Prevalence and interrelationships of hoof lesions and lameness in Swedish dairy cows

Preventive Veterinary Medicine 54 (2002) 247±263 Prevalence and interrelationships of hoof lesions and lameness in Swedish dairy cows Thomas Manske *, Jan Hultgren, Christer Bergsten Department of Animal Environment and Health, Section of Field Research, Swedish University of Agricultural Sciences, P.O. Box 234, SE-532 23 Skara, Sweden Received 12 September 2001; accepted 24 March 2002 Abstract The prevalence of hoof lesions and lameness in 4899 heifers and cows was determined at claw trimming one time in a cross-sectional study on 101 Swedish dairy farms, 1996±1998. The percentage of affected animals was 41% for heel-horn erosion, 30% for sole haemorrhages, 27% for erosive dermatitis, 21% for abnormal claw shape, 14% for white-line haemorrhages, 8.8% for white-line ssures, 8.6% for sole ulcers, 3.3% for double soles, 2.3% for verrucose dermatitis, and 1.8% for interdigital hyperplasia (IH). Seventy-two percent of all animals had at least one hoof lesion. The prevalence of lameness was 5.1%; most hoof lesions did not cause lameness. Differences between herds were substantial; the herd-speci c, animal-level prevalence of lesions ranged from 25 to 98% and of lameness from 0 to 33%. Sole haemorrhages were found in all herds. The proportion of variance at the between-herd level was particularly high for heel-horn erosion (62%) and abnormal claw shape (54%). Strong correlations between lesions were found within hooves (and animals), e.g. for heel-horn erosion and dermatitis (Spearman's rank correlation, r s ˆ 0:36 and 0.37, respectively) and for sole and white-line haemorrhages (r s ˆ 0:25 and 0.28). Most hoof lesions affected hind and front hooves bilaterally, whereas the correlation between hind and front hooves generally was lower. Herds that ranked high for prevalence of sole ulcer also ranked high for sole haemorrhages and for abnormal claw shape and herds that ranked high for dermatitis also ranked high for heel-horn erosion, verrucose dermatitis and IH. Abnormal claw shape was strongly associated with sole ulcer (r s ˆ 0:41 at cow level)ðsuggesting the importance of maintaining a correct claw shape for the prevention of hoof-horn lesions. # 2002 Elsevier Science B.V. All rights reserved. Keywords: Dairy cattle; Hoof lesion; Lameness; Claw trimming * Corresponding author. Tel.: 46-5116-7123; fax: 46-5116-7134. E-mail address: thomas.manske@hmh.slu.se (T. Manske). 0167-5877/02/$ ± see front matter # 2002 Elsevier Science B.V. All rights reserved. PII: S 0167-5877(02)00018-1

248 T. Manske et al. / Preventive Veterinary Medicine 54 (2002) 247±263 1. Introduction In an earlier Swedish study, approximately three out of four slaughtered dairy cows had at least one hoof lesion (Andersson and LundstroÈm, 1981). Hoof lesions cause most lamenesses in dairy cattle (Murray et al., 1996) and lameness is an important cause of culling (Collick et al., 1989; Sprecher et al., 1997). Therefore, studies restricted to abattoir material might overestimate the real prevalence of hoof lesions. Investigations of the hoofhealth status in live dairy cattle have, with some notable exceptions (e.g. Enevoldsen et al., 1991a,b; Frankena et al., 1992; Smits et al., 1992; Smilie et al., 1996, 1999; Vaarst et al., 1998), focused on lameness and on the hoof-health status in lame cows. However, hoof lesions do not always cause lameness (Smits et al., 1992; Logue et al., 1994), and inferences from studies on the prevalence of hoof lesions restricted to lame animals might therefore lead to an underestimation of the prevalence of lesions and an over-estimation of their effect. A convenient method for gathering data on hoof-health is to make use of protocols from maintenance claw trimming (Enevoldsen et al., 1991a,b; Vaarst et al., 1998). Such records are not only important in the investigation of herd-health problems (Mills et al., 1986; Bargai, 2000), but might also play an important role in the evaluation of breeding animals for hoof-health and soundness in locomotion (Ral et al., 2001). A 2-year project was designed to describe the hoof-health status of dairy cattle, using