145 Routine data registries as a basis to analyse and improve the quality of antimicrobial prescription in Primary Care 6 Introduction Antimicrobial resistance (AMR) is increasing worldwide and is a major threat to global health (2). The leading driver of AMR is the use of antimicrobials (3). The vast majority (between 80 and 90%) of antimicrobials for use in humans is prescribed in primary care (4). Although development of multi-resistant bacteria and other consequences of AMR occur mainly in hospitals, the role of primary care as the source of the increase in AMR is larger than previously assumed, presumably through antimicrobial selection pressure in the wider population (5). Improving the quality of antimicrobial prescription in primary care may play an important part in avoiding further increase of AMR. Healthcare registries harbouring routinely collected healthcare data, such as electronic medical records (EMR) composed in primary care practices, are increasingly made available for research purposes. Combining those with several other large public dataset sources, do arise new opportunities for AMR research and data-driven healthcare. However, the responsible utilization of large registries that consist of routinely collected healthcare data presents challenges, such as non-ordered and unstructured crude data as well as the need to bring together data from different sources at the patient level. Currently, there is limited understanding of how large healthcare registries of routinely collected data can be combined and used in AMR research. In our current study we explore the feasibility and describe methods that can be used regardless of prescription rates, making our findings applicable for countries with either high or low antimicrobial prescription rates. Although the number of antimicrobial prescriptions in The Netherlands is low compared to most other European countries (6), AMR has even increased in The Netherlands over the last 10 years (7). To illustrate our definition of a low prescription rate: the number of antimicrobial prescriptions in Dutch primary care was 8.7 defined daily doses (DDD) per 1000 patients per year. By contrast, the average number of prescriptions in European primary care was 16.7 DDD/1000 patients per year (6). To improve prudent antimicrobial prescribing, we need to identify determinants of (in)appropriate antimicrobial prescribing on patient and practice level. These determinants may then allow us to define specific risk groups and to identify specific elements in a primary care practice that might be the target of antimicrobial stewardship interventions. Previously established determinants include female gender and presence of comorbidities (8-10). However, information on socioeconomic context and primary care practice characteristics as potential determinants is lacking.
RkJQdWJsaXNoZXIy MTk4NDMw