Comparison of multiple international metrics for benchmarking antibiotic usage (ABU) using UK beef and sheep data.

BACKGROUND: Accurate surveillance of livestock antibiotic usage (ABU) at the farm level is an increasingly important part of national antibiotic stewardship initiatives. Numerous ABU indicators or metrics have been developed in Europe and North America but the comparability of these metrics is poorly understood. For policymakers, understanding the relationship between metrics is important when considering the risks posed by ABU and how to regulate them, at the national level, and regulate international trade access in livestock products between countries who use different ABU metrics. OBJECTIVES: To quantify the patterns of ABU among beef (cattle) and lamb (sheep) production systems. To explore ABU variation between farm types across seven ABU metrics developed in Europe and North America using a common dataset of sheep and beef farms' antibiotic purchases from the UK. METHODS: A dataset of >16 200 antibiotic sales events to 686 farm enterprises of different types underwent quantitative analysis. Correlation matrixes were generated for seven international ABU metrics. RESULTS: ABU was significantly higher among calf-rearers. Across all farm types, tetracyclines and ß-lactams were the predominant groups by mass, but represent a similar dose equivalent to macrolides across most farm types. Good agreement (>0.9) was observed between metrics. CONCLUSIONS: Reliable metrics to accurately benchmark farms are crucial for maintaining confidence of farmers in the fairness of any surveillance system, especially when the ranking of any given system may be linked to financial subsidies or penalties and also when negotiating import and export access for livestock products between countries.

Schmallenberg virus neutralising antibody responses in sheep.

BACKGROUND: Schmallenberg virus (SBV) is a midge borne virus of cattle and sheep. Infection is typically asymptomatic in adult sheep but fetal infection during pregnancy can result in abortion, stillbirth, neurological disorders and malformations of variable severity in newborn animals. It was first identified in Germany and the Netherlands in 2011 and then circulated throughout Europe in 2012 and 2013. Circulation in subsequent years was low or non-existent until summer and autumn 2016, leading to an increased incidence of deformed newborn lambs and calves in 2016-17. This study reports SBV circulation in October 2016 within a group of 24 ewes and 13 rams. The ewes were monitored at 3 times points over an 11 week period (September to December 2016). RESULTS: Most ewes displayed an increase in SBV VNT with antibody titre increases greater in older, previously exposed ewes. Two ewes had SBV RNA detectable by RT-qPCR, one on 30/09/16 and one on 04/11/16. Of these ewes, one had detectable serum SBV RNA (indicating viraemia) despite pre-existing antibody. The rams had been previously vaccinated with a commercial inactivated SBV vaccine, they showed minimal neutralising antibody titres against SBV 8 months post-vaccination and all displayed increased titre in October 2016. CONCLUSION: This data suggests that SBV circulated for a minimum period of 5 weeks in September to October 2016 in central England. Ewes previously exposed to virus showed an enhanced antibody response compared to naïve animals. Pre-existing antibody titre did not prevent re-infection in at least one animal, implying immunity to SBV upon natural exposure may not be life-long. In addition, data suggests that immunity provided by killed adjuvanted SBV vaccines only provides short term protection (< 8 months) from virus.

Associations between small ruminant lentivirus infection and total milk yield and somatic cell count in a dairy sheep flock.

BACKGROUND: Small ruminant lentiviruses (SRLVs) are lentiviruses of sheep and goats, formerly known as maedi-visna (MV) in sheep and caprine encephalitis and arthritis in goats. In sheep, SRLVs commonly cause progressive pneumonia, wasting and indurative mastitis. SRLVs have a long latent period, and chronic production losses are often not recognised until very late. Few studies quantifying the production losses in ewes have been published, and none have been published under UK flock husbandry conditions. METHODS: Production records of milk yield and somatic cell count (SCC) from a dairy flock of 319 milking East Friesian × Lacaune ewes identified as MV infected via routine serological screening for SRLV antibodies were used in multivariable linear regression modelling to estimate the impact of SRLV status on total milk yield and SCC. RESULTS: Milk yield was reduced in seropositive ewes by 8.1%-9.2% over an entire lactation. SCC counts were not significantly different in SRLV-infected and unifected animals. LIMITATIONS: Further parameters, such as body condition score or clinical mastitis, that were not available may have clarified the underlying cause of milk yield drop. CONCLUSIONS: The study demonstrates substantial production losses in an SRLV-affected flock and highlights the impact of the virus on a farm's economic viability.