hoof-health records obtained at claw trimming. The prevalence of different hoof lesions in dairy cattle from herds in south-western Sweden (as recorded at claw trimming in the housing seasons 1996±1997 and 1997±1998) is presented here. We also present the associations between different lesions within hooves, cows, and herds and between lesions of the same type in different hooves within cow. 2. Materials and methods 2.1. Selection of herds and individual cows All dairy farmers in south-western Sweden (4204 herds) received a questionnaire regarding general management, housing system, claw trimming routines and interest in participating in a hoof-health study. Herds to be admitted for inclusion were selected from the replies. Inclusion criteria were membership in the of cial milk recording scheme (OMRS), a herd-size greater than 25 cows with most animals either Swedish Red and White (SRB) or Swedish Holstein (SLB), no plans of rebuilding the dairy herd's housing facilities within 2 years time, and willingness to participate in the study. Altogether, 101 herds were included. Herds were drawn randomly (by lottery) from a list of all herds within strata representing different housing systems: short or long tie-stalls or cubicles. Cows in short-stalls had unlimited access to the feeding table, whereas cows in long-stalls were allowed access for limited amounts of time (Table 1). The relative sizes of the strata re ected the current proportions of different housing systems according to the replies to the questionnaire, but with some modi cation

T. Manske et al. / Preventive Veterinary Medicine 54 (2002) 247±263 249 Table 1 Distribution of 101 Swedish dairy herds included in a study of hoof-health 1996±1998 on different levels of factors related to housing, management and feeding Long tie-stalls (n ˆ 31 herds) Short tie-stalls (n ˆ 55 herds) Cubicles (n ˆ 15 herds) Introduction to milking herd 1 week before calving 6 18 4 1 2 weeks before calving 4 6 2 2±3 weeks before calving 4 7 2 3 weeks before calving 17 24 7 Lying surface Rubber mats 12 37 8 Littered concrete 19 18 7 Electric cow-trainers ± a Yes 6 16 No 25 39 Gutter grates ± a Yes 4 21 No 27 34 Flooring ± a ± a Slatted concrete 10 Solid concrete 5 Abrasiveness of flooring ± a ± a Slippery 7 Moderately rough 8 Very rough 0 Grouping ± a ± a No grouping 8 Production level 4 Udder health 3 Access to feed <6 h daily 8 0 0 6±8 h daily 9 0 0 >8 h daily 14 55 15 Feeding of roughages Restricted 28 15 1 Ad lib. 3 40 14 Calculation of feed ration Individual 11 30 12 Standardised 14 18 2 Other 6 7 1 a Not applicable. according to contemporary building trends, favouring herds with short tie-stalls or cubicles. The low number of eligible herds in the stratum representing cubicle herds did not make a random selection possible and hence all cubicle herds ful lling the inclusion criteria were included.

250 T. Manske et al. / Preventive Veterinary Medicine 54 (2002) 247±263 In each herd, heifers 7 months of gestation and all cows regardless of stage of lactation were eligible for examination. However, certain animals were excluded for practical reasons (not available at time of trimming or impending slaughter; n ˆ 282), or for humane reasons (cows and heifers in close proximity to calving, recently calved cows, or animals with concurrent diseases are not related to the locomotor system; n ˆ 12). Exclusions of some cattle close to slaughter were made due to the farmers' unwillingness to trim these animals. Miscoding or other human errors n ˆ 76 and other unspeci ed reasons n ˆ 50 accounted for further losses. Seventy-seven cows had been at pasture for a period 30 days before the examination; these animals were omitted from the analyses. To avoid interdependence of recordings, each animal was examined only once, on its rst appearance in the project. Due to a simultaneous experiment on the impact of different claw trimming frequencies on hoof-health, only 50% of eligible cattle were trimmed and examined in the autumn in 80 herds; the rest were examined in the following spring. Withdrawal of two of these herds for accidental reasons (one farm burned down, the other was inadvertently trimmed without recording of the hoof-health) before the spring visit caused only 50% of the animals in these herds to be examined. In the remaining 21 herds, all animals were examined at autumn trimming. Records on the hoof-health status of 4899 animals in 101 herds were used in the analyses. The number of animals examined per herd varied between 12 and 148 (median 43). 