Comparison of antimicrobial usage estimated using records-based software with estimates based on an inventory of drug...

Measurement of antimicrobial usage (AMU) on dairy farms is an important aspect of antimicrobial stewardship and has been performed using a variety of methods. The objectives of this observational study were to compare AMU estimated using data extracted from computerized herd records with estimates based on drug packaging inventories and to assess consistency in AMU during different periods. Data were collected between May 2023 and January 2024 from a convenience sample of 25 conventional dairy farms that recorded treatments in selected herd management software. During an initial farm visit, researchers used a custom online antimicrobial benchmarking program to retrospectively summarize AMU for mature cows over the previous 12 mo and provided containers to collect drug packaging waste. Researchers visited farms biweekly during a 4-mo period to collect and count discarded packaging and generated a second benchmark report at the end of the period. The software expressed AMU as defined daily dose (DDD) per cow per year, and the same metrics were calculated manually for health-related events that occurred during the prospective 4-mo period that drug packaging waste was collected. A t-test was used to evaluate differences between AMU estimated using the software for mo 1 and 4, and a 1-way ANOVA was used to assess differences in AMU among methods. Herds contained between 158 and 12,500 dairy cows, with bulk tank SCC ranging between 64,000 and 250,000 cells/mL and produced 38.6 (±1.56 SE) kg/cow per day of milk. Total AMU for the previous 12 mo estimated using software was 4.5 DDD/cow per year and did not differ between mo 1 and 4. Using the estimated AMU from the benchmarking software, no differences were identified for route or indication. No differences in AMU estimates were found based on method, and the interclass correlation coefficient varied from moderate to good agreement between methods. Within herds, a greater proportion of first-parity animals was associated with less AMU, whereas an increased proportion of third-parity animals was associated with greater injectable antibiotic usage. The estimates based on the benchmarking program, using 1-year retrospective data, demonstrated sufficient precision for understanding and managing AMU on dairy farms.