Longitudinal study of antimicrobial use patterns, vaccination and disease prevalence in British sheep flocks.

BACKGROUND: The aim of the study was to describe the longitudinal dynamics of antimicrobial use (AMU) on sheep farms and explore associations between AMU and management factors, vaccination strategies, reproductive performance and prevalence of lameness. METHODS: Antimicrobial supply data were collected for 272 British sheep farms for 3-6 consecutive years between 2015 and 2021. These data were obtained from the farms' veterinary practices. RESULTS: Annual median AMU ranged from 8.1 to 11.8 mg/kg population corrected unit. AMU was skewed in each year with a small proportion of very high users. AMU within farms varied substantially between years. High AMU farms in 1 year were not necessarily high in other years. No associations between AMU and either vaccine usage or lameness prevalence were found. LIMITATIONS: The study design requires veterinarians and farmers to volunteer their data. This unavoidably introduces the potential for a participation bias. CONCLUSIONS: AMU on sheep farms is generally low, with a small number of farms being responsible for high usage. Targeting antimicrobial stewardship effort towards the small minority of persistently high users may be more appropriate than a focus on generic, industry-wide attempts to reduce overall AMU.

To scan or not to scan? The economics of transthoracic ultrasonography for 'whole-flock' ovine pulmonary adenocarcinoma screening in UK sheep flocks.

BACKGROUND: Transthoracic ultrasonography (TTUS) is currently the only widely used method to diagnose suspected preclinical or subclinical cases of ovine pulmonary adenocarcinoma/Jaagsiekte (OPA) in live sheep. However, the economic impact of using TTUS as a screening test has not been described previously. METHODS: Test characteristics for TTUS in a low-prevalence situation were obtained from a previous study of 1074 breeding ewes that underwent TTUS with an experienced operator. The economic impact was modelled using a 10,000-iteration partial budget simulation and probability sensitivity analysis to explore the relative influence of model variables. RESULTS: In flocks of this size, culling true-positive and false-positive cases resulted in an estimated median net loss of £4647 (interquartile range: £3537-£6006), determined primarily by replacement ewe value and the cost of TTUS per ewe. CONCLUSION: The results of this study emphasise that great caution should be exercised by practitioners when determining if TTUS is appropriate as a screening test for OPA in low-prevalence flocks or subpopulations within a flock, such as younger age groups, where the losses incurred through the inadvertent culling of healthy sheep may significantly exceed any benefit derived from culling preclinical/subclinical cases.

To scan or not to scan? Efficacy of transthoracic ultrasonography for ovine pulmonary adenocarcinoma screening in a large commercial UK sheep flock.

BACKGROUND: Transthoracic ultrasonography (TTUS) is currently the only widely used method to diagnose preclinical or subclinical ovine pulmonary adenocarcinoma (OPA) in the live sheep. However, little is known about the test characteristics of TTUS. METHODS: One thousand and seventy-four breeding ewes in a flock with evidence of low OPA prevalence underwent TTUS by an experienced operator. Fifty-one sheep were diagnosed with OPA and underwent gross postmortem examination (PME). RESULTS: Lesions consistent with OPA were found in only 24% (12/51) of the culled ewes. Thirty-five percent (18/51) of culled ewes had gross lesions consistent with other pulmonary disease and 41% (21/51) had no detectable gross lesions on PME. Histopathology and immunohistochemistry confirmed OPA in only the 12 animals identified with OPA lesions from PME. CONCLUSION: Great caution should be exercised when deciding if TTUS is an appropriate screening test in groups of sheep where OPA prevalence may be anticipated to be low. TTUS is a subjective test and thus individual operator ability will influence the sensitivity and specificity of TTUS for OPA diagnosis while the underlying prevalence influences the eventual positive predictive value.