2.2. Herds: animals, housing and management Information on breed, parity and calving date was collected from the OMRS. Of the examined animals, 2700 were Swedish Holsteins, 2030 Swedish Red and White and 169 other dairy breeds or crosses between the two major breeds. Information on parity was available for 4882 cattle; of these, 5% were pregnant heifers, 45% were in parity 1, 22% in parity 2, 13% in parity 3 and 15% in parity 4. Of the cows with at least one calving n ˆ 4597, 22% were in the rst 2 months of lactation, 24% in the third or fourth month and the remaining 54% in the fth month of lactation or more or dried-off at the time of examination. Information on housing facilities, diet, feeding and management routines were obtained at one separate visit to the herds by one of the authors (TM or CB). Lactating cows were housed in long tie-stalls (n ˆ 31 herds), short tie-stalls (n ˆ 55 herds), or in cubicles (n ˆ 15 herds). Herds with tie-stalls were generally smaller (median 42 examined animals per herd) than herds with cubicles (median 71 examined animals per herd). To assess how well the included herds represented the dairy herd population in the area, comparisons were made with of cial databases. In all included herds, cattle were on maintenance claw trimming schemes already before the study. Claw trimming was performed once per year in 49 herds, at least twice per year in 39 herds, and on an irregular basis ``when needed'' in 13 herds. In 99 herds, the animals were grazed for varying lengths of time in the summer. In the remaining two herds, the cattle had daily access to open-air exercise pens during summer. The distribution of herds according to some relevant housing, management, and feeding factors is presented in Table 1.

2.3. Examination of cattle Locomotion score and presence and severity of hoof lesions were recorded at claw trimming from October 1996 to May 1997 and from September 1997 to June 1998. Professional hoof trimmers trimmed the cows in stand-up chutes. Diagnoses of lesions were made by a specially trained technician (n ˆ 3595) or by one of the authors (TM, n ˆ 1006; CB, n ˆ 375). There was a complete confounding of herd and observer; hence, possible herd effects on the hoof-health were indistinguishable from incongruities in scoring. For instance, the technician only recorded hoof-health on herds that prior to the study had only one yearly maintenance trimming, whereas TM and CB recorded hoofhealth on herds that were on a twice yearly trimming-plan. The gait of each animal was assessed before trimming, when the animal was moved to the trimming chute. Due to impracticalities associated with the housing facilities, locomotion was not assessed in two herds. The hoof-health was recorded as presence or absence of speci ed hoof lesions. Lesions were diagnosed on the basis of macroscopic examination, regardless of hypothesised aetiology. Although locomotion and some hoof lesions originally were scored on ordinal scales from 0 to 3, these scores were reclassi ed for the analyses as the presence or absence of lameness and lesions (Table 2). The congruity in scoring hoof lesions between different observers was improved through a group scoring exercise prior to the study. Moreover, photographic slides were obtained from examined hooves and were later used to study inter- and intraobserver agreement in scoring. Brie y, the intraobserver repeatability in scoring different lesions on the original 4-level ordinal scale was substantial, with average kappa(k) coef cients 60% for all lesions. The interobserver repeatability was excellent for the scoring of sole ulcer (k ˆ 86%), and good to fair for heel-horn erosion (k ˆ 62%), sole haemorrhage (k ˆ 51%), dermatitis (k ˆ 46%), white-line ssure (k ˆ 44%) and white-line haemorrhage (k ˆ 40%). 