Technical note: Assessing lameness in tie-stalls using live stall lameness scoring

Video stall lameness scoring (SLS) has been shown to be comparable to video locomotion scoring for evaluating lameness in dairy cows housed in tie-stalls and may be a more practical and easier method to measure lameness in a herd. We compared live SLS to video SLS and to live locomotion scoring. A total of 685 lactating cows subsampled from 27 commercial dairy herds were examined for lameness through live and video SLS. Cows scored with the live or video SLS system were scored for 4 behavioral indicators while still in their stall: weight shifting (shift), standing on the edge of the stall (edge), uneven weight bearing while standing (rest), and uneven weight bearing while moving from side to side (uneven). Two observers live scored and video scored for SLS. Lameness prevalence from video SLS and live SLS were similar (31 vs. 30%, respectively). Prevalence of the behavioral indicators varied from 0.59 to 58.2%. Sensitivity and specificity of live SLS was calculated using video SLS as the gold standard for lameness detection in tie-stalls. Sensitivity of live SLS was 0.83 and specificity was 0.94. False positives and false negatives for lameness were 14.4 and 16.8%, respectively. When comparing the prevalence of lameness measured through video or live SLS at the herd level, live SLS for lameness was correlated to video SLS (r = 0.91) with a Cohen's kappa coefficient of 0.79 (95% confidence limit = 0.73–0.84). Average exact agreement in the behavior indicators observed ranged from 80 to 100%. A subsample of 250 cows from 5 herds were scored for live SLS and live locomotion by a third observer. Intra- and interobserver reliability for live SLS and live locomotion scoring were found to have a kappa coefficient of 0.53 (95% confidence limit = 0.43–0.64) when determining a cow as lame through SLS or locomotion scoring. Live SLS was correlated with live locomotion scoring (r = 0.92). However, lameness prevalence was lower when using live SLS (28.4%) compared with locomotion scoring (38%). In summary, live SLS may be an acceptable method to replace video SLS to identify lame cows and rank tie-stall herds in terms of lameness prevalence without having to remove the cows from their stalls or view videos to score; however, it may underestimate lameness prevalence compared with locomotion scoring methods.     

The impact of clinical lameness on the milk yield of dairy cows

This paper investigates the impact of lameness on milk yield. The dataset includes approximately 8000 test-day milk yields from 900 cows on five farms in Gloucester, UK, collected over 18 mo from 1997 to 1999. The data were structured to account for repeated measures of test-day yield (1 to 10 per cow) and analyzed to account for this autocorrelation. Factors affecting milk yield included: farm of origin, stage of lactation, parity, and whether a cow ever became lame. In clinically lame cows, milk yield was reduced from up to 4 mo before a case of lameness was diagnosed and treated and for the 5 mo after treatment. The total mean estimated reduction in milk yield per 305-d lactation was approximately 360 kg. We conclude that clinical lameness has a significant impact on milk production. This is important information for assessing the economic impact of clinical lameness and its impact on cow health. It adds weight to the importance of early identification of clinical lameness and the urgency of techniques to improve the definition of this highly subjective diagnosis.     

