If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
UBC Dairy Education and Research Centre, University of British Columbia, Agassiz, British Columbia, Canada, V0M 1A0Department of Animal Environment and Health, Swedish University of Agricultural Sciences, 532 23 Skara, Sweden
Lameness is a major concern to animal health and welfare within the dairy industry. Our objectives were to describe the prevalence of lameness in high-producing cows on farms with automated milking systems (AMS) and to identify the main risk factors for lameness at the animal and farm level. We visited 36 AMS farms across Canada and Michigan. Farm-level factors related to stall design, bedding use, flooring, and stocking rates were recorded by trained observers. Cows were scored for lameness, leg injuries, body condition (BCS), and body size (hip width and rump height; n = 1,378; 25–40 cows/farm). Mean herd prevalence of clinical lameness was 15% (range = 2.5–46%). Stall width relative to cow size and parity was found to be the most important factor associated with lameness. Not fitting the average stall width increased the odds of being lame 3.7 times in primiparous cows. A narrow feed alley [<430 cm; odds ratio (OR) = 1.9], obstructed lunge space (OR = 1.7), a low BCS (OR = 2.1 for BCS ≤2.25 compared with BCS 2.75–3.0), and presence of hock lesions (OR = 1.6) were also identified as important risk factors for lameness. Only 1 of 36 farms had stalls of adequate width and length for the cows on their farm. For lameness prevention, it can be concluded that more emphasis needs be placed on either building stalls of appropriate width or selecting for smaller-framed cows that fit the existing stalls.
Lameness remains one of the most important concerns to animal health and welfare within the dairy industry. Lameness is considered to be painful and is associated with reduced milk yields, negative effects on reproductive performance, and higher culling rates, resulting in substantial production losses (
). One key to successful milking on farms with automatic milking systems (AMS) is voluntary attendance at the milking robot. Prevention of lameness is therefore important in herds with AMS also from an economic point of view, as lame cows voluntarily approach the AMS less frequently (
An important risk factor for lameness in freestall systems with parlor milking is a lack of cow comfort, resulting in increased time spent standing by the cows, which is especially problematic if flooring is hard (
). This lack of cow comfort could be due to inadequate stall design or bedding management. These risk factors are likely to be present also in AMS herds. However, some potential lameness risk factors are specific to AMS herds, such as overstocking of cows at the AMS or management of cow traffic, as these also affect cows’ time budget (
). Studies reporting on prevalence of lameness and risk factors for lameness in AMS herds are, however, limited. The objective of the current study was to describe the prevalence of lameness in high-producing cows on AMS farms within Canada and Michigan, and to identify potential risk factors for lameness related to animal-based measures and to facility design and management practices on AMS farms.
Materials and Methods
This study was part of a larger study aimed at evaluating the comfort of cows on AMS farms across Canada and Michigan (
). The study was approved by the Institutional Animal Care Committees and Research Ethics Boards at Université Laval, the University of Guelph, the University of Calgary, and Michigan State University.
Herd Selection
Between June 2010 and November 2012, we visited 36 farms with AMS in Quebec (QC; n = 10), Ontario (ON; n = 10), British Columbia (BC; n = 4), and Alberta (AB; n = 5), Canada, as well as Michigan (MI; n = 7). Farms had to have operated the AMS for at least 6 mo. Farms were invited by mail to participate in the study. When letters were returned indicating the willingness of the producer to participate, they were interviewed by telephone to determine if they met the additional study inclusion criteria, which included having cows housed in freestalls in their present barn for at least 1 yr and no access to outdoor exercise area or pasture. The mean (±SD) number of lactating cows in the participating farms was 125 ± 108 (maximum 495 cows) and the mean annual milk production was 9,346 ± 773 kg [retrieved from Valacta (Sainte-Anne-de-Bellevue, QC, Canada) and CanWest DHI Herd Recording (Guelph, ON, Canada) data, available for 17 farms]. The number of AMS ranged from 1 to 8 per farm, with the majority of farms having 1 (47%) or 2 (30%) AMS units. Twenty-eight farms had a free cow traffic system to access the AMS, whereas 8 farms had a directed traffic system.
