotics was lower than 1%. Specifically, the rate of switch to another antibiotic, when it was prescribed to treat low respiratory tract, dental, and middle ear infections, was lower among users of co-amoxiclav than those prescribed with other β-lactams.