Lying behavior as an indicator of lameness in dairy cows

Lameness is widely recognized as one of the most serious welfare and production concerns in the dairy industry. Our objectives were to evaluate the associations between lying behavior and lameness, and to determine whether lying behavior can be used as a diagnostic tool for lameness. Electronic data loggers recorded lying behavior of 1,319 cows from 28 farms at 1-min intervals for 5 d. These cows were gait scored according to a 5-point Numerical Rating System (NRS), and categorized as NRS ≤2, NRS = 3, or NRS = 4; no cow was scored as NRS = 5. Lameness was dichotomized twice: LAME (NRS ≥3) and SEVLAME (NRS = 4). Data were divided into 2 groups: 11 farms using deep-bedded stalls (DB) and 17 farms using mattress stalls (MAT). Differences in the daily lying time (h/d), frequency of lying bouts (n/d), duration of lying bouts (min/bout), and the standard deviation of bout duration (min/bout) between LAME or SEVLAME cows and those that were not were tested using mixed models. Receiver operating characteristic curves were constructed to identify behavioral thresholds to distinguish SEVLAME cows from the rest. Odds ratios for SEVLAME were estimated using logistic regression. Overall, 28.5% of cows were LAME including 7.3% that were SEVLAME. The prevalence of SEVLAME was higher on MAT farms than on DB farms (9.3 ± 1.3 vs. 4.4 ± 1.2%, respectively). SEVLAME cows on DB farms spent 12.8 [confidence interval (CI): 12.0 to 13.7] h/d lying down compared with 11.2 (CI: 10.7 to 11.8) h/d for cows that were not SEVLAME. These cows had longer duration of lying bouts [95.3 (CI: 84.6 to 107.3) vs. 80.3 (CI: 74.9 to 86.1) min/bout] and greater SD of bout duration [44.4 (CI: 41.1 to 48.0) vs. 50.7 (CI: 44.1 to 58.3) min/bout]. There were no behavioral differences among lameness categories on MAT farms. Within DB farms, cows with lying times >14.5 h/d had 16.2 (5.8 to 45.2) times higher odds of being SEVLAME. Cows with average lying bouts >90 min/bout were at 3.0 (1.2 to 7.4) times higher odds of being SEVLAME, and cows with average SD of bout duration >55 min/bout were at 4.1 (1.7 to 9.9) times higher odds of being SEVLAME. These results show that high lying times, long lying bouts, and variability in the duration of lying bouts were associated with lameness, and that stall surface influenced the behavioral responses of lame cows.     

Effects of routine treatment with nonsteroidal anti-inflammatory drugs at calving and when lame on the future probability of lameness and culling in dairy cows: A randomized controlled trial

Claw horn lesions (CHL) are reported as the most common cause of lameness in intensive dairy systems. Despite their prevalence, the underlying pathological mechanisms and preventive strategies for CHL remain poorly understood. Recent advances have pointed to the role of inflammation in disease aetiopathogenesis. Moderating inflammation from first calving may lead to long-term benefits and a viable intervention for treating and preventing disease. We conducted a 34-mo randomized controlled trial to investigate the effects of routine treatment with the nonsteroidal anti-inflammatory drug ketoprofen at calving and during treatment for lameness, on the future probability of lameness and culling, caused by exposure to normal farm conditions. A cohort of dairy heifers were recruited from a single, commercial dairy herd between January 8, 2018, and June 22, 2020, and randomly allocated to one of 4 treatment groups before first calving. The lactating herd was lameness scored every 2 wk on a 0 to 3 scale, to identify animals that became lame (single score ≥2a) and hence required treatment. Animals in group 1 received a therapeutic trim and a hoof block on the sound claw (if deemed necessary) every time they were treated for lameness. Animals in group 2 received the same treatment as group 1 with the addition of a 3-d course of ketoprofen (single dose daily) every time they were treated for lameness. Animals in group 3 received the same treatment as group 2 with the addition of a 3-d course of ketoprofen (single dose daily) starting 24 to 36 h after each calving. Animals in group 4 received a 3-d course of ketoprofen (single dose daily) every time they were identified with lameness. No therapeutic trim was administered to this group, unless they were identified as severely lame (a single score ≥3a). Animals were followed for the duration of the study (ending October 23, 2020). Probability of lameness was assessed by a lameness outcome score collected every 14 d. Data on culling was extracted from farm records. One hundred thirty-two animals were recruited to each group, with data from 438 animals included in the final analysis (111 in group 1, 117 in group 2, 100 in group 3, and 110 in group 4). Mixed effect logistic regression models were used to evaluate the effect of treatment group on the ongoing probability of lameness. Compared with the control group (group 1), animals in group 3 were less likely to become lame (odds ratio: 0.66) and severely lame (odds ratio: 0.28). A Cox proportional hazards survival model was used to investigate the effect of treatment group on time to culling. Compared with group 1, animals in groups 2 and 3 were at reduced risk of culling (hazard ratios: 0.55 and 0.56, respectively). The lameness effect size we identified was large and indicated that treating a cohort of animals with the group 3 protocol, would lead to an absolute reduction in population lameness prevalence of approximately 10% and severe lameness prevalence of 3%, compared with animals treated in accordance with conventional best practice (group 1).     