2.4. Statistical analysis T. Manske et al. / Preventive Veterinary Medicine 54 (2002) 247±263 251 The original records from the claw trimmings were entered into FileMaker Pro 3.0 (Claris Corp., 1996). Subsequent editing of the data (checks for logical compatibility and biological plausibility) and statistical calculations were made using the statistical software package JMP 4.0 (SAS Institute Inc., 2000) and SAS 8.01 (SAS Institute Inc., 2000). With the intention of simplifying the analysis and interpretation of this study and forthcoming on the same data, we made an a priori decision to aggregate hind hoof lesions into three cow level lesion complexes: HAEMORRHAGES, DERMATITIS and SEPARA- TIONS (Table 2). Moreover, the observation of a sole ulcer in one or both hind hooves was scored on cow level as SOLE ULCERS. The aggregation of lesions was based on empirical knowledge and literature and also con rmed using principal-component analysis with varimax rotation (SAS Institute Inc., 2000). Prevalence of hoof lesions and lameness was calculated as the number of affected animals divided by the number examined. Variance components analysis was performed using the GLIMMIX macro in SAS, modelling each of the hoof lesions and lameness separately including only herd as a random factor and a random intercept. Because of the non-binomial distribution of residuals when modelling on hoof level, hooves could not be used as the unit of observation. Instead,

252 T. Manske et al. / Preventive Veterinary Medicine 54 (2002) 247±263 Table 2 Definitions of hoof traits as recorded at claw trimming in 101 Swedish dairy herds 1996±1998 or as aggregated for the analyses Lesion DERMATITIS Erosive dermatitis Verrucose dermatitis Heel-horn erosion HAEMORRHAGES Haemorrhages in sole and white-line horn SEPARATIONS Double sole White-line fissure SOLE ULCERS Sole ulcer Abnormal claw shape IH Lameness Score definition Erosive or verrucose dermatitis or heel-horn erosion in a hind hoof 0: Intact skin in ID cleft and adjacent areas of the foot 1: Hyperaemic, exudative lesions in the skin; erosions or ulcerations; skin thickening in the ID cleft 0: Intact skin or erosions/ulcerations bordering the claw capsule 1: Proliferative (papillomatous) lesions in the skin bordering the claw capsule 0: Intact heel-horn or superficial loss of heel-horn (``pitting'') 1: Loss of horn of the heel bulb (fissures or craters) Haemorrhages in the sole or white-line of a hind hoof 0: No haemorrhage or slight haemorrhagic discoloration of horn tissue 1: Haemorrhagic discoloration either covering 50% of the sole area or a smaller area with a deep intense colour White-line fissure or double sole in a hind hoof 0: Continuous sole horn, from the surface 1: Under-run sole horn 0: Sole horn continuous with wall horn via an intact white-line 1: Separation of sole from wall; disrupted continuity of white-line Sole ulcer in a hind hoof 0: Intact sole horn 1: Exposed corium on the solar surface of the claw capsule 0: No deviation from normal claw shape 1: Corkscrew, scissor-beak or other deformation of the claw capsule (e.g. severe overgrowth, causing loss of contact between toe and ground) 0: No proliferation of the ID skin fold 1: Fibrous proliferation of the ID skin fold 0: Normal gait 1: Asymmetric gait, avoiding weight-bearing on one or more limbs hoof-health status was collapsed over cow, assuming a cow to be affected if she had at least one affected hoof. When analysing on cow level, the residual variation was 0.69±0.97, indicating some remaining underdispersion. In the nal analyses, this was adjusted by constraining the residuals to 1.0 (corresponding to no extra-binomial variation). Associations between lesions were studied using Spearman's rank correlations. At the hoof and cow levels, correlations were calculated on binary scores, whereas correlations at the herd level were calculated on the herd-speci c animal-level prevalence of the different lesions. To compensate for some of the increased risk of Type-1 errors with multiple comparisons, the a-level was set to 0.01 in this analysis. 3. Results The size distribution of included herds was comparable to that of all herds in the OMRS in the relevant area of Sweden. Although herd breed-composition was an inclusion

