Dirofilarae identification with MALDI-TOF MS database

mic approach on real-world urine antimicrobial susceptibility testing and antimicrobial prescribing data.Methods: Open-source microbiology and prescribing data for patients with positive urine culture results on systemic antimicrobial treatment (from the Boston, MA retrospective, open-source, pseudonymised, hospital-wide Medical Information Mart for Intensive care [MIMIC]-IV dataset) were analysed.Algorithmic linking and analysis of data was performed using the open-source RStudio integrated development environment.The outcomes of interest were the number of decisions to stop and start Access, Watch, and Reserve antimicrobial agents in the 48 hours following the availability of a urine antimicrobial susceptibility result.Results: In the 679 patients analysed, the commonest systemic antimicrobial decisions taken in the 48 hours following a urine result were for ceftriaxone (n = 222), ciprofloxacin (n = 166), and vancomycin (n = 143) -there was a net change of + 11 Access agents (n = 134 started vs n = 145 stopped), -90 Watch agents (n = 373 started vs n = 463 stopped) and + 4 Reserve agents (n = 16 started vs n = 12 stopped).Discussion: The largest net observed change in antimicrobial treatment in the 48 hours following antimicrobial susceptibility results becoming available was a reduction in the number of Watch agent courses.However, three times more Watch agents were started in the 48 hours following antimicrobial therapy than Access agents.Although causality could not be determined from an analysis of this kind, using linked, pseudonymised healthcare data to observe trends in this way provides a quick, simple, practical way for healthcare providers to better understand what happens in real-world care settings following the provision of diagnostic information to clinicians.Linkage of further electronic healthcare record datasets (e.g., clinical coding) could help explore this in more detail -this would also help understand where diagnostic resources should be focused to maximise their impact. Conclusion:A diagnostic-linked algorithmic approach to antimicrobial surveillance using linked, pseudonymised prescribing and diagnostic data could help clinicians, laboratory managers, and service commissioners better quantify the clinical impact, value, and required resourcing of diagnostic services to help maximise their effect on antimicrobial stewardship.