Cessation of amoxicillin-clavulanate selective reporting and association with antibiotic prescribing and C. difficile infection in Ontario, Canada, 2017-2024: an ecological study

Objectives — To discourage unnecessary use of amoxicillin-clavulanate (a broader-spectrum antibiotic with high Clostridioides difficile infection (CDI) risk), selective reporting is used to suppress susceptibility results when the isolate is susceptible to narrower-spectrum antibiotics like amoxicillin. In Ontario, Canada, surveillance data revealed a rapid increase in amoxicillin-clavulanate susceptibility reporting in December 2021, consistent with a cessation of selective reporting in some community laboratories. Our objective was to examine whether increased amoxicillin-clavulanate susceptibility reporting among amoxicillin-susceptible urinary isolates was associated with increased prescribing of amoxicillin-clavulanate and community-associated CDI (CA-CDI) at a population level.

Methods — We conducted an ecological longitudinal study of Ontario residents from January 2017 to June 2024, using monthly province-wide aggregated data. The primary exposure was monthly percentage of amoxicillin-clavulanate reporting among amoxicillin-susceptible Escherichia coli and Proteus mirabilis isolates. Outcomes included monthly amoxicillin-clavulanate prescribing rates and CA-CDI incidence, while prescribing of amoxicillin, nitrofurantoin, trimethoprim-sulfamethoxazole, ciprofloxacin, and first-generation cephalosporins served as comparators. Analyses used negative binomial regression stratified by age and sex, adjusted for respiratory virus activity.

Results — Following an increase in amoxicillin-clavulanate susceptibility reporting from 4.1% (Jan 2017 – Nov 2021) to 56.7% (Dec 2021 – Jun 2024) among amoxicillin-susceptible isolates, amoxicillin-clavulanate prescribing increased (Incidence Rate Ratio (IRR): 1.29, 95%CI: 1.17-1.43), while comparator antibiotic prescribing remained stable (amoxicillin IRR: 0.97, 95%CI: 0.87-1.09). Compared to amoxicillin, amoxicillin-clavulanate prescribing increased by 36% (IRR: 1.36, 95%CI: 1.23-1.51). The change in reporting was associated with 459 (95%CI: 336-585) modelled excess cases of CA-CDI from Dec 2021–Jun 2024, compared with the projected CA-CDI incidence if there had been no change in amoxicillin-clavulanate reporting.

Conclusion — Our study suggests that reporting amoxicillin-clavulanate susceptibility among amoxicillin-susceptible urinary isolates is associated with increased amoxicillin-clavulanate prescribing and higher CA-CDI rates based on model-derived estimates, highlighting the impact of selective antibiotic susceptibility reporting by microbiology laboratories on patient outcomes.