Cow Selection
In each herd, we selected 40 lactating Holstein focal cows for detailed cow-based measurements. Where possible, we maximized the number of focal cows that were ≤120 DIM. If the milking herd had less than 40 cows ≤120 DIM, the selection criterion was expanded above 120 DIM until a sample of 40 cows was obtained. The sample of focal cows was also chosen to reflect the ratio of primiparous to multiparous cows in the herd. In 5 herds, fewer than 40 lactating cows were available; thus, fewer cows were selected as focal animals (range = 25–39 cows for these 5 herds). Data were successfully obtained from 1,378 cows in total.
Animal-Based Measures
All focal cows were assessed for lameness, for the presence of lesions on the hock (tarsus joint) and knees (carpal joints), leg cleanliness, BCS, hip width, and rump height. In addition, the parity and DIM of each focal cow were retrieved from farm records. Standard operating procedures were used to collect each animal based measure as described by
), and these are available on the Canadian Dairy Research Portal (https://www.dairyresearch.ca/animal-comfort-tool.php). Measures were taken while cows were locked in self-locking head gates at the feed alley. Cows were scored for hock and knee injuries according to the criteria described by
. Both the left and right limbs were scored for hock and knee injuries on a 4-point scale (Table 1). Cleanliness on the lateral lower hind right leg (from the top of the claw to the middle of the hock) was assessed using a 4-point scoring system adapted from
. Scores were based on the level of manure contamination where 0 = fresh manure for <50% of the area; 1 = fresh manure for >50% of the area; 2 = dried caked and fresh manure for <50% of the area; and 3 = entire area with dried caked manure. Body condition scores were recorded on a 5-point scale in 0.25 increments using the Elanco Animal Health body condition scoring chart for dairy cattle [
]. Claw length was assessed by estimating the angle of the dorsal surface of the left and right lateral claws in relation to the ground. We defined claw length as no overgrowth (angle ≥45°) or overgrown (angle <45°). Hip width was measured between the points of the 2 hook bones with a flexible measuring tape when the cow was in a standing position. Rump height was measured from the ground to the spine parallel to the hook bone using a height stick. To identify lame cows, all cows were individually video recorded while walking in the feeding alley after they were released, one at a time, from the head locks. Cows were later assessed using the video to determine whether they had a head bob or had an obvious limp, defined as uneven weight bearing of one or more limbs, adapted from
Data were collected during a single visit by 2 assessors. Each assessor took the same measures on all cows. In total, 11 assessors collected data on the different farms. To ensure high repeatability, all assessors had taken part in the same training program before the study started, which is described in detail by
. Briefly, the assessors underwent an intensive training program and the repeatability for each assessor was assessed against 2 standard trainers, with only assessors that reached the target level of a weighted kappa (Kw) ≥0.6 being used (
). Once field data collection was in progress, the trainees were reassessed twice (via a refresher course 3 to 4 wk after initial training and a midway assessment) to ensure that they remained objective and repeatable in their scoring.
Farm-Level Measurements
Pen Design and Stocking Density
Farm-level measures were taken in all pens containing focal cows. In 34 farms, all focal cows were located in a single pen. On the remaining 2 farms, focal cows were spread across 2 and 3 pens. In each pen, overall pen area (m2/cow) was defined as the total length by the width of the pen divided by the number of cows in the pen. Stall stocking density was calculated as the number of cows per useable lying stall. The length of the feed bunk was measured, and feeding space per cow was calculated by dividing the feed bunk space by the number of cows in the pen. The type of feed barrier and flooring type in the cow alley adjacent to the feed bunk were recorded. Cleanliness of the feed alley floor was assessed by walking the entire length of the feed bunk alley 20 min before and after scraping and measuring the height of manure that collected on the heel of a clean rubber boot. Floor cleanliness was scored as clean (<1 cm of manure) or dirty (≥1 cm of manure). If the scraping system was manual, feed alley cleanliness was scored at the beginning and at the end of the visit. The width of the alley where cows stood to feed was measured as the distance from the feed bunk to the rear curb or the front of the lying stalls.