Clearance of Maedi-visna infection in a longitudinal study of naturally infected rams is associated with homozygosity for the TMEM154 resistance allele.

Maedi-visna (MV) is a lentiviral disease of sheep responsible for severe production losses in affected flocks. There are no vaccination or treatment options with control reliant on test and cull strategies. The most common diagnostic methods used at present are combination ELISAs for Gag and Env proteins with virus variability making PCR diagnostics still largely an experimental tool. To assess variability in viral loads and diagnostic tests results, serology, DNA and RNA viral loads were measured in the blood of 12 naturally infected rams repeatedly blood sampled over 16 months. Six animals tested negative in one or more tests at one or more time points and would have been missed on screening programmes reliant on one test method or a single time point. In addition the one animal homozygous for the 'K' allele of the TMEM154 E35K SNP maintained very low viral loads in all assays and apparently cleared infection to below detectable limits at the final time point it was sampled. This adds crucial data to the strong epidemiological evidence that this locus represents a genuine resistance marker for MV infection and is a strong candidate for selective breeding of sheep for resistance to disease.

Failure to detect Schmallenberg virus RNA in ram semen in the UK (2016-2018).

Background: Schmallenberg virus (SBV) is a midge-borne arbovirus that first emerged in the European ruminant population in 2011 and has since settled to an endemic pattern of disease outbreaks on an approximately 4-year cycle when herd immunity from the previous circulation drops to a point allowing renewed widescale virus circulation. The impacts of trade restrictions on genetic products (semen, embryos) from affected areas were severe, particularly after the discovery that the virus is intermittently shed in the semen of a small number of bulls. The trade in small ruminant (ram and goat) semen is less than that of bulls; nonetheless, there has been no study into the shedding rate of SBV in ram semen. Methods: Semen samples (n = 65) were collected as part of UK ram trials and artificial insemination studies around the period of the 2016-2018 SBV recirculation. Semen was preserved in RNAlater for shipping, and RNA extraction with RNeasy and S gene RT-quantitative PCR performed for SBV nucleic acid detection. Results: No SBV RNA was detected in any samples. Conclusions: While larger numbers of animals would be needed to completely exclude the possibility of SBV shedding in ram semen, this trial nonetheless highlights that this is likely a rare event if it occurs at all and is unlikely to play a role in disease transmission.

The impact of glutaraldehyde based footbaths on Dichelobacter nodosus prevalence and the antimicrobial resistant community of the ovine interdigital skin.

Ovine footrot, is a highly contagious polymicrobial bacterial infection, primarily caused by Dichelobacter nodosus. Preventative bactericidal footbaths are commonly used in the sheep industry to reduce the spread of bacteria. However, their effect on the bacterial community is poorly understood. This is the first study to investigate the impact of 2% Digicur (ProGiene,UK) footbath on the bacterial community of the ovine interdigital skin following a common UK footbathing routine. Swab samples were analysed by qPCR to determine prevalence and load of D. nodosus and numerated on MacConkey agar in the presence or absence of tetracycline and ampicillin to determine phenotypic antimicrobial resistance. Metagenomics were used to determine the impact of a single footbath on the bacterial community and genotypic antimicrobial resistance. The results suggest 2% Digicur is ineffective at reducing the load of D. nodosus when applied as a one off or weekly footbath, however sheep may act as a reservoir for multi-drug resistant bacteria creating opportunities to spread antimicrobial resistance to other sheep and their environment.

Variable selection for inferential models with relatively high-dimensional data: Between method heterogeneity and covariate stability as adjuncts to robust selection.

