Quantification of antimicrobial use on Irish dairy farms: A comparison of three recording methods

Abstract

Antimicrobial use (AMU) data are essential for monitoring usage over time, facilitating reduction strategies to combat the threat of antimicrobial resistance (AMR) to both human and animal health. The objective of this study was to measure and describe AMU over a 12-mo period in Irish dairy herds and compare 3 different recording methods to a reference method. A sample of 33 Irish dairy herds were randomly selected from 6 private veterinary practices across Ireland. The herds were followed for a 12-mo period and their AMU was monitored using 3 recording methods: 1. Veterinary prescription data (VET), 2. The inventory of medicine bins on the farms (BIN), and 3. Farmer treatment records from herd recording software (APP). Each recording method was compared with a previously developed reference method for AMU. The reference method used was based on pre- and poststudy medicine stock on the farms combined with veterinary prescription data. Antimicrobial use was analyzed using both mass- and dosed-based metrics, including mass (mg) of antimicrobial active ingredient per population correction unit (mg/PCU), defined daily doses for animals (DDDVET) and defined course doses for animals (DCDVET). Median AMU was 16.24, 10.47, 8.87 and 15.55 mg/PCU by mass, and 2.43, 1.55, 1.19 and 2.26 DDDVET by dose for VET, BIN, APP, and reference method data, respectively. Reliability of the agreement between each pair of methods was quantified using the concordance correlation coefficient (CCC). When compared with the reference method, VET data had excellent reliability [95% confidence interval (CI) of CCC: 0.992-0.998]. The BIN data had good to excellent reliability [95% CI of CCC: 0.776-0.936]. The APP data had poor reliability when compared with the reference method [95% CI of CCC: -0.167-0.156]. Our results highlight that a small number of herds were contributing most to overall use and farmers showed varying levels of consistency in recording AMU. Veterinary data were the most reliable approach for assessing AMU when compared with a reference method of AMU. This is an important finding for the future monitoring of AMU at a national level.