Stall Design and Bedding Maintenance
Data on 8 dimensions per stall, including stall length, bed length (distance from rear curb to brisket board), brisket board height, neck rail height, curb height, distance of neck rail from rear curb, height of upper edge of bottom divide rail, and lunge space, were estimated for an average of 6 ± 1 stalls per pen (range = 4 to 10 stalls). The measures were obtained from the first and the last usable stalls in 3 representative rows in each pen containing focal cows. If <3 rows were present, the middle stall in each row was also assessed. In the same stalls, bedding quantity, cleanliness, and dryness were also assessed. Stall width was measured as the average width of 3 consecutive stalls on either side of the middle stall of each row that was measured (range 9 to 24 stalls per pen).
To take these measures, we again used the standard operating procedures described by
, which are available on the Canadian Dairy Research Portal (www.dairyresearch.ca/animal-comfort-tool.php). In short, lunge space was assessed as adequate if no obstruction was present in the bob-zone, ≤76 cm forward from the top of the brisket board as proposed by
. If no brisket board was present, lunge space was measured horizontally below the neck rail and 76 cm forward, 10 cm above the stall surface. Bedding quantity was evaluated as none (unable to rake bedding), ≤2 cm (when raked flat for organic bedding, or below the curb for sand bedding), or >2 cm (when raked flat for organic bedding, or even with or above the curb for sand bedding). Bedding cleanliness was assessed on a 5-point scale, where 0 = none; 1 = little manure or visible wet areas; 2 = manure-free area larger than contaminated area; 3 = contaminated area larger than manure-free area; and 4 = entire area contaminated. Dryness was scored as 0 = dry, 1 = wet, and 2 = very wet. The type of bedding and stall base (surface under bedding) were also recorded. If different types of stall bases were used in the same pen, the number of stalls with each base type was counted.
General Management
During the visit, all producers were interviewed about their general management routines including, for example, stall and pen management and herd health issues such as lameness monitoring and hoof trimming practices. They were also asked how many times per day cows were fetched to be milked. The complete management questionnaire is available on the Canadian Dairy Research Portal (https://www.dairyresearch.ca/animal-comfort-tool.php).
Data Handling and Statistical Analysis
Data Handling
Data were entered into a relational database (Access 2010; Microsoft Corp., Redmond, WA), and then exported to Excel files (Excel 2010; Microsoft Corp.) and into Stata/IC 12 (StataCorp LP, College Station, TX) for statistical analyses. Due to missing data, the number of cows on each farm varied from measure to measure (Table 2). Data on parity and knee lesions were only obtained from 34 and 33 herds, respectively. If a categorical predictor had too few observations in any category, categories were combined based on biological cut-points or to equalize the number of observations across categories. All farm-level categorical predictors considered for analyses had at least 5 pens or farms per category (Table 3). Days in milk were categorized in quartiles. Parity was categorized as 1 and ≥2. Feed alley width was dichotomized as <430 and ≥430 cm, as a width of ≥430 cm is recommend in the Canadian Dairy Code of Practice (
Dairy Farmers of Canada and the National Farm Animald Care Council (DFC-NFACC). 2009. Code of practices for the care and handling of dairy cattle. Dairy Farmers of Canada, Ottawa, Ontarioa, Canada. Accessed May 10, 2015. http://www.nfacc.ca/codes-of-practice
). In alignment with these recommendations, feeding space was dichotomized as <60 and ≥60 cm. A cow was classified to not fit the stall width if the average stall width was narrower than 2× the cow’s hip width, as proposed by
. Cows were categorized as lame if they had an obvious limp, and severe lameness was categorized as having both an obvious limp and a head bob. Cows were classified as having a hock or a knee lesion if they had a score of ≥2 on at least one leg.