Associations between housing and management practices and the prevalence of lameness,hock lesions,and thin cows on US dairy operations

The objective of this study was to determine the association among different housing and management practices on the prevalence of lameness, hock lesions, and thin cows on US dairy operations. This study was conducted as part of the National Animal Health Monitoring System's Dairy 2014 study, which included dairy operations in 17 states. Size categories were assigned as follows: small (30–99 cows), medium (100–499 cows), and large (≥500 cows). Trained assessors visited 191 dairy operations from March through July 2014 and recorded locomotion and hock scores (on a 3-point scale), and the number of thin cows (body condition score ≤2.25) from a total of 22,622 cows (average 118 cows per farm). The majority of cows (90.4%) were considered to be sound (locomotion score = 1), 6.9% were mild/moderately lame (locomotion score = 2), and 2.7% were severely lame (locomotion score = 3). Similarly, most cows (87.3%) had no hock lesions (hock score = 1), 10.1% had mild lesions (hock score = 2), and 2.6% had severe hock lesions (hock score = 3). A low percentage of cows (4.2%) were thin. Univariate comparisons were performed using PROC LOGLINK, which accounts for study design and weighting. Variables meeting the univariate screening criterion of P < 0.20 were eligible for entry into multivariable models. Statistical significance in the multivariable models was declared at P < 0.05. Large operations had a lower within-herd prevalence of cows with locomotion score ≥2 and locomotion score = 3 compared with small or medium-sized operations. Operations on which cows were kept primarily on pasture had a lower percentage of locomotion score = 3 than those housed in freestall or open/dry lot operations. The use of sand bedding was associated with a lower within-herd prevalence of locomotion score ≥2 than straw/hay or dry/composted manure as the primary bedding material. Sand bedding was also associated with a lower within-herd prevalence of locomotion score = 3 than other bedding types except for rubber mats or mattresses. Operations that housed cows in an open/dry lot had a lower percentage of hock score ≥2 and hock score = 3 than other housing types. Providing sprinklers for heat abatement and having a nutritionist balance rations for cows was associated with a lower percentage of thin cows. Results from this study highlight management practices that may reduce the prevalence of lameness, hock lesions, and thin cows on dairy operations in the United States.     

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.     

Risk factors for lameness in freestall-housed dairy cows across two breeds, farming systems, and countries

Lameness poses a considerable problem in modern dairy farming. Several new developments (e.g., herd health plans) strive to help farmers improve the health and welfare of their herd. It was thus our aim to identify lameness risk factors common across regions, breeds, and farming systems for freestall-housed dairy cows. We analyzed data from 103 nonorganic and organic dairy farms in Germany and Austria that kept 24 to 145 Holstein Friesian or Fleckvieh cows in the milking herd (mean = 48). Data on housing, management, behavior, and lameness scores for a total of 3,514 cows were collected through direct observations and an interview. Mean lameness prevalence was 34% (range = 0-81%). Data were analyzed applying logistic regression with generalized estimating equations in a split-sample design. The final model contained 1 animal-based parameter and 3 risk factors related to lying as well as 1 nutritional animal-based parameter, while correcting for the significant confounders parity and data subset. Risk for lameness increased with decreasing lying comfort, that is, more frequent abnormal lying behavior, mats or mattresses used as a stall base compared with deep-bedded stall bases, the presence of head lunge impediments, or neck rail-curb diagonals that were too short. Cows in the lowest body condition quartile (1.25-2.50 for Holstein Friesian and 2.50-3.50 for Fleckvieh) had the highest risk of being lame. In cross-validation the model correctly classified 71 and 70% of observations in the model-building and validation samples, respectively. Only 2 out of 15 significant odds ratios (including contrasts) changed direction. They pertained to the 2 variables with the highest P-values in the model. In conclusion, lying comfort and nutrition are key risk areas for lameness in freestall-housed dairy cows. Abnormal lying behavior in particular proved to be a good predictor of lameness risk and should thus be included in on-farm protocols. The study is part of the European Commission's Welfare Quality® project.     

