Antimicrobial use quantification in adult dairy cows - Part 2 - Developing a foundation for pharmacoepidemiology by comparing measurement methods.

As stewardship programmes seek to influence therapeutic choices and reduce resistance selection pressure, it is essential to understand the multiple factors which may influence any given antimicrobial use measure. This article compares 9 measures of antimicrobial use pertaining to adult cows in US dairy systems based on treatment records standardized to a single standardized regimen (SReg) format. There are 3 fundamental parameters underlying the nine measures: amount of active substance (AMOUNT), number of SRegs (ANIMALS) and treatment time (TIME). Additionally, three use categories were considered in the analysis: control of mastitis (Dry Cow), treatment of detected mastitis (Clinical Mastitis) and treatment of other diseases (Other Treatment). The assumptions necessary for calculation of each of the nine measures are identified. The influence of measure choice on farm rank was explored at the farm level both within and across fundamental parameters. Across all use categories, when measures of TIME and ANIMALS were compared, correlations between measures were greater than 0.91. However, when measures of AMOUNT were compared to either measures of TIME or ANIMALS, the correlation between measures was more variable across use categories (R = 0.31-0.91). These comparisons demonstrate that the selection of antimicrobial use metric can affect which dairies are considered the highest users of antimicrobials. Measurement selection also influences which use category has the greatest potential for use reduction. This indicates that measure selection can influence which use category is prioritized as a target for reduction of antimicrobial use. Utilization of this information may allow those developing dairy antimicrobial stewardship programmes to better understand the potential effects of measurement selection on driving changes in antimicrobial use.