Table 2Distribution of all cow-level explanatory variables hypothesized to be associated with lameness as measured on 1,378 cows from 36 dairy farms with automated milking systems
Table 3Distribution of all pen and farm-level explanatory variables hypothesized to be associated with lameness as measured on 1,378 cows in 39 pens from 36 dairy farms with automated milking systems
Lunge space was classified as adequate at the pen level if no obstruction was present in ≥75% of the measured stalls. Similarly, bedding quantity, cleanliness, and dryness was classified as ≥2 cm, clean (score 0–1), and dry (score 0) if ≥75% of the measured stalls in the pen received the corresponding scores.
Statistical Analysis
Model building was performed in 2 steps. The outcome of interest was the presence of lameness (i.e., having an obvious limp) at the cow level. First, simple logistic regression analysis was performed to assess the association between lameness and each predictor variable, applying the LOGIT procedure in Stata (StataCorp LP). Hypothesized and tested variables are listed in Tables 2 and 3. Predictors with an association with lameness at P ≤ 0.25 were considered for further modeling (listed in Table 4).
Table 4Univariable associations expressed as odds ratios of the presence of lameness with individual and herd-level factors in 1,378 cows on 36 dairy farms with automated milking system
Next, we built 2 separate multivariable mixed-effects logistic regression models with the selected predictor variables (Table 4). One model considered cow-level measures and the other examined management and explanatory variables at the farm and pen level. As the focal cows were nested within pen and within farm, the random effects of pen within farm were forced into the models using the XTMELOGIT procedure in Stata (StataCorp LP). Both models were constructed using manual backward stepwise elimination, retaining variables with P ≤ 0.05. Continuous variables were centered on the overall mean and tested with their quadratic terms. The variable stall base was tested using only 2 categories (sand vs. all others) as no difference was noted between the non-sand categories in the univariable analyses. In addition, the fixed effect of geographic location (AB and BC, ON, QC, and MI) was tested in both models but not retained (P > 0.05). Finally, previously excluded variables were forced into the model again and retained if P ≤ 0.05 or if they were judged as confounders (changed the estimate of any other variable by >30%). Biologically plausible 2-way interactions between the variables retained in the models were tested. An interaction between the variables parity and fit stall width was found to be significant (P ≤ 0.05) and was therefore kept in the final model concerning cow-level measures. Coefficient estimates were transformed and are presented as odds ratios (OR). The final models were validated by examination of influential covariate patterns and by running the model with and without these. We found no reason to exclude any observations, and the model fits were considered satisfactory.
Results
Lameness Prevalence
The mean herd prevalence of lameness was 15%, but ranged from 2.5 to 46% among the participating farms. Severe lameness was less common, averaging 4%; however, in 5 herds, 10 to 12% of the focal cows were severely lame (i.e., obviously limping and having a head bob). All producers with >20% lameness in their herd (8 farms) stated that they had a major or moderate problem with lameness when answering the questionnaire. However, 7 out of 8 producers with a herd prevalence between 15 and 20% answered that they had a minor problem with lameness. Of the 36 participating farms, 10 kept records of all identified cases of lameness, whereas on 13 farms treated cases only were recorded. Four farms used hoof trimmer records as the single documentation of lameness. The remaining 9 farms did not keep any record of lameness at all. Almost all farms trimmed their cows routinely (≥1 time per year). Only 2 farmers stated that they called the trimmer out or trimmed themselves only when a cow needed it. Producers stated that they fetched cows to the AMS unit on average 3 times/d (range = 0–12). A positive correlation was noted between the herd prevalence of lameness and the number of times per day producers stated that cows were fetched to be milked (r = 0.45, P = 0.007).
Cow-Level Factors
Results of single variable analyses (P ≤ 0.25) are presented in Table 4 and results of the multivariable analysis in Table 5. Body condition score, presence of hock lesion, and the interaction between parity and not fitting the average stall width were associated with lameness in the final model concerning cow-level factors (Table 5). A BCS of ≤2.25 was associated with twice the odds of being lame (OR = 2.11, P = 0.005) compared with a BCS of 2.75 to 3.0. In total, 13% of the cows had a BCS ≤2.25 but the proportion ranged from 0 to 40% across the herds. Farms situated in QC had, on average, 29% of their focal cows within BCS category ≤2.25, compared with 10% in ON and MI and 3% in AB and BC. Only 19 cows (1%) had a BCS of 2.0. The proportion of lame cows within different BCS categories is presented in Figure 1.