Assessment of lameness prevalence and associated risk factors in dairy herds in England and Wales

Visits were made to 205 dairy farms in England and Wales between October 2006 and May 2007 by 1 or more of 4 researchers. At each visit, all milking cows were locomotion scored (lameness scored) using a 4-point scale (0 = sound locomotion, 1 = imperfect locomotion, 2 = lame, 3 = severely lame). The mean prevalence of lameness (scores 2 and 3) across the study farms was 36.8% (range = 0-79.2%). On each farm, the presence within the housing and grazing environments of commonly reported risks for increased lameness was recorded. Each farmer was interviewed to gauge the ability of the farm staff to detect and treat lameness. A multivariable linear regression model was fitted. Risk factors for increased lameness were the presence of damaged concrete in yards, cows pushing each other or turning sharply near the parlor entrance or exit, cattle grazing pasture also grazed by sheep, the use of automatic scrapers, not treating lame cows within 48 h of detection, and cows being housed for 61 d or longer at the time they were locomotion scored by the visiting researcher. Having a herd consisting entirely of a breed or breeds other than Holstein-Friesian was associated with a reduction in lameness prevalence compared with having a herd consisting entirely of Holstein-Friesians.     

New insights into the association between lameness,behavior, and performance in Simmental cows

In a herd of 100 milking Simmental cows, data of performance and behavior parameters were collected automatically with different systems such as pedometers, an automatic milking system, and automatic weighing troughs for 1 yr. Performance measures were several milking-related parameters, live weight, as well as feed intake. Behavior-associated measures were feeding behavior (e.g. feeding duration, number of visits to the trough, and feeding pace) as well as activity such as lying duration, number of lying bouts, and overall activity. In the same time, lameness status of every cow was assessed with weekly locomotion scoring. According to the score animals were then classified lame (score 4 or 5) or nonlame (score 1, 2, or 3). From these data in total, 25 parameters summarized to daily values were evaluated for their ability to determine the lameness status of a cow. Data were analyzed with a regularized regression method called elastic net with the outcome lame or nonlame. The final model had a high prediction accuracy with an area under the curve of 0.91 [95% confidence interval (CI) = 0.88–0.94]. Specificity was 0.81 (95% CI = 0.73–0.85) and sensitivity was 0.94 (95% CI = 0.88–1.00). The most important factors associated with a cow being lame were number of meals, average feed intake per meal, and average duration of a meal. Lame cows fed in fewer and shorter meals with a decreased intake per meal. Milk yield and lying-behavior-associated parameters were relevant in the model, too, but only as parts of interaction terms demonstrating their strong dependence on other factors. A higher milk yield only resulted in higher risk of being lame if feed intake was decreased. The same accounts for lying duration: only if lying time was below the 50% quantile did an increased milk yield result in a higher risk of being lame. The association of lameness and daily lying duration was influenced by daily feeding duration and feeding duration at daytime. The results of the study give deeper insights on how the association between behavior and performance parameters and lameness is influenced by intrinsic factors in particular and that many of these have to be considered when trying to predict lameness based on such data. The findings lead to a better understanding why, for instance, lying duration or milk yield seem to be highly correlated with lameness in cows but still have not been overly useful as parameters in other lameness detection models.     

The hidden cost of disease: II. Impact of the first incidence of lameness on production and economic indicators of primiparous dairy cows