T. Manske et al. / Preventive Veterinary Medicine 54 (2002) 247±263 253 Table 3 Overall and herd-specific animal-level (minimum, maximum, and quartiles, Q) prevalence of hoof lesions and lameness, and variance components due to between-herd variation in 4899 Swedish dairy cattle from 101 herds, examined at routine claw trimming, 1996±1998 Overall (%) Herd-specific (%) Front hooves Hind hooves All hooves Min. Q 1 Q 2 Q 3 Max. Variance component DERMATITIS ± 47 ± 0.0 26 43 66 98 53 Erosive dermatitis 1.0 26 27 0.0 13 25 36 69 40 Verrucose dermatitis 0.1 2.2 2.3 0.0 0.0 1.6 3.6 19 41 Heel-horn erosion 13 38 41 0.0 17 33 63 91 62 HAEMORRHAGES ± 32 ± 6.7 23 32 41 58 17 Sole haemorrhage 9.8 27 30 3.4 21 30 39 56 19 White-line haemorrhage 4.0 12 14 0.0 7.4 12 19 37 19 SEPARATIONS ± 9.8 ± 0.0 3.9 7.7 14 38 35 Double sole 0.8 3.3 3.3 0.0 0.0 2.4 6.4 27 42 White-line fissure 1.6 7.7 8.8 0.0 2.6 6.2 12 47 40 SOLE ULCERS ± 8.3 ± 0.0 3.7 7.7 12 26 26 Sole ulcer 0.5 8.3 8.6 0.0 3.8 9.4 14 33 32 Abnormal claw shape 7.9 18 21 0.0 5.6 21 32 62 54 IH 0.0 1.8 1.8 0.0 0.0 0.0 3.2 13 32 Any lesion 31 63 72 18 57 70 79 98 33 Lameness ± ± 5.1 0.0 0.0 3.7 7.9 33 40 criterion, the proportion of breeds in the study (42% SRB, 55% SLB, 3% other breeds or crosses) corresponded reasonably well to the proportion in herds in the OMRS (45% SRB, 52% SLB, 3% other breeds or crosses) (Swedish Dairy Association, 2000). The overall and herd-speci c animal-level prevalence of different traits and the variance component due to between-herd variation is given in Table 3. The prevalence of hoof lesions varied greatly between herds. Sole haemorrhages were found in all herds. Most animals had at least one hoof lesion. The most common lesions were heel-horn erosion, sole haemorrhages and dermatitis. Most lesions were approximately 2±4 times more prevalent in the hind than in the front hoovesðdermatitis and interdigital hyperplasia (IH) even more so. Correlations between lesions at different levels and between hooves within cow are presented in Table 4. The correlation patterns at different levels are further illustrated in Fig. 1. Correlations between lesions were generally positive, with a notable exception at the herd level, where ranking for prevalence of white-line ssure and abnormal claw shape were negatively correlated. Correlations between lesions in the right and left hind and front hooves were higher than between lesions in any front and any hind hoof. Most animals with abnormal claw shape (66%) and heel-horn erosion (63%) in one hind hoof had the same type of lesion in the contra-lateral hind hoof; the corresponding gure for sole ulcer was 29%. All hoof lesions (except dermatitis) were more prevalent (P < 0:001) in lame than in non-lame cattle. Hence, the risk of lameness was increased (odds ratio with associated 95%

254 T. Manske et al. / Preventive Veterinary Medicine 54 (2002) 247±263 Table 4 Correlations (Spearman's rho) between different types of hoof lesions at the hoof (n ˆ 19; 596), cow (n ˆ 4899) and herd (n ˆ 101) level, and between presence of the same type of hoof lesions in hind hooves, in front hooves, and in any hind and any front hoof (data from routine claw trimming in Swedish dairy herds, 1996±1998) Lesion 1 2 3 4 5 6 7 8 9 10 1. Erosive dermatitis 2. Verrucose dermatitis 0.15 a 0.19 a 0.42 a 3. Heel-horn erosion 0.36 a 0.08 a 0.37 a 0.10 a 0.60 a 0.26 b 4. Sole haemorrhage 0.09 a 0.03 a 0.11 a 0.04 b 0.02 0.09 a 0.04 0.13 0.05 5. White-line haemorrhage 0.04 a 0.01 0.08 a 0.25 a 0.01 0.00 0.07 a 0.28 a 0.03 0.08 0.13 0.66 a 6. Double sole 0.03 a 0.01 0.06 a 0.06 a 0.03 a 0.03 c 0.00 0.08 a 0.11 a 0.05 a 0.13 0.15 0.37 a 0.29 b 0.27 b 7. White-line fissure 0.03 a 0.00 0.05 a 0.04 a 0.06 a 0.16 a 0.00 0.02 0.05 a 0.06 a 0.08 a 0.20 a 0.07 0.17 0.06 0.03 0.18 0.26 b 8. Sole ulcer 0.04 a 0.02 b 0.13 a 0.19 a 0.06 a 0.14 a 0.03 a 0.01 0.00 0.12 a 0.26 a 0.08 a 0.19 a 0.05 b 0.09 0.02 0.25 c 0.41 a 0.17 0.48 a 0.14 9. Abnormal claw shape 0.03 a 0.01 0.11 a 0.11 a 0.05 a 0.06 a 0.03 a 0.13 a 0.01 0.02 0.12 a 0.20 a 0.07 a 0.11 a 0.04 b 0.41 a 0.04 0.08 0.23 c 0.13 0.03 0.17 0.36 a 0.54 a 