Table 5Results of a multivariable mixed-effects logistic regression model
Figure 1Proportion of lame cows within different BCS categories (n = 1,375 cows). A BCS of ≤2.25 was associated with twice the odds of being lame (odds ratio = 2.1, P = 0.005) compared with a BCS of 2.75 to 3.0.
Cows had a mean (±SD) hip width of 62 ± 7 cm (n = 1,311) and a mean rump height of 148 ± 5 cm (n = 1,313). The width of the stalls was 120 ± 4 cm, on average, and mean bed length was 178 ± 10 cm (Table 3). As a result, 62% of the cows were too big to fit the average stall width (2× the cow’s hip width needed) and 42% did not fit the average bed length (1.2× the cow’s rump height needed). Only 30% of the multiparous cows fit the stall width. Due to a significant interaction with parity, not fitting the average stall width increased the odds of being lame 3.7 times in primiparous (P = 0.004) and 1.3 times in multiparous cows (P < 0.001; Table 5; Figure 2). We found no association between lameness and the cow not fitting the stall bed length (P = 0.83). Out of 36 farms, 10 farms succeeded in fitting ≥75% of the focal animals to stall width and 19 farms in fitting cows to bed length. Only 1 farm had stalls wide and long enough to fit all focal cows. It was cow size (hip width), rather than stall width, that varied across farms with a different proportion of cows fitting the stall width (Table 6).
Figure 2Proportion of lame cows with different parity presented in relation to whether the cow fits the average stall width or not (n = 1,215 cows). Not fitting the average stall width increased the odds of being lame by 3.7 times in primiparous (P = 0.004) and 1.3 times in multiparous cows (P < 0.001).
Of the housing and management predictors tested, a narrow feed alley (<430 cm; OR = 1.85; P = 0.006) and an obstructed lunge space (OR = 1.71, P = 0.02) were associated with higher odds of lameness in the final multivariable model (Table 7). The width of the feed alley averaged 406 cm (range = 301–640 cm across pens), and the proportion of lame cows within pens with different feed alley widths is presented in Figure 3. Twenty-seven farms had adequate lunge space in ≥75% their measured stalls. The mean prevalence of lameness in these farms was 13% compared with 19% in farms that had more stalls with obstructed lunge spaces.
Table 7Results of a multivariable mixed-effects logistic regression model
Figure 3Proportion of lame cows in 39 pens from 36 dairy farms with automatic milking systems, presented in relation to width of the feed alley. Dotted line shows minimum feed alley width recommended in the Canadian Dairy Code of Practice (
Dairy Farmers of Canada and the National Farm Animald Care Council (DFC-NFACC). 2009. Code of practices for the care and handling of dairy cattle. Dairy Farmers of Canada, Ottawa, Ontarioa, Canada. Accessed May 10, 2015. http://www.nfacc.ca/codes-of-practice
Stall base was associated with lameness in the screening of single variables (P = 0.04), and a tendency was noted for sand bedding to be associated with lower odds of lameness in the final multivariable model (P = 0.06; Table 7). More pen space per cow was associated with decreased odds, and increasing stall stocking density was associated with higher odds for lameness in the simple analysis only (Table 4). On 10 farms, stocking density was higher than 1 cow/lying stall, which is the recommended best practice according to the Canadian Dairy Code of Practice (
Dairy Farmers of Canada and the National Farm Animald Care Council (DFC-NFACC). 2009. Code of practices for the care and handling of dairy cattle. Dairy Farmers of Canada, Ottawa, Ontarioa, Canada. Accessed May 10, 2015. http://www.nfacc.ca/codes-of-practice
). Only 1 farm exceeded the Canadian Dairy Code requirement of a maximum stocking density of 1.2 cows/stall.
Neither the AMS traffic system used nor the number of cows per AMS unit was related to lameness. Half of the farms provided the recommended 60 cm of feeding space for each cow, but higher feed bunk density was not associated with lameness.