Lameness is a persistent and underreported health and welfare problem in the dairy industry, resulting in reduced cow performance and profitability as well as early culling. The study objectives were (1) to quantify the impact of the first instance of lameness, at different stages of lactation, on production and economic performance, and (2) to further quantify the impacts of the first instance of lameness when only cows that remain in the herd for at least 100 d in milk (DIM) and those that remain for 305 DIM are included in the analysis. A retrospective longitudinal study was conducted using pre-existing data from animal health records and Dairy Herd Improvement Association records. Data were edited based on selected inclusion criteria, yielding a data set containing records from 15,159 first-lactation Holstein cows from 120 herds with year of first calving between 2003 and 2014. Lame cows were assigned to 1 of 4 groups based on when in the lactation the first event of lameness occurred: transition (1–21 DIM), early lactation (22–100 DIM), mid-lactation (101–200 DIM), or late lactation (201+ DIM). Mid- and late-lactation lame cows were also stratified by cumulative milk yield before the lameness event. Healthy cows (i.e., no recorded lameness event) were randomly assigned for each lactation stage, with mid-lactation healthy and late-lactation healthy cows similarly stratified. Production performance (cumulative milk, fat, and protein yield) and economic performance [milk value, margin over feed cost (MOFC), and gross profit] were analyzed using a mixed model with herd as a random effect. Cumulative milk yields were 811 to 1,290 kg lower for lame cows than for healthy cows, with milk component yields undergoing similar reductions. Reductions in milk yield contributed to losses in milk value (?Can$527 to ?Can$1,083; ?US$419 to ?US$862) and MOFC (?Can$510 to ?Can$774; ?US$406 to ?US$616). Higher losses were reported using gross profit (?Can$753 to ?Can$1,052; ?US$599 to ?US$837), which includes all lameness-related costs. Production and performance losses were smaller when 100 DIM and 305 DIM thresholds were applied (i.e., exclusion of cows culled before 100 and 305 DIM, respectively), however, mid- and late-lactation lame cows maintained high levels of significant losses for all 6 variables analyzed. Lameness also led to higher levels of culling, masking losses for transition and early-lactation lame cows in the 305-DIM analysis. Increasing producer understanding of the costs associated with lameness not only serves to provide insight to producers for more informed culling decisions, but may also help producers weigh the costs of adopting new methods and technologies targeted at reducing on-farm lameness.     

Stocking density,milking duration,and lying times of lactating cows on Canadian freestall dairy farms

Lying time is an important measure of cow comfort, and the lying behavior of dairy cattle can now be recorded automatically with the use of accelerometers. To assess the effect that stall stocking density and the time that cows spend away from the home pen being milked has on the lying behavior of Holstein cattle, a total of 111 commercial freestall dairy farms were visited in Canada. Accelerometers were used to automatically record the lying behavior of 40 focal cows per farm. Total duration of lying, lying bout frequency, and the mean duration of lying bouts were calculated. Pen population was the total number of cows in the pen. To calculate stall stocking density (%) the number of cows in the pen and the number of useable stalls were counted and multiplied by 100, and the length × width of the pen was divided by the number of cows in the pen to calculate area/cow (m²). Time away from the pen per day was recorded from when the first cow in each pen was taken out of the home pen for milking until the last cow returned to the home pen after milking, and this time was multiplied by daily milking frequency. The median value for lying duration at the farm level was 10.6 h/d, with 10.5 lying bouts/d, and a median lying bout duration of 1.2 h. Stall stocking density ranged from 52.2 to 160.0%, with very few farms (7%) stocking at greater than 120%. Although stall stocking density was not significantly correlated with lying behavior, the results showed that no farm with stocking density greater that 100% achieved an average herd lying duration of 12 h/d or higher, whereas 21.6% of farms with a stocking density of 100% or less did achieve the target lying time of ≥12 h/d, as recommended by the Canadian Code of Practice (χ² = 4.86, degrees of freedom = 1). Area/cow (m²) was not correlated with any aspect of lying behavior, but regardless of space per cow, pen population was correlated with daily frequency and duration of lying bouts. As the number of cows in the pen increased, lying daily bout frequency increased (correlation coefficient = 0.24) and lying bout duration decreased (correlation coefficient = −0.30). Lying behavior was affected by the time the cows were away from the pen being milked. When cows were away from the pen for longer than 3.7 h/d, no farm achieved the recommended herd median lying time of 12 h/d or longer. These results suggest that providing 1 stall for each cow in the pen and minimizing time away from the pen are important factors if cattle are to achieve the recommended daily lying duration of 12 h/d.     