Variable selection in inferential modelling is problematic when the number of variables is large relative to the number of data points, especially when multicollinearity is present. A variety of techniques have been described to identify 'important' subsets of variables from within a large parameter space but these may produce different results which creates difficulties with inference and reproducibility. Our aim was evaluate the extent to which variable selection would change depending on statistical approach and whether triangulation across methods could enhance data interpretation. A real dataset containing 408 subjects, 337 explanatory variables and a normally distributed outcome was used. We show that with model hyperparameters optimised to minimise cross validation error, ten methods of automated variable selection produced markedly different results; different variables were selected and model sparsity varied greatly. Comparison between multiple methods provided valuable additional insights. Two variables that were consistently selected and stable across all methods accounted for the majority of the explainable variability; these were the most plausible important candidate variables. Further variables of importance were identified from evaluating selection stability across all methods. In conclusion, triangulation of results across methods, including use of covariate stability, can greatly enhance data interpretation and confidence in variable selection.

Use of bootstrapped, regularised regression to identify factors associated with lamb-derived revenue on commercial sheep farms.

The profitability of UK sheep farms is variable with many farms making a net loss. For economic sustainability, farms have to be profitable, therefore it is important to maximise income whilst controlling costs. The most important source of income in sheep flocks is from lamb production but there is little information on factors that explain variability between farms in revenue from lamb sales. The aim of this research was to identify farm, farmer and management factors likely to have the largest, most reliable associations with lamb-derived revenue. From a population of 830 sheep farms, 408 farmers completed an online questionnaire comprising over 300 variables. Total lamb-derived revenue was calculated for each farm using abattoir information including carcass classification. The median flock size was 560 ewes, median land size 265 acres, median revenue per acre from lambs sold was £197 (IQR = 120-296) and median revenue per ewe £95 (IQR = 72-123). A robust analytic approach using regularised (elastic net) regression with bootstrapping was implemented to account for multicollinearity in the data and to reduce the likelihood of model over-fitting. To provide model inference and allow ranking of variables in terms of relevance, covariate stability and coefficient distributions were evaluated. Factors with high stability and relatively large positive associations with revenue per acre were (median effect size (£); 95 % bootstrap probability interval); an increased stocking rate of 0.2 ewe/acre (13; 6-17), fertilizer being used on most of the grazing land (18; 0.1-37), the use of rotational grazing (13; 0.3-34), decreased proportion of ewes with prolapses (4; 0.3-9), separation of lame sheep from the rest of the flock (16; 0.9-37), selecting ewes for culling based on prolapses (20; 0.2-55) and infertility (20; 2-46), conducting body condition scoring of ewes at lambing (28; 3-58), early lactation (21; 1-54) or weaning (25; 2-70), increased farmer education (20; 2-54) and farmers with a positive business attitude (15; 0.2-38). Additional factors with a high stability and relatively large associations with increased revenue per ewe were; never trimming diseased feet of lame ewes (9; 0.8-22) and making use of farm records (5; 0.3-12). This is the first study in animal health epidemiology to use bootstrapped regularised regression to evaluate a wide dataset to provide a ranking of the importance of explanatory covariates. We conclude that the relatively small set of variables identified, with a potentially large influence on lamb-derived revenue, should be considered prime candidates for future intervention studies.

Sustainable lamb production: Evaluation of factors affecting lamb growth using hierarchical, cross classified and multiple memberships models.