Discussion
Prevalence of Lameness in AMS Farms
The mean prevalence of lameness in our study was 15%. This is similar to the 14% reported for AMS farms in Denmark (
). The latter 2 studies followed the same data collection protocol as described in the current paper. Our results, thus, indicate that prevalence of lameness seems to be lower in AMS herds in comparison to Canadian tiestall and freestall herds without AMS. This may be due to factors associated with the AMS, such as desynchronized milking times, or it may reflect the management styles of producers who adopt AMS, including a greater awareness of the importance of controlling lameness. On the other hand, visited AMS farms were few and it is possible that these herds had a low prevalence of lameness before transition to AMS. Also, cows on visited AMS farms were often housed in newer facilities compared with the majority of participating freestall farms without AMS. In other studies, differences in locomotion scores associated with converting to AMS have not been found when holding management and facility design constant (
in: Meijering A. Hogeveen H. Koning C.J.A.M.d. Automatic Milking-A Better Understanding. Wageningen Academic Publishers,
Wageningen, the Netherlands2004: 125-134
The influence of an automatic milking system on claw health and lameness of dairy cows.
in: Meijering A. Hogeveen H. Koning C.J.A.M.d. Automatic Milking—A Better Understanding. Wageningen Academic Publishers,
Wageningen, the Netherlands2004: 155-160
In the present study we found a positive correlation between the herd prevalence of lameness and the number of times the farmer stated that cows had to be fetched for milking. The number of times cows were fetched was only stated by the farmers, but not measured directly. However, we have no reason to believe that farmers did not answer this question truthfully with estimates close to the actual number; therefore, we consider these data valid for concluding there is a correlation between lameness and fetching. Fetching cows is very time consuming, and the increased workload possibly makes herd managers more motivated to reduce lameness in AMS compared with non-AMS farms. Some farms in the current study were indeed very successful at controlling lameness, having ≤5% lame cows. This demonstrates that a lameness prevalence of 5% or less is an achievable goal.
Cow-Level Factors Associated with Lameness
Not fitting the average stall width was the strongest animal-based risk factor for lameness, increasing the odds of lameness 3.7 times in primiparous cows. Difficulties in performing lying down movements may make the cow more hesitant to lie down, possibly resulting in greater standing time. It is also possible that rest is disrupted by neighboring cows more often if stalls are too narrow. A 3-dimensional biomechanical study by
revealed that the lateral displacement of a cow during lying down movements reaches up to 180% of the hip width. A stall width that is 200% of the hip width is therefore proposed as adequate to accommodate most lying movements (
found that cows reduced their lying time by 1.2 h/d in narrow stalls (112 vs. 132 cm wide). Reduced lying time acts as an exacerbating factor in the development of claw lesions (
), and poor cow comfort resulting in reduced lying time is believed to have a critical role in both the development of lameness and the speed of recovery (
Stalls in our study were, on average, 120 cm wide and the majority of both primiparous and multiparous cows were too large to fit these properly. In the Canadian Dairy Code of Practice, 117 cm is the recommended minimum stall width for cows weighing 545 kg, 122 cm for cows weighing 636 kg, and 127 cm for 727 kg cows (
Dairy Farmers of Canada and the National Farm Animald Care Council (DFC-NFACC). 2009. Code of practices for the care and handling of dairy cattle. Dairy Farmers of Canada, Ottawa, Ontarioa, Canada. Accessed May 10, 2015. http://www.nfacc.ca/codes-of-practice
). The fact that cows too large for their stalls in our study were at greater risk of being lame suggests that the guidelines within the Canadian Dairy Code of Practice (
Dairy Farmers of Canada and the National Farm Animald Care Council (DFC-NFACC). 2009. Code of practices for the care and handling of dairy cattle. Dairy Farmers of Canada, Ottawa, Ontarioa, Canada. Accessed May 10, 2015. http://www.nfacc.ca/codes-of-practice
) are reasonably accurate. Experimental studies are needed to explore the association between lameness and stall width in detail. As most farms failed to provide stalls of adequate width to accommodate the size of their cows, correct body size does not appear to be taken into account when building stalls. This could be addressed by constructing wider stalls or, alternatively, by selecting for smaller-framed cows that would fit into the pre-existing stalls.