Comparison of models to identify lame cows based on gait and lesion scores, and limb movement variables

Bovine lameness results in pain and suffering in cattle and economic loss for producers. A system for automatically detecting lame cows was developed recently that measures vertical force components attributable to individual limbs. These measurements can be used to calculate a number of limb movement variables. The objective of this investigation was to explore whether gait scores, lesion scores, or combined gait and lesion scores were more effectively captured by a set of 5 limb movement variables. A set of 700 hind limb examinations was used to create gait-based, lesion-based, and combined (gait- and lesion-based) models. Logistic regression models were constructed using 1, 2, or 3 d of measurements. Resulting models were tested on cows not used in modeling. The accuracy of lesion-score models was superior to that of gait-score models; lesion-based models generated greater values of areas under the receiving operating characteristic curves (range 0.75 to 0.84) and lower mean-squared errors (0.13 to 0.16) compared with corresponding values for the gait-based models (0.63 to 0.73 and 0.26 to 0.31 for receiving operating characteristic and mean-squared errors, respectively). These results indicate that further model development and investigation could generate automated and objective methods of lameness detection in dairy cattle.     

Tie-stall design and its relationship to lameness, injury, and cleanliness on 317 Ontario dairy farms

The objective of this study was to identify relationships between tie-stall design and selected cow-based injury, lameness, and cleanliness measurements. All lactating dairy cows (n = 17,893) from 317 Ontario tie-stall dairy farms were evaluated once between March and September 2003. Stall dimensions were recorded and cows were scored for neck and hock lesions, broken tails, back arch, hind claw rotation, and udder and hind limb cleanliness. Neck lesions were significantly associated with tie-rail height. Hock lesions were positively associated with presence of an electric trainer and hind limb and udder cleanliness and negatively associated with tie-chain length. A negative association was found between broken tails and tie-rail height and a positive association between broken tails and udder and hind limb cleanliness. As mean stall length increased, fewer cows had hind-claw rotation. Having more dirty cows was associated negatively with stall length and chain length, and positively associated with the presence of an electric trainer. Proportion of cows with clean udders increased with the percentage of cows with clean hind limbs and with tie-rail height. As the prevalence of clean udders increased the prevalence of broken tails decreased. This study provides insight into how tie-stall dimensions may influence aspects of dairy cow welfare. Application of these principles could improve tie-stall design.     

Short communication: Milking order consistency of dairy cows in large Australian herds

We used on-farm records from dairy infrastructure to examine the consistency of the milking order over 150 d in 5 Australian dairy herds that were milking more than 500 cows as a single group. Within a single day the difference in milking order rank position was less than 20 percentage points for 72% of cows. The correlation coefficient comparing milking rank position in the morning and afternoon was 0.72, with the position of cows at the beginning and end of the milking order being more consistent than cows toward the middle of the milking order. Over a period of 150 d, cows with a mean position in the first and last 20% of the milking order maintained their position more consistently than cows in the middle of the milking order. Milking position of cows between one month and the next was highly correlated (r = 0.88). In large herds, subpopulations of cows are regularly milked toward the beginning and the end of the milking order. It is common for cows to be collected from the paddock as a group, to wait as a group in the dairy yard to be milked, and to return individually to the paddock or feed pad immediately after they have been milked. Thus, cows milked later in the milking order are likely to be away from the paddock for several hours longer than cows milked earlier in the milking order. This may affect their welfare though differences in time available for lying down, equality of pasture eaten, and time spent standing in the dairy yard.     

Perceptions and risk factors for lameness on organic and small conventional dairy farms

Lameness is an important multifactorial disease that affects dairy cattle on both organic (ORG) and conventional (CON) farms. The objective of this study was to characterize perception of lameness and identify risk factors for lameness prevalence on ORG and similarly-sized CON farms. Dairy herds (n = 292) were enrolled across 3 states (NY, OR, WI), with CON herds enrolled based on similar herd size and location of ORG herds. During a single herd visit, information was collected about management practices and lameness events occurring in the previous 60 d, and paperwork was left to record lameness events during the 60 d after the visit. During the herd visit, study personnel scored cows for body condition, lameness, and hock condition. For analysis, CON herds were further divided into CON grazing and CON nongrazing. A Poisson regression model was used to assess risk factors for prevalence of cows scored lame. On these relatively small, lower producing farms, the prevalence of lameness in cows scored by study personnel was less than previously reported for larger, higher producing dairy herds located in the United States. Prevalence of lameness was weakly and positively correlated with the rate of lameness calculated using farmer records. Researchers observed lame cows on some farms where farmers perceived that lameness never occurred. An increased prevalence of cows scored lame by study personnel was associated with an increased prevalence of hock lesions, use of CON nongrazing management, and routine use of a footbath. Multiple strategies may be used to manage lameness on farms, including increasing farmer perception of lameness and reduction in exposure of cows to risk factors that contribute to development of lameness.     

