Minimizing antimicrobial use (AMU) in livestock is needed to control antimicrobial resistance (AMR). In the Netherlands, the livestock sector reduced AMU by almost 70 % since 2009, but this reduction stagnated in recent years. With only therapeutic AMU allowed, it is clear that besides socio-economic and behavioral factors, also the farm technical characteristics influence the conditions under which farmers need AMU. These characteristics pertain to farm management, including biosecurity, vaccination schemes, nutrition, micro-climate and husbandry practices. Identifying farm-related risk factors for AMU is needed to control AMR in a sustainable and pragmatic way. This need, often concerns the overall contribution of seemingly related (rather than individualized) factors. Here, risk factors for AMU in pig and calf farms were determined using two approaches: a typical risk factor analysis based on generalized estimating equations (GEEs) or hierarchical mixed-effects models and a multiblock partial least-squares regression analysis. These methods were applied to longitudinal data from two previous studies, i.e. a panel study and an intervention study involving 36 multiplier pig farms and 51 veal calf farms in the Netherlands, respectively. The multiblock analysis allowed us to quantify the importance of each factor and their respective block (i.e. farm management domain). For pigs, factors related to internal biosecurity had the highest impact on AMU, while for calves, these were mainly related to micro-climate. Structural characteristics, such as farm size and production type, followed in importance for both sectors. While both methods provided similar outcomes, the multiblock approach provided further insights by grouping and comparing factors believed to be inter-related.
Keywords: Antibiotics; Antimicrobial use; Calves; Generalized estimating equation; Multiblock partial least-squares regression; Pigs.
Copyright © 2021 The Author(s). Published by Elsevier B.V. All rights reserved.