187 Discussion 7 Impact of antimicrobial prescribing in primary care The impact of antimicrobial prescribing in primary care on the development of AMR has not been previously established at country level. As already discussed in detail in the introduction, one could reasonably argue that the impact of primary care on AMR is likely to be low, as narrow-spectrum penicillins are presumably chosen for early disease stages. Results in chapter 4 underline the necessity of actually assessing impact, as our study showed that some GPs believe that antimicrobial prescribing in primary care does not contribute to the development of AMR (1, 2) and that only hospital and veterinary care are responsible for AMR development. Analysis of antimicrobial prescriptions in chapter 2 showed that these prescriptions are not primarily confined to narrow-spectrum penicillins, with proportions of penicillin prescriptions ranging from as low as 29% up to 65% in the 12 European countries included in the study. These findings were confirmed in chapter 6, where we found that penicillins represent only 44% of antimicrobial prescriptions in Dutch primary care. Furthermore, 11% of all antimicrobial prescriptions were for macrolides, a broadspectrum antimicrobial, and 77.2% of these prescriptions were not first or second choice antimicrobials as defined in guidelines. In chapter 2 we used the antibiotic spectrum index (ASI), a proxy indicator for antimicrobial selection pressure, to assess the impact of antimicrobial prescribing in primary care. The ASI incorporates the volume of antimicrobials used as well as their activity against microorganisms, expressed as an index number representing the spectrum of microorganisms susceptible to that drug (3). This is a novel method to assess the impact of antimicrobial prescribing. The common method is to assess volumes using defined daily doses (DDD). A major advantage of the ASI compared to DDD is the incorporation of an antimicrobial activity spectrum. In our analysis we found a better correlation between ASI and the prevalence of AMR compared to DDD. Between 80-90% of the cumulative ASI in a country originates from antimicrobial prescriptions in primary care, demonstrating that the impact of primary care on antimicrobial selection pressure is much larger than previously assumed. Our findings are supported by previous studies. A review of 243 studies showed a positive association between the volume of antimicrobial consumption in a country and the prevalence of AMR (4). Another review (n=24 studies) showed that antimicrobial prescriptions for individuals with a UTI in primary care lead to development of AMR to that antimicrobial, which may persist for up to 12 months (5). Compared to previous studies, ours was the first to use ASI to measure impact on antimicrobial selection pressure at the country level.
RkJQdWJsaXNoZXIy MTk4NDMw