In light of current concerns about the sustainability of red meat production in a world with increasing global demand for food from animal origin there is a need for a better understanding of factors that influence the growth rate and feed conversion efficiency of animals on commercial farms. The primary objective of this observational study was to use longitudinal data to quantify the simultaneous effects of multiple ewe and lamb factors on lamb growth rate. A secondary aim was to evaluate model structures that specifically account for lamb grouping effects during the growth period and compare these to classical hierarchical growth rate models. A total of 4172 weight recordings from 805 lambs and data on disease events were collected over a 6-month period from a commercial pedigree sheep flock. Three mixed model structures were compared, hierarchical, cross classified and multiple membership, and final estimates determined within a Bayesian framework. The multiple membership structure provided the best model fit and was used for final inference; taking account of the effect of lamb grouping over time provided the best estimates of lamb growth rate. Ewe lameness and mastitis cases had a deleterious impact on lamb growth. Lambs from ewes identified with mastitis during lactation were on average 3.0 (standard error (SE) 1.6) kg lighter during the four month growth period than lambs from unaffected ewes. Lambs from ewes that were not lame during pregnancy were 3.0 (SE 1.2) kg heavier at eight weeks of age than lambs from ewes with a least one lameness case during the same period. Lambs from ewes lame either during the first 4 weeks or between 4-8 weeks of a lamb's life (but not lame during pregnancy) were also significantly heavier at 56 days of age, than lambs reared by ewes that were lame during pregnancy (2.8 (SE 1.2) and 3.4 (SE 1.2) kg respectively). Cases of pneumonia and bacterial arthritis in lambs had a significant negative impact on lamb growth with affected lambs being on average 5.5 (SE 1.1) kg and 2.2 (SE 1.2) kg less than non-affected lambs respectively after the disease event. Prior to a case of lameness or pneumonia, lambs were significantly heavier than unaffected lambs suggesting a possible trade-off between growth and immune function. Overall, the study provides evidence that that a combination of ewe and lamb characteristics and disease events play an important role in determining lamb growth rate and that heavier lambs may be more susceptible to disease.

DirtyGenes: testing for significant changes in gene or bacterial population compositions from a small number of samples.

High throughput genomics technologies are applied widely to microbiomes in humans, animals, soil and water, to detect changes in bacterial communities or the genes they carry, between different environments or treatments. We describe a method to test the statistical significance of differences in bacterial population or gene composition, applicable to metagenomic or quantitative polymerase chain reaction data. Our method goes beyond previous published work in being universally most powerful, thus better able to detect statistically significant differences, and through being more reliable for smaller sample sizes. It can also be used for experimental design, to estimate how many samples to use in future experiments, again with the advantage of being universally most powerful. We present three example analyses in the area of antimicrobial resistance. The first is to published data on bacterial communities and antimicrobial resistance genes (ARGs) in the environment; we show that there are significant changes in both ARG and community composition. The second is to new data on seasonality in bacterial communities and ARGs in hooves from four sheep. While the observed differences are not significant, we show that a minimum group size of eight sheep would provide sufficient power to observe significance of similar changes in further experiments. The third is to published data on bacterial communities surrounding rice crops. This is a much larger data set and is used to verify the new method. Our method has broad uses for statistical testing and experimental design in research on changing microbiomes, including studies on antimicrobial resistance.

Discrimination of contagious and environmental strains of Streptococcus uberis in dairy herds by means of mass spectrometry and machine-learning.

Streptococcus uberis is one of the most common pathogens of clinical mastitis in the dairy industry. Knowledge of pathogen transmission route is essential for the selection of the most suitable intervention. Here we show that spectral profiles acquired from clinical isolates using matrix-assisted laser desorption ionization/time of flight (MALDI-TOF) can be used to implement diagnostic classifiers based on machine learning for the successful discrimination of environmental and contagious S. uberis strains. Classifiers dedicated to individual farms achieved up to 97.81% accuracy at cross-validation when using a genetic algorithm, with Cohen's kappa coefficient of 0.94. This indicates the potential of the proposed methodology to successfully support screening at the herd level. A global classifier developed on merged data from 19 farms achieved 95.88% accuracy at cross-validation (kappa 0.93) and 70.67% accuracy at external validation (kappa 0.34), using data from another 10 farms left as holdout. This indicates that more work is needed to develop a screening solution successful at the population level. Significant MALDI-TOF spectral peaks were extracted from the trained classifiers. The peaks were found to correspond to bacteriocin and ribosomal proteins, suggesting that immunity, growth and competition over nutrients may be correlated to the different transmission routes.