10. Interdigital hyperplasia 0.09 a 0.07 a 0.08 a 0.02 c 0.01 0.03 a 0.04 a 0.04 a 0.05 a 0.11 a 0.09 a 0.09 a 0.00 0.00 0.02 0.05 a 0.05 a 0.06 a 0.39 a 0.37 a 0.41 a 0.20 c 0.24 c 0.29 b 0.31 b 0.15 0.03 Same type of lesion Between hind hooves 0.64 a 0.35 a 0.67 a 0.56 a 0.39 a 0.26 a 0.30 a 0.41 a 0.76 a 0.35 a Between front hooves 0.43 a 0.00 0.69 a 0.45 a 0.35 a 0.30 a 0.27 a 0.50 a 0.79 a ± Between hind and front hooves 0.12 a 0.12 a 0.33 a 0.30 a 0.21 a 0.01 0.12 a 0.07 a 0.33 a 0.07 a a P 0:001. P 0:01. P 0:05. con dence interval) for cows with SOLE ULCERS (6.02 [4.56, 8.14]), SEPARATIONS (2.77 [1.99, 3.85]), HAEMORRHAGES (2.35 [1.80, 3.06]), and DERMATITIS (1.54 [1.18, 1.93]). When lesion aggregates were analysed together, the association between dermatitis and lameness was no longer signi cant (P ˆ 0:11).

T. Manske et al. / Preventive Veterinary Medicine 54 (2002) 247±263 255 Fig. 1. Correlation patterns between hoof lesions (AS: abnormal claw shape, DS: double sole, D: erosive dermatitis, HE: heel-horn erosion, IH: interdigital hyperplasia, SH: sole haemorrhage, SU: sole ulcer, W: verrucose dermatitis, WLF: white-line fissure, and WLH: white-line haemorrhage) at three levels. (a) The hoof (n ˆ 19; 596); (b) the animal (n ˆ 4899); (c) the herd (n ˆ 101). The line-thickness is proportional to the magnitude of the Spearman's rank correlation (r s 0:10, P 0:01); a dashed line indicates a negative association. Data collected at routine claw trimming in Swedish dairy herds 1996±1998. 4. Discussion 4.1. Representativeness Our study included herds selected from a questionnaire with a reply frequency of slightly >50% (Hultgren et al., 1998). Herdsmen answering the questionnaire and agreeing to enter the study might have been more interested in hoof-health matters, and their herds thus might have had a better hoof-health than other herds in the population (Mill and Ward, 1994). On the other hand, herds that had experienced hoof-health problems might also have been likely to enrol in order to get quali ed help. Andersson and LundstroÈm (1981) estimated that 40% of Swedish dairy cows were trimmed. This 20-year-old result probably

256 T. Manske et al. / Preventive Veterinary Medicine 54 (2002) 247±263 underestimates the proportion of cattle that are trimmed routinely today. According to Hultgren et al. (1998), 78% of Swedish dairy herds are on a regular claw trimming scheme. In the present study, cattle in all herds were trimmed routinely. Because maintenance claw trimming is not performed in all Swedish dairy herds, there might have been a positive selection for herdsmen with a relatively high awareness of hoof-health matters in the present study. In summary, the included herds were believed to represent all dairy herds in south-western Sweden well. Nine percent (497/5396) of eligible cattle were withheld from examination for different reasons. Many of these animals were at a low risk of hoof lesions (dry cows, cows at a late stage of lactation or pregnant heifers). Thus, these omissions might have caused a slight over-estimation of the prevalence of hoof lesions. On the other hand, some animals that were intended for culling were withheld from trimming (51% of all omissions). An initial attempt at collecting and examining the hooves of all culled animals had to be abandoned due to lack of resources. In contrast to the results of Barkema et al. (1994), Collick et al. (1989) and Sprecher et al. (1997) found hoof lesions (lameness) to be an important cause of culling. If the risk of culling in our study was increased by lameness, an underestimation of the true prevalence of hoof lesions and lameness is possible. An initial attempt to examine the hooves of culled animals was abandoned for practical reasons. In only ve herds, animals were examined at only one occasion. The risk of spurious herd effects due to examination (trimming) in different seasons was therefore probably minimised. 