Lower BCS (≤2.25) were associated with increased odds for lameness. A correlation between low BCS and lameness has been demonstrated in numerous studies (reviewed by
). Historically, it was assumed that lame cows lost weight as a consequence of reduced feed intake. A growing number of studies, however, suggest that a low BCS may predispose cattle to lameness (
Lameness in dairy cattle: A debilitating disease or a disease of debilitated cattle? A cross-sectional study of lameness prevalence and thickness of the digital cushion.
J. Dairy Sci.2009; 92 (http://dx.doi.org/10.3168/jds.2008-1827): 3175-3184
). Cows in AMS herds also have to compete for access to the milking robot and low-ranking cows in general spend more time queuing in the waiting area (
). Simulations of waiting time show that low-ranking cows may wait to milk about 1 h on a typical day, compared with 10 min for a middle-rank animal and 3.5 min for the dominant ones (
). The daily lying time of the cows participating in this study was up to 1 h shorter in individuals with BCS ≤2.25 compared with cows with BCS ≥3.25 (
Dairy Farmers of Canada and the National Farm Animald Care Council (DFC-NFACC). 2009. Code of practices for the care and handling of dairy cattle. Dairy Farmers of Canada, Ottawa, Ontarioa, Canada. Accessed May 10, 2015. http://www.nfacc.ca/codes-of-practice
). Only 1% of examined cows fell within this category; thus, our results show that at a BCS of 2.25, animals are at higher odds of being lame. To take corrective action sooner is therefore warranted.
An association between lameness and presence of hock lesions has been found in several studies of freestall systems with milking parlors (
). As risk factors and the outcome of interest were measured at the same time in our study, we cannot draw conclusions about causality. The presence of hock lesions is, however, strongly related to time spent lying on abrasive surfaces and prolonged high local pressure or friction of the hock on hard surfaces, and it is suggested that lameness aggravates the risk of hock lesions rather than the other way around (
Housing and Management Factors Associated with Lameness
Of the other housing and management factors tested, the width of the feed alley, lunge space, and stall base (sand vs. all others) were the most important in relation to probability of being lame. To allow easy movement of cattle in the feeding area, a feed alley width of 430 cm is recommended in the Dairy Code of Practice (
Dairy Farmers of Canada and the National Farm Animald Care Council (DFC-NFACC). 2009. Code of practices for the care and handling of dairy cattle. Dairy Farmers of Canada, Ottawa, Ontarioa, Canada. Accessed May 10, 2015. http://www.nfacc.ca/codes-of-practice
). Farms fulfilling this criterion in our study had fewer lame cows, suggesting that this recommendation is relevant. If walking alleys are too narrow, cows cannot move around without interference from other cows. Reducing the general space per cow in cubicle systems has been shown to increase the number of agonistic interactions among cows (
, agonistic interactions possibly induce claw trauma as a result of rapidly moving to avoid or give way to dominant cows. In alignment with our findings,
also found an association between a narrow feed alley and an increased prevalence of lameness (<320 cm compared with >340 cm). Other recent, large studies of herd-level risk factors for lameness seem to have overlooked the width of the feed alley (
). We therefore, propose that width of the feed alley be included as a potential risk factor for lameness in future studies.
Stall designs that fail to provide for the movements associated with lying and rising and the need of a soft, cushioned surface reduce the use of the stall and thus increase the risk of lameness (
). Therefore, it is no surprise that in our study an obstructed lunge space was associated with higher odds for lameness and sand bedding in stalls was associated with lower odds for lameness. Deep-bedded sand, because of its ability to supply cushion and traction, allows cows to perform the processes of rising and lying down more easily, without fear of slipping (
). The majority of farms in our study had fewer than 60 cows/unit and the number of cows per AMS unit was not associated with lameness. The type of traffic management surrounding the AMS (free vs. directed traffic) was also not associated with lameness in our study. Only 8 of 36 farms had a directed traffic system in the current study; further, the design of the directed traffic systems was highly variable. More detailed studies are therefore needed to fully understand the effect of traffic system on lameness in AMS herds.
Stall stocking density was only positively correlated with lameness in the univariable analysis. It has previously been demonstrated that overstocking reduces lying time in a linear fashion (
), which could explain a possible link to lameness. However, in the present study, all but one farm followed the Canadian guidelines of ≤1.2 cows/stall. Thus, data from overcrowded herds are needed to more fully evaluate the risk level for lameness due to overstocking.
Conclusions
The observed prevalence of lameness in our study population of AMS herds was lower than that observed previously in Canadian tiestall and freestall herds without AMS. The main risk factors for lameness in AMS farms involve stall design and, thus, appear to be similar to those on freestall farms with conventional milking parlors, rather than being related to management of the AMS. In particular, stall width relative to cow size was found to be the most important factor associated with lameness. Only 1 of 36 farms had stalls of both adequate width and length for the cows on their farm. Narrow feed alleys were also identified as a risk factor. The strong association between stall width and prevalence of lameness indicates that more emphasis needs be placed on either building stalls of appropriate width or selecting for smaller framed cows which fit the existing stalls. Experimental studies are needed to specifically explore the association between lameness and stall width.
Acknowledgments
We thank the participating farmers from the United States and Canada and the collaborators, technicians, students, and co-op students from Michigan State University, the University of British Columbia, University of Calgary, University of Guelph, Laval University, and Valacta Inc. (Sainte-Anne-de-Bellevue, QC, Canada). We thank Jenny Gibbons (AHDB Dairy, Warwickshire, UK) and Gemma Charlton (Harper Adams College, Newport, UK) for invaluable help with establishing the database. This study was funded by the Dairy Farmers of Canada (Ottawa, ON, Canada), the Canadian Dairy Commission (Ottawa, ON, Canada), Agriculture and Agri-Food Canada (Ottawa, ON, Canada), as part of the Dairy Science Cluster initiative, Agriculture and Agri-Food Canada and the BC Ministry of Agriculture (Victoria, BC, Canada) through the Canada-BC Agri-Innovation Program under Growing Forward 2, a federal-provincial-territorial initiative (delivered by the Investment Agriculture Foundation of BC), Alberta Milk (Edmonton, AB, Canada), the Alberta Livestock and Meat Agency (Edmonton, AB, Canada), and the Natural Sciences and Engineering Research Council of Canada (Ottawa, ON, Canada). We also thank the Michigan State University Department of Animal Science for supporting this research through an award from the Elwood Kirkpatrick Dairy Research Science Endowment to Janice Siegford (East Lansing, MI).
References
Anderson N.
Free Stall Dimensions. Infosheet - Livestock Technology. Ministry of Agriculture, Food and Rural Affairs,
Fergus, Ontario, Canada2007
Lameness in dairy cattle: A debilitating disease or a disease of debilitated cattle? A cross-sectional study of lameness prevalence and thickness of the digital cushion.
J. Dairy Sci.2009; 92 (http://dx.doi.org/10.3168/jds.2008-1827): 3175-3184
Dairy Farmers of Canada and the National Farm Animald Care Council (DFC-NFACC). 2009. Code of practices for the care and handling of dairy cattle. Dairy Farmers of Canada, Ottawa, Ontarioa, Canada. Accessed May 10, 2015. http://www.nfacc.ca/codes-of-practice
in: Meijering A. Hogeveen H. Koning C.J.A.M.d. Automatic Milking-A Better Understanding. Wageningen Academic Publishers,
Wageningen, the Netherlands2004: 125-134 ()
The influence of an automatic milking system on claw health and lameness of dairy cows.
in: Meijering A. Hogeveen H. Koning C.J.A.M.d. Automatic Milking—A Better Understanding. Wageningen Academic Publishers,
Wageningen, the Netherlands2004: 155-160
00178-8&id=gr1.jpg){kind=link}
00178-8&id=gr2.jpg){kind=link}
00178-8&id=gr3.jpg){kind=link}