Drivers for precision livestock technology adoption: A study of factors associated with adoption of electronic identification technology by commercial sheep farmers in England and Wales.

The UK is the largest lamb meat producer in Europe. However, the low profitability of sheep farming sector suggests production efficiency could be improved. Although the use of technologies such as Electronic Identification (EID) tools could allow a better use of flock resources, anecdotal evidence suggests they are not widely used. The aim of this study was to assess uptake of EID technology, and explore drivers and barriers of adoption of related tools among English and Welsh farmers. Farm beliefs and management practices associated with adoption of this technology were investigated. A total of 2000 questionnaires were sent, with a response rate of 22%. Among the respondents, 87 had adopted EID tools for recording flock information, 97 intended to adopt it in the future, and 222 neither had adopted it, neither intended to adopt it. Exploratory factor analysis (EFA) and multivariable logistic regression modelling were used to identify farmer beliefs and management practices significantly associated with adoption of EID technology. EFA identified three factors expressing farmer's beliefs-external pressure and negative feelings, usefulness and practicality. Our results suggest farmer's beliefs play a significant role in technology uptake. Non-adopters were more likely than adopters to believe that 'government pressurise farmers to adopt technology'. In contrast, adopters were significantly more likely than non-adopters to see EID as practical and useful (p≤0.05). Farmers with higher information technologies literacy and intending to intensify production in the future were significantly more likely to adopt EID technology (p≤0.05). Importantly, flocks managed with EID tools had significantly lower farmer- reported flock lameness levels (p≤0.05). These findings bring insights on the dynamics of adoption of EID tools. Communicating evidence of the positive effects EID tools on flock performance and strengthening farmer's capability in use of technology are likely to enhance the uptake of this technology in sheep farms.

The Applied Development of a Tiered Multilocus Sequence Typing (MLST) Scheme for Dichelobacter nodosus.

Dichelobacter nodosus (D. nodosus) is the causative pathogen of ovine footrot, a disease that has a significant welfare and financial impact on the global sheep industry. Previous studies into the phylogenetics of D. nodosus have focused on Australia and Scandinavia, meaning the current diversity in the United Kingdom (U.K.) population and its relationship globally, is poorly understood. Numerous epidemiological methods are available for bacterial typing; however, few account for whole genome diversity or provide the opportunity for future application of new computational techniques. Multilocus sequence typing (MLST) measures nucleotide variations within several loci with slow accumulation of variation to enable the designation of allele numbers to determine a sequence type. The usage of whole genome sequence data enables the application of MLST, but also core and whole genome MLST for higher levels of strain discrimination with a negligible increase in experimental cost. An MLST database was developed alongside a seven loci scheme using publically available whole genome data from the sequence read archive. Sequence type designation and strain discrimination was compared to previously published data to ensure reproducibility. Multiple D. nodosus isolates from U.K. farms were directly compared to populations from other countries. The U.K. isolates define new clades within the global population of D. nodosus and predominantly consist of serogroups A, B and H, however serogroups C, D, E, and I were also found. The scheme is publically available at https://pubmlst.org/dnodosus/.

Prediction of Streptococcus uberis clinical mastitis risk using Matrix-assisted laser desorption ionization time of flight mass spectrometry (MALDI-TOF MS) in dairy herds.