4.2. Scoring and statistical analysis Hitherto, the repeatability of scoring systems for hoof lesions largely has been overlooked. Without inter- and intraobserver repeatability, the results of surveys can be misleading (Murray et al., 1994). In our study, the agreement between lesion scores from different observers was enhanced through a group scoring exercise prior to the start of the study and also assessed and deemed adequate in a post-hoc study. The complete confounding between observer and herd and between observer and previous claw trimming history caused uncertain results in the correlation analyses (see Section 4.3.5). Correlation analyses on data sets containing subsets of observations can cause spurious results (Altman, 1991) due to the exploratory nature of the analyses, we still chose to present the crude correlation estimates. 4.3. Hoof-health 4.3.1. Lameness The prevalence of lameness in our study was 5%. The results from earlier cross-sectional prevalence studies vary greatly, from the 1.2% reported in a study of Dutch zero-grazing herds (Smits et al., 1992) to the 13.7±16.7% reported from the USA (Wells et al., 1993) and the 20.6% reported from England (Clarkson et al., 1996). We assessed the gait when animals were led from stalls or groups to the trimming chute and the conditions for examination were often sub-optimal; our estimate of lameness prevalence might thus be an underestimation of the real prevalence.

T. Manske et al. / Preventive Veterinary Medicine 54 (2002) 247±263 257 If the duration of lameness is short relative to the length of lactation, the observed prevalence re ects an even-higher lactation incidence. The of cially reported lactation incidence of veterinary treatments for hoof and leg diseases in Sweden was 2% (Swedish Dairy Association, 2000). The low incidence of veterinary-treated cases of lameness compared to the incidence estimated from our study is in agreement with Murray et al. (1996) and indicates that many lame cattle are either treated by farmers or hoof trimmers or are left untreated. The between-herd proportion of variability in lameness of 40% indicates that herd-level factors are important causes of lameness and con rms the high variability in lameness prevalence between herds reported by Clarkson et al. (1996) and Whitaker et al. (2000). We made no attempt at diagnosing the cause of lameness, but SOLE ULCERS had the strongest association with lameness followed by SEPARATIONS (i.e. double sole or whiteline ssure). DERMATITIS was often a coincidental nding. The result of our crosssectional study is in agreement with records from lameness treatments in UK dairy herds, where sole ulcers and white-line lesions were the most common lesions in lame cows, with 28 and 22% (Murray et al., 1996). 4.3.2. Overall prevalence of lesions Although all examinations were performed during the housing season, the animals were examined at different lengths of time after being housed. Because hoof-health commonly improves when cattle are on pasture (Peterse, 1985), cattle examined close to the time of housing generally have a better hoof-health than cattle examined later in the housing season (Andersson and LundstroÈm, 1981). A higher prevalence of lesions would have been expected, had all animals been examined in spring. Smits et al. (1992) studied the hoof-health of zero-grazed dairy cattle and found lesions in 96% of the animals. Philipot et al. (1994) examined only one hind hoof; 89% of the animals had at least one hoof lesion. Vaarst et al. (1998) reported the results from ecological and ordinary Danish dairy herds and found that 37% of the animals had lesions. Due to different objectives of these studies, the types of recorded lesions differed from our study and the results are thus not completely comparable. The variation in herd prevalence of hoof lesions is in agreement with earlier studies (Smits et al., 1992; Smilie et al., 1996, 1999; Murray et al., 1996; Vaarst et al., 1998), and indicates that herd-level factors are important for the prevalence of hoof lesions (Frankena et al., 1992). In our study, the proportion of variability due to between-herd (relative to within-herd) variation was particularly large for heel-horn erosion and abnormal claw shape. Different levels of hygiene in the hoof-environment can explain the large between-herd variation for heel-horn erosion (Bergsten and Pettersson, 1992; Hultgren and Bergsten, 2001). