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

Factors associated with heartworm preventative use in the golden retriever lifetime study

Introduction

Heartworm disease is preventable with use of heartworm preventatives, but the reported prevalence of heartworm preventative use in the United States is low, some estimates falling around 50% of dogs. However, there are very few estimates of prevalence and its associated factors.

Methods

We aimed to estimate prevalence and evaluate factors, including vaccination status, demographics, lifestyle, physical conditions, medications and supplements, and environment and living conditions, for their association with heartworm preventative use in a large dataset from the Golden Retriever Lifetime Study (N = 2,998). Due to the large number of predictors evaluated, we built a bootstrapped elastic net logistic regression model, which is robust to overfitting and multicollinearity. Variables were evaluated by calculating covariate stability (>80%) and statistical significance (p<0.02).

Results

In our sample, the prevalence of heartworm use was 39.5%. In our elastic net model, receiving vaccinations (rabies, Bordetella, or any other vaccine), being located in the Southern U.S., being altered, having an infectious disease or ear/ nose/throat system disease diagnosis, being on heartworm preventatives in the past, currently being on tick preventative, having sun exposure in an area with concrete flooring, living in a house with more rooms with carpeted floors, and spending time on hardwood flooring inside were associated with greater odds of heartworm preventative use. Supplementation use and being in the top quartile of height were associated with lower odds of heartworm preventative use.

Discussion

The explanatory factors we identified can be used to improve client communication. In addition, target populations for educational interventions and outreach can be identified. Future studies can validate the findings in a more diverse population of dogs.

Hoof lesions in partly housed pasture-based dairy cows

Lameness is a symptom of a painful disorder affecting the limbs, which impacts dairy cow welfare and productivity. Lameness is primarily caused by hoof lesions. The prevalence of different lesion types can differ depending on environmental conditions and farm management practices. The aims of this observational study were to establish the cow-level and herd-level lesion prevalence during both housing and grazing periods in a partly housed, pasture-based system, establish the prevalence of lesions always associated with pain ("alarm" lesion), identify the lesions associated with a higher lameness score, determine relationships between lesions, and identify risk factors for digital dermatitis. On 98 farms during the grazing period and on 74 of the same farms during the housing period, every cow was lameness scored (0-3 lameness scoring scale), and the hind hooves of lame cows (score 2 and 3) were examined (maximum 20 cows per visit) and the prevalence of each lesion type recorded. To gather data on potential predictors for the risk factor analysis, a questionnaire with the farmer was conducted on lameness management practices and infrastructure measurements were taken at each visit. Cow-level data were also collected (e.g., parity, breed, milk yield, and so on). Noninfectious lesions were found to be more prevalent than infectious lesions in this system type. The most prevalent lesion types during both grazing and housing periods were white line separation, sole hemorrhages and overgrown claws; all remaining lesions had a cow-level prevalence of less than 15%. The cow-level prevalence of alarm lesions was 19% during the grazing period and 25% during the housing period; the most prevalent alarm lesion was sole ulcers during both periods. We found significantly more foreign bodies within the hoof sole (grazing = 14%, housing = 7%) and overgrown claws (grazing = 71%, housing = 55%) during the grazing period compared with the housing period. Cows with foul of the foot, sole ulcer, white line abscess, toe necrosis or an amputated claw had higher odds of being more severely lame, compared with mildly lame. The strongest correlation between lesions were between toe necrosis and digital dermatitis (r = 0.40), overgrown claws and corkscrew claws (r = 0.33), and interdigital hyperplasia and digital dermatitis (r = 0.31) at herd level. At the cow level, the strongest correlation was between overgrown claws and corkscrew claws (r = 0.27), and digital dermatitis and heel erosion (r = 0.22). The farmers' perception of the presence of digital dermatitis (and lameness) was significantly correlated with the actual presence of digital dermatitis recorded. Additional risk factors for the presence of digital dermatitis were cow track and verge width near the collecting yard, and stone presence on the cow tracks. Results from this study help further our understanding of the causes of lameness in partly housed, pasture-based dairy cows, and can be used to guide prevention and treatment protocols.

Stability selection for mixed effect models with large numbers of predictor variables: A simulation study

Covariate selection when the number of available variables is large relative to the number of observations is problematic in epidemiology and remains the focus of continued research. Whilst a variety of statistical methods have been developed to attempt to overcome this issue, at present very few methods are available for wide data that include a clustered outcome. The purpose of this research was to make an empirical evaluation of a new method for covariate selection in wide data settings when the dependent variable is clustered. We used 3300 simulated datasets with a variety of defined structures and known sets of true predictor variables to conduct an empirical evaluation of a mixed model stability selection procedure. Comparison was made with an alternative method based on regularisation using the least absolute shrinkage and selection operator (Lasso) penalty. Model performance was assessed using several metrics including the true positive rate (proportion of true covariates selected in a final model) and false discovery rate (proportion of variables selected in a final model that were non-true (false) variables). For stability selection, the false discovery rate was consistently low, generally remaining ≤ 0.02 indicating that on average fewer than 1 in 50 of the variables selected in a final model were false variables. This was in contrast to the Lasso-based method in which the false discovery rate was between 0.59 and 0.72, indicating that generally more than 60% of variables selected in a final model were false variables. In contrast however, the Lasso method attained higher true positive rates than stability selection, although both methods achieved good results. For the Lasso method, true positive rates remained ≥ 0.93 whereas for stability selection the true positive rate was 0.73-0.97. Our results suggest both methods may be of value for covariate selection with high dimensional data with a clustered outcome. When high specificity is needed for identification of true covariates, stability selection appeared to offer the better solution, although with a slight loss of sensitivity. Conversely when high sensitivity is needed, the Lasso approach may be useful, even if accompanied by a substantial loss of specificity. Overall, the results indicated the loss of sensitivity when employing stability selection is relatively small compared to the loss of specificity when using the Lasso and therefore stability selection may provide the better option for the analyst when evaluating data of this type.

Cow- and herd-level risk factors for lameness in partly housed pasture-based dairy cows

Lameness in dairy cows is a major animal welfare concern and has substantial economic impact through reduced production and fertility. Previous risk factor analyses have focused on housed systems, rather than those where cows were grazed for the majority of the year and housed only for the winter period. Therefore, the aim of this observational study was to identify a robust set of cow-level and herd-level risk factors for lameness in a pasture-based system, based on predictors from the housing and grazing periods. Ninety-nine farms were visited during the grazing period (April 2019-September 2019), and 85 farms were revisited during the housing period (October 2019-February 2020). At each visit, all lactating cows were scored for lameness (0 = good mobility, 1 = imperfect mobility, 2 = impaired mobility, 3 = severely impaired mobility), and potential herd-level risk factors were recorded through questionnaires and infrastructure measurements. Routine cow-level management data were also collected. Important risk factors for lameness were derived though triangulation of results from elastic net regression, and from logistic regression model selection using modified Bayesian information criterion. Both selection methods were implemented using bootstrapping. This novel approach has not previously been used in a cow-level or herd-level risk factor analysis in dairy cows, to the authors' knowledge. The binary outcome variable was lameness status, whereby cows with a lameness score of 0 or 1 were classed as non-lame and cows with a score of 2 or 3 were classed as lame. Cow-level risk factors for increased lameness prevalence were age and genetic predicted transmitting ability for lameness. Herd-level risk factors included farm and herd size, stones in paddock gateways, slats on cow tracks near the collecting yard, a sharper turn at the parlor exit, presence of digital dermatitis on the farm, and the farmers' perception of whether lameness was a problem on the farm. This large-scale study identified the most important associations between risk factors and lameness, based on the entire year (grazing and housing periods), providing a focus for future randomized clinical trials.

Evaluation of the use of antibiotic waste bins and medicine records to quantify antibiotic use on sheep, beef, and mixed species farms: A mixed methods study

There is a lack of robust data on antibiotic use on sheep and beef farms in the UK, particularly for farms with mixed species. On mixed farms, quantification of antibiotic use is reliant on accurate farmers' records as veterinary prescription data does not provide information at the species level. Previous studies that have investigated multiple antibiotic use collection methods were conducted on single species farms and failed to collect data on the reasons why differences in methods may exist. This study aimed to evaluate sheep and beef farmers' antibiotic recording practices by comparing quantities of antibiotics measured from medicine records and empty antibiotic packaging collection bins, and identify barriers and facilitators of the antibiotic use collection methods. Thirty-five farms were followed for a year period. Farmers were asked to record their antibiotic treatments and deposit empty antibiotic packaging used in sheep or beef cattle into a bin. Semi-structured qualitative interviews were conducted to understand the experiences of farmers taking part in the study and explore the possible differences in methods. Bins and medicine records were collected and the mass of active ingredient (mg) was calculated. The level of agreement between the two antibiotic use collection methods was measured using rank parameters of Kendall's T_a. The bins were 67 % (CI = 29-87 %) more likely to measure more antibiotic use than the medicine records. The scale of variability of the measurements between two random farms was 33 % (CI = 6-56 %) larger for the antibiotic waste bins than the scale of variability between the medicine records. Sheep farmers often missed neonatal lamb treatments off their medicine records, with a median of 32.5 missing treatments per farm (IQR = 18-130). Of the mixed species farms, 28 % of treatment entries were missing the species the antibiotic was used in. Farmers reported that the bin was easy to use but they also reported that there was a tendency to under-report actual use where there were multiple workers on the farm or where treatments were administered by the veterinarian. The qualitative analysis identified contextual barriers to accurate medicine recording, such as difficulties with animal identification, with recording in the field, and with recording during lambing time. This study demonstrated that there were significant differences in antibiotic use measured by the bins and the medicine records. The mixed-methods approach provided an understanding of the contextual barriers that impacted farmers' medicine recording and use of the bin. This information on the contextual barriers can be used to inform the design of data collection methods to improve antibiotic consumption data in the sheep and beef sectors.

Quantifying the value of on-farm measurements to inform the selection of key performance indicators for livestock production systems

The use of key performance indicators (KPIs) to assist on-farm decision making has long been seen as a promising strategy to improve operational efficiency of agriculture. The potential benefit of KPIs, however, is heavily dependent on the economic relevance of the metrics used, and an overabundance of ambiguously defined KPIs in the livestock industry has disincentivised many farmers to collect information beyond a minimum requirement. Using high-resolution sheep production data from the North Wyke Farm Platform, a system-scale grazing trial in southwest United Kingdom, this paper proposes a novel framework to quantify the information values of industry recommended KPIs, with the ultimate aim of compiling a list of variables to measure and not to measure. The results demonstrated a substantial financial benefit associated with a careful selection of metrics, with top-ranked variables exhibiting up to 3.5 times the information value of those randomly chosen. When individual metrics were used in isolation, ewe weight at lambing had the greatest ability to predict the subsequent lamb value at slaughter, surpassing all mid-season measures representing the lamb's own performance. When information from multiple metrics was combined to inform on-farm decisions, the peak benefit was observed under four metrics, with inclusion of variables beyond this point shown to be detrimental to farm profitability regardless of the combination selected. The framework developed herein is readily extendable to other livestock species, and with minimal modifications to arable and mixed agriculture as well.

Multiple model triangulation to identify factors associated with lameness in British sheep flocks

•

Multiple model triangulation identifies variables that are likely true positives.

•

Triangulation increases confidence in which managements to recommend in practice.

•

Effective management of ewes can lower prevalence of lameness in ewes and lambs.

Identification of factors associated with an outcome can be challenging when the number of explanatory variables is large in relation to the number of observations. Multiple model triangulation, where results from several model types are combined, improves the likelihood of identifying true predictor variables. The aim of this study was to use triangulation to identify covariates likely to be truly associated with the prevalence of lameness in sheep flocks in Great Britain.

Data were collected using a questionnaire sent to 3200 sheep farmers in Great Britain in 2018. The useable response rate was 14.1 %. The geometric mean prevalence of lameness was 1.4 % (95 % CI 1.2−1.7) for ewes, and 0.6 % (95 % CI 0.5−0.9) for lambs, however, approximately 60 % flocks had >2% prevalence of lameness in ewes.

Four model types were investigated, two generalised linear models (negative binomial and quasi-Poisson) built using stepwise selection, and two elastic net models (Poisson and Gaussian distributions) refined with selection stability estimation.

Triangulated covariates were those selected in three or all four models – 10 for ewes and 12 for lambs. Higher prevalence of lameness in ewes was associated with 5−100% feet bleeding during routine foot trimming compared with not foot trimming, footbathing the flock to treat severe footrot (SFR) and always using formalin in footbaths, both compared with not footbathing, using FootVax™ for <1 year compared with not using FootVax™, and never quarantining new or returning sheep to the farm for >3 weeks compared with always. Lower prevalence of lameness in ewes was associated with vaccinating with FootVax™ for >5 years compared with not vaccinating, peat soil compared with no peat soil, and having no lame ewes to treat.

Higher prevalence of lameness in lambs was associated with 5−100% feet bleeding during routine foot trimming, always foot trimming ewes with SFR, not knowingly selecting replacement ewes from ewes that were never lame compared with always, replacement sheep purchased and homebred compared with only homebred, treating lambs >3 days after recognition of lameness compared with 0-3 days and footbathing the flock to treat interdigital dermatitis compared with not footbathing at all. Lower prevalence of lameness in lambs was associated with peat soil, flocks in Scotland versus England, an altitude of >230−500 m compared with ≤230 m, never using antibiotic injection to treat lambs with SFR compared with always, and having no lame lambs to treat.

We conclude triangulation identified reliable management practices for farmers to implement to minimise lameness in sheep.

From the other perspective: Behavioural factors associated with UK sheep farmers' attitudes towards antibiotic use and antibiotic resistance

Research suggests that many sheep farmers continue to carry out traditional antibiotic use practices despite new ’good practice’ recommendations. The aim of this study was to group farmers depending on their attitudes around antibiotic use and antibiotic resistance, and determine the behaviours that are associated with the farmers in these groups. In 2017, a flock health survey was sent to British sheep farmers. K-means cluster analysis was used to identify groups of farmers with similar attitudes towards antibiotic use and resistance. A multivariable logistic regression model was built to determine the associations between farmers’ past behaviours and their antibiotic attitude group. There were 461 responses. Two groups of farmers were identified based on their antibiotic attitudes. Cluster 1 were defined as the "discordant" group who had positive views of using antibiotics prophylactically and negative views of reducing antibiotic use. Cluster 2 were defined as the "concordant" group who were positive about reducing antibiotic use and had negative views about using antibiotics prophylactically. Using antibiotics in all lambs (OR = 2.689, CI = 1.571, 4.603), using antibiotics in all ewes (OR = 3.388, CI = 1.318, 8.706), always trimming diseased feet over the past three years (OR = 2.487, CI = 1.459, 4.238), not using a computer to record information over the past three years (OR = 1.996, CI = 1.179, 3.381), not changing worming practices over the past three years (OR = 1.879, CI = 1.144, 3.087), and farmers’ perceptions that their sheep flock did not make a financial loss in the past three years (OR = 2.088, CI = 1.079, 4.040) were significantly associated with belonging to the discordant group. Talking to their veterinarian about antibiotic use or the frequency of veterinary visits were not associated with antibiotic attitude group. These results suggest that farmers who had attitudes relating to antibiotic use that did not align with current recommendations carried out more traditional practices, which were strengthened by their positive perceptions of profitability.

Factors associated with daily weight gain in preweaned calves on dairy farms

The preweaning period is vital in the development of calves on dairy farms and improving daily liveweight gain (DLWG) is important to both financial and carbon efficiency; minimising rearing costs and improving first lactation milk yields. In order to improve DLWG, veterinary advisors should provide advice that has both a large effect size as well as being consistently important on the majority of farms. Whilst a variety of factors have previously been identified as influencing the DLWG of preweaned calves, it can be challenging to determine their relative importance, which is essential for optimal on-farm management decisions. Regularised regression methods such as ridge or lasso regression provide a solution by penalising variable coefficients unless there is a proportional improvement in model performance. Elastic net regression incorporates both lasso and ridge penalties and was used in this research to provide a sparse model to accommodate strongly correlated predictors and provide robust coefficient estimates. Sixty randomly selected British dairy farms were enrolled to collect weigh tape data from preweaned calves at birth and weaning, resulting in data being available for 1014 calves from 30 farms after filtering to remove poor quality data, with a mean DLWG of 0.79 kg/d (range 0.49-1.06 kg/d, SD 0.13). Farm management practices (e.g. colostrum, feeding, hygiene protocols), building dimensions, temperature/humidity and colostrum quality/bacteriology data were collected, resulting in 293 potential variables affecting farm level DLWG. Bootstrapped elastic net regression models identified 17 variables as having both a large effect size and high stability. Increasing the maximum preweaned age within the first housing group (0.001 kg/d per 1d increase, 90 % bootstrap confidence interval (BCI): 0.000-0.002), increased mean environmental temperature within the first month of life (0.012 kg/d per 1 °C increase, 90 % BCI: 0.002-0.037) and increased mean volume of milk feeding (0.012 kg/d per 1 L increase, 90 % BCI: 0.001-0.024) were associated with increased DLWG. An increase in the number of days between the cleaning out of calving pen (-0.001 kg/d per 1d increase, 90 % BCI: -0.001-0.000) and group housing pens (-0.001 kg/d per 1d increase, 90 % BCI: -0.002-0.000) were both associated with decreased DLWG. Through bootstrapped elastic net regression, a small number of stable variables have been identified as most likely to have the largest effect size on DLWG in preweaned calves. Many of these variables represent practical aspects of management with a focus around stocking demographics, milk/colostrum feeding, environmental hygiene and environmental temperature; these variables should now be tested in a randomised controlled trial to elucidate causality.

Model selection for inferential models with high dimensional data: synthesis and graphical representation of multiple techniques

Inferential research commonly involves identification of causal factors from within high dimensional data but selection of the 'correct' variables can be problematic. One specific problem is that results vary depending on statistical method employed and it has been argued that triangulation of multiple methods is advantageous to safely identify the correct, important variables. To date, no formal method of triangulation has been reported that incorporates both model stability and coefficient estimates; in this paper we develop an adaptable, straightforward method to achieve this. Six methods of variable selection were evaluated using simulated datasets of different dimensions with known underlying relationships. We used a bootstrap methodology to combine stability matrices across methods and estimate aggregated coefficient distributions. Novel graphical approaches provided a transparent route to visualise and compare results between methods. The proposed aggregated method provides a flexible route to formally triangulate results across any chosen number of variable selection methods and provides a combined result that incorporates uncertainty arising from between-method variability. In these simulated datasets, the combined method generally performed as well or better than the individual methods, with low error rates and clearer demarcation of the true causal variables than for the individual methods.

Quantitative Analysis of Colostrum Bacteriology on British Dairy Farms

Total bacterial counts (TBC) and coliform counts (CC) were estimated for 328 colostrum samples from 56 British dairy farms. Samples collected directly from cows' teats had lower mean TBC (32,079) and CC (21) than those collected from both colostrum collection buckets (TBC: 327,879, CC: 13,294) and feeding equipment (TBC: 439,438, CC: 17,859). Mixed effects models were built using an automated backwards stepwise process in conjunction with repeated bootstrap sampling to provide robust estimates of both effect size and 95% bootstrap confidence intervals (BCI) as well as an estimate of the reproducibility of a variable effect within a target population (stability). Colostrum collected using parlor (2.06 log cfu/ml, 95% BCI: 0.35–3.71) or robot (3.38 log cfu/ml, 95% BCI: 1.29–5.80) milking systems, and samples collected from feeding equipment (2.36 log cfu/ml, 95% BCI: 0.77–5.45) were associated with higher TBC than those collected from the teat, suggesting interventions to reduce bacterial contamination should focus on the hygiene of collection and feeding equipment. The use of hot water to clean feeding equipment (−2.54 log cfu/ml, 95% BCI: −3.76 to −1.74) was associated with reductions in TBC, and the use of peracetic acid (−2.04 log cfu/ml, 95% BCI: −3.49 to −0.56) or hypochlorite (−1.60 log cfu/ml, 95% BCI: −3.01 to 0.27) to clean collection equipment was associated with reductions in TBC compared with water. Cleaning collection equipment less frequently than every use (1.75 log cfu/ml, 95% BCI: 1.30–2.49) was associated with increased TBC, the use of pre-milking teat disinfection prior to colostrum collection (−1.85 log cfu/ml, 95% BCI: −3.39 to 2.23) and the pasteurization of colostrum (−3.79 log cfu/ml, 95% BCI: −5.87 to −2.93) were associated with reduced TBC. Colostrum collection protocols should include the cleaning of colostrum collection and feeding equipment after every use with hot water as opposed to cold water, and hypochlorite or peracetic acid as opposed to water or parlor wash. Cows' teats should be prepared with a pre-milking teat disinfectant and wiped with a clean, dry paper towel prior to colostrum collection, and colostrum should be pasteurized where possible.

Uptake of the lameness Five-Point Plan and its association with farmer-reported lameness prevalence: A cross-sectional study of 532 UK sheep farmers

The aims of this research were to determine the uptake of a national strategy to reduce lameness in the UK flock, known as the Five-Point Plan (5 P P); explore the association between footrot vaccination (Footvax®) use and 5 P P adoption; investigate the management practices associated with farmer-reported percentage lameness through risk factor analysis; and identify the population attributable fractions of these management practices. In 2014, the 5 P P was launched to provide a practical, farm-level framework to help farmers reduce lameness to reach Farm Animal Welfare Committee (FAWC) targets. No published studies have explicitly explored its uptake in UK flocks nor its association with lameness prevalence. Understanding what parts of the 5 P P farmers adopt and which elements contribute towards the greatest reduction in lameness are integral in informing future strategies. Between November 2018 and February 2019, 532 UK sheep farmers completed a cross-sectional online and paper-based survey. The geometric mean of farmer-reported percentage lameness in ewes was 3.2 % (95 % CI: 2.8-3.6). Farmers adopted a median of 3 points of the plan, but was only fully-adopted by 5.8 % of farmers. The number of points adopted increased with flock size, with larger commercial flocks more likely to cull and vaccinate against footrot, but smaller, pedigree flocks were more likely to treat individual lame sheep. Vaccination was poorly associated with the uptake of other points of the 5 P P. Eight flock management factors were associated with significantly higher percentage lameness in ewes; not carrying out measures to avoid lameness transmission, not quarantining bought in stock, not treating individual lame sheep within three days, maintaining an open flock and foot trimming were all associated with a higher risk of lameness in flocks studied. In addition, using Footvax® for ≤5 years was associated with a higher risk of lameness, although vaccination could be a consequence of high flock lameness or these farmers were not implementing other effective managements, such as treating promptly. The highest PAFs were calculated for trimming lame sheep (16.9 %), maintaining an open flock (13.5 %) and not carrying out measures to avoid lameness transmission (11.8 %). We provide new evidence documenting the benefits of adopting parts of the 5 P P on reducing lameness prevalence in UK flocks, although uptake of these measures could be improved in flocks. Encouraging uptake of these measures could make an important contribution towards reducing the prevalence of lameness and reaching 2021 FAWC ≤ 2% lameness prevalence targets.

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.