Applying additive logistic regression to data derived from sensors monitoring behavioral and physiological characteristics of dairy cows to detect lameness

The hypothesis was that sensors currently available on farm that monitor behavioral and physiological characteristics have potential for the detection of lameness in dairy cows. This was tested by applying additive logistic regression to variables derived from sensor data. Data were collected between November 2010 and June 2012 on 5 commercial pasture-based dairy farms. Sensor data from weigh scales (liveweight), pedometers (activity), and milk meters (milking order, unadjusted and adjusted milk yield in the first 2 min of milking, total milk yield, and milking duration) were collected at every milking from 4,904 cows. Lameness events were recorded by farmers who were trained in detecting lameness before the study commenced. A total of 318 lameness events affecting 292 cows were available for statistical analyses. For each lameness event, the lame cow’s sensor data for a time period of 14 d before observation date were randomly matched by farm and date to 10 healthy cows (i.e., cows that were not lame and had no other health event recorded for the matched time period). Sensor data relating to the 14-d time periods were used for developing univariable (using one source of sensor data) and multivariable (using multiple sources of sensor data) models. Model development involved the use of additive logistic regression by applying the LogitBoost algorithm with a regression tree as base learner. The model’s output was a probability estimate for lameness, given the sensor data collected during the 14-d time period. Models were validated using leave-one-farm-out cross-validation and, as a result of this validation, each cow in the data set (318 lame and 3,180 nonlame cows) received a probability estimate for lameness. Based on the area under the curve (AUC), results indicated that univariable models had low predictive potential, with the highest AUC values found for liveweight (AUC = 0.66), activity (AUC = 0.60), and milking order (AUC = 0.65). Combining these 3 sensors improved AUC to 0.74. Detection performance of this combined model varied between farms but it consistently and significantly outperformed univariable models across farms at a fixed specificity of 80%. Still, detection performance was not high enough to be implemented in practice on large, pasture-based dairy farms. Future research may improve performance by developing variables based on sensor data of liveweight, activity, and milking order, but that better describe changes in sensor data patterns when cows go lame.     

Validation of the M-stage scoring system for digital dermatitis on dairy cows in the milking parlor

A high prevalence of digital dermatitis (DD) and the benefits of early topical treatment highlight the need for simple tools for routine DD detection. The objective of this study was to determine the accuracy of scoring DD lesions using the 5 M-stage scoring system in the milking parlor compared with the trimming chute as the gold standard. Three observers inspected 3,585 cows and 6,991 hind feet from 9 farms in the milking parlor using a mirror (glued to a plastic kitchen spatula) and a headlamp, followed by inspection in a trimming chute within 5 d. Interobserver agreement for scoring DD in various settings was ≥82% (kappa >0.74; weighted kappa >0.76). At trimming chute inspections, 68% of cows had at least 1 DD lesion, 19% had 1 hind leg affected, and 49% had both hind legs affected. Within-herd DD prevalence ranged from 16 to 81% of cows affected. True within-herd prevalence was 2, 6, 0, 36, and 14% for M1, M2, M3, M4, and M4.1 lesions, respectively. At the foot level, DD prevalence was the same (58%) in the milking parlor and trimming chute inspection, but distribution of M-stages differed. Milking parlor inspection as a means of identifying the presence of DD lesions had a sensitivity of 92% and specificity of 88%, with positive and negative predictive values of 91 and 89%, respectively. Agreement between milking parlor and trimming chute inspections was 73% (kappa = 0.59, weighted kappa = 0.65) for the 5 M-stage scoring system and 90% (kappa = 0.80) if only the presence of a lesion was noted. Test characteristics varied greatly among M-stages, with the highest sensitivity for detecting M4 (82%) and M2 (62%) lesions, and the lowest for detecting M4.1 (20%), M1 (7%), and M3 (0%) lesions. In the milking parlor, 20% of M2 lesions were misclassified as M4.1, 8% of M4 lesions were misclassified as M0, and 68% of M4.1 lesions were misclassified as M4. The majority (87%) of DD lesions were located between the heel bulbs; 10 and 2% of DD lesions affected the interdigital space and the front of the foot, respectively. The sensitivity to detect the presence of a lesion when it occurred between the heel bulbs was 93%, but <67% if it occurred elsewhere on the foot. We concluded that inspection of the rear feet in the milking parlor was an inexpensive and simple method of detecting and scoring DD lesions. If the objective is to determine herd-level DD prevalence and routine monitoring, this method was adequately reliable. However, if the objective is to follow up DD in cows with history of interdigital hyperplasia or to detect M1 or M4.1 lesions, this method was not sufficiently reliable. Although DD scoring in the milking parlor as a routine practice should facilitate early detection, prompt treatment interventions, and herd monitoring, it was not sufficiently reliable to replace definitive identification of M-stages in the trimming chute.     

