Back

A triple bottom line approach to antimicrobial stewardship: the effects of simplifying treatment

Balm, M.; Bupha-Intr, O.; Sinha, T.; Kelly, M.; Stewart, L.; Stephen, R.; Blackmore, T.; Bloomfield, M.

2023-12-24 infectious diseases
10.1101/2023.12.21.23300392 medRxiv
Show abstract

AimOur antimicrobial guidelines (AGs) were changed in 2021 to recommend once-daily ceftriaxone in place of three-times-daily cefuroxime as preferred cephalosporin. This analysis sought to assess the effects of this on incidence of Clostridioides difficile infection (CDI), third-generation cephalosporin resistant Enterobacterales (3GCR-E), and resource utilisation. MethodBefore and after analysis of 30-day CDI and 3GCR-E incidence following receipt of cefuroxime/ceftriaxone pre- and post-AG change. Total nursing time and waste production relating to cefuroxime/ceftriaxone delivery were calculated pre- and post-change. ResultsCDI incidence was 0.6% pre- and 1.0% post-change (adjusted odds ratio [aOR] 1.44, p=0.07) and 3GCR-E incidence 3.5% and 3.1% (aOR 0.90, p=0.33). Mean per-quarter estimated nursing administration time decreased from 2065 to 1163 hours (902 nurse-hour reduction) and antibiotic-related waste generation from 1131kg to 748kg (383kg reduction). Overall days of therapy per-quarter of cefuroxime/ceftriaxone were unchanged between periods. ConclusionThis simplification of our AG from a three-times-daily to a once-daily antibiotic resulted in considerable savings for our hospital (roughly 1.7 full-time equivalent nurses and over a tonne of waste yearly), with no significant increases in CDI or 3GCR-E. The impact of dosing schedules on non-antibiotic-spectrum factors, such nursing time and resource usage, is worthy of consideration when designing AGs.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.