The purpose of this study was to evaluate whether the risk of Streptococcus uberis clinical mastitis at cow level could be predicted from the historical presence of specific strains of S. uberis on dairy farms. Matrix-assisted laser desorption ionization time of flight mass spectrometry was used to identify S. uberis isolates potentially capable of contagious transmission. Data were available from 10,652 cows from 52 English and Welsh dairy farms over a 14 month period, and 521 isolates of S. uberis from clinical mastitis cases were available for analysis. As well as the temporal herd history of clinical mastitis associated with particular S. uberis strains, other exposure variables included cow parity, stage of lactation, milk yield, and somatic cell count. Observations were structured longitudinally as repeated weekly measures through the study period for each cow. Data were analyzed in a Bayesian framework using multilevel logistic regression models. Similarity of mass spectral profiles between isolates of S. uberis from consecutive clinical cases of mastitis in herds was used to indicate potential for contagious phenotypic characteristics. Cross validation showed that new isolates with these characteristics could be identified with an accuracy of 90% based on bacterial protein mass spectral characteristics alone. The cow-level risk in any week of these S. uberis clinical mastitis cases increased with the presence of the same specific strains of S. uberis in other cows in the herd during the previous 2 weeks. The final statistical model indicated there would be a 2-3 fold increase in the risk of S. uberis clinical mastitis associated with particular strains if these occurred in the herd 1 and 2 weeks previously. The results suggest that specific strains of S. uberis may be involved with contagious transmission, and predictions based on their occurrence could be used as an early warning surveillance system to enhance the control of S. uberis mastitis.

Quantitative analysis of antibiotic usage in British sheep flocks.

The aim of this study was to examine the variation in antibiotic usage between 207 commercial sheep flocks using their veterinary practice prescribing records. Mean and median prescribed mass per population corrected unit (mg/PCU) was 11.38 and 5.95, respectively and closely correlated with animal defined daily dose (ADDD) 1.47 (mean), 0.74 (median) (R2=0.84, P<0.001). This is low in comparison with the suggested target (an average across all the UK livestock sectors) of 50 mg/PCU. In total, 80 per cent of all antibiotic usage occurred in the 39 per cent of flocks where per animal usage was greater than 9.0 mg/PCU. Parenteral antibiotics, principally oxytetracycline, represented 82 per cent of the total prescribed mass, 65.5 per cent of antibiotics (mg/PCU) were prescribed for the treatment of lameness. Oral antibiotics were prescribed to 49 per cent of flocks, 64 per cent of predicted lamb crop/farm. Lowland flocks were prescribed significantly more antibiotics than hill flocks. Variance partitioning apportioned 79 per cent of variation in total antibiotic usage (mg/PCU) to the farm level and 21 per cent to the veterinary practice indicating that veterinary practices have a substantial impact on overall antimicrobial usage. Reducing antibiotic usage in the sheep sector should be possible with better understanding of the drivers of high usage in individual flocks and of veterinary prescribing practices.

Quantitative analysis of antimicrobial use on British dairy farms.

Antimicrobial resistance has been reported to represent a growing threat to both human and animal health, and concerns have been raised around levels of antimicrobial usage (AMU) within the livestock industry. To provide a benchmark for dairy cattle AMU and identify factors associated with high AMU, data from a convenience sample of 358 dairy farms were analysed using both mass-based and dose-based metrics following standard methodologies proposed by the European Surveillance of Veterinary Antimicrobial Consumption project. Metrics calculated were mass (mg) of antimicrobial active ingredient per population correction unit (mg/PCU), defined daily doses (DDDvet) and defined course doses (DCDvet). AMU on dairy farms ranged from 0.36 to 97.79 mg/PCU, with a median and mean of 15.97 and 20.62 mg/PCU, respectively. Dose-based analysis ranged from 0.05 to 20.29 DDDvet, with a median and mean of 4.03 and 4.60 DDDvet, respectively. Multivariable analysis highlighted that usage of antibiotics via oral and footbath routes increased the odds of a farm being in the top quartile (>27.9 mg/PCU) of antimicrobial users. While dairy cattle farm AMU appeared to be lower than UK livestock average, there were a selection of outlying farms with extremely high AMU, with the top 25 per cent of farms contributing greater than 50 per cent of AMU by mass. Identification of these high use farms may enable targeted AMU reduction strategies and facilitate a significant reduction in overall dairy cattle AMU.