Randomized controlled trial comparison of analgesic drugs for control of pain associated with induced lameness in lactating dairy cattle

Both the economic loss and welfare implications of lameness affect the dairy industry. Currently no analgesic drugs are approved to alleviate lameness-associated pain in lactating dairy cattle in the United States. In this randomized controlled trial, 48 lactating Holsteins were enrolled to evaluate the effect of oral meloxicam and i.v. flunixin meglumine on induced lameness. Cows were allocated to 1 of 4 treatment groups (n = 12 per group): lameness and flunixin meglumine (LAME + FLU); lameness and meloxicam (LAME + MEL); lameness and placebo (LAME + PLBO); or sham induction and placebo (SHAM + PLBO). Six hours before treatment, arthritis-synovitis was induced in the distal interphalangeal joint with 20 mg of amphotericin B, whereas SHAM cows were given an intra-articular injection of an equal volume (4 mL) of isotonic saline. Cows in LAME + FLU received 2.2 mg/kg flunixin meglumine i.v. and whey protein placebo orally; LAME + MEL were administered 1 mg/kg meloxicam orally and 2 mL/45 kg sterile saline placebo i.v.; LAME + PLBO were administered 2 mL/45 kg sterile saline placebo i.v. and whey protein placebo orally; and SHAM + PLBO received 2 mL/45 kg sterile saline placebo i.v. and whey protein placebo orally. The initial treatment of MEL, FLU, or PLBO was identified as time 0 h and followed by a second dose 24 h later with data collection for 120 h. The methods used to assess analgesic efficacy were electronic pressure mat, visual lameness assessment, visual analog score, plasma cortisol concentration, plasma substance P concentration, mechanical nociception threshold, and infrared thermography imaging. Linear mixed effect modeling was the primary method of statistical analysis. Visual lameness scoring indicated a lower proportion of the FLU + LAME group was lame at the T2 h and T8 h time points in comparison to the positive controls, whereas MEL therapy resulted in a lower proportion of lame cows at the T8 h time point. Cortisol area under the effect curve was lower following FLU therapy compared with LAME + PBLO for the 0–2 h (LSM difference = 35.1 ng·h/mL, 95% CI: 6.8, 63.3 ng·h/mL), 2–8 h (LSM difference = 120.6 ng·h/mL, 95% CI: 77.2, 164.0 ng·h/mL), and 0–24 h (LSM difference = 226.0 ng·h/mL, 95% CI: 103.3, 348.8 ng·h/mL) time intervals. Following MEL therapy, cortisol area under the effect curve was lower than LAME + PLBO for both the 2 to 8 h (LSM difference = 93.6 ng·h/mL, 95% CI: 50.2, 137.0 ng·h/mL) and 0 to 24 h time intervals (LSM difference = 187.6 ng·h/mL, 95% CI: 64.9, 310.4 ng·h/mL). Analysis of data from other assessment modalities failed to discern biologically relevant differences between treatment groups. We conclude that meaningful differences were evident for visual lameness assessment and cortisol from MEL and FLU treatment versus the positive control. Further clinical research is needed toward development of a model that will create reproducible events that are more pronounced in severity and duration of lameness which can be validated as a substitute for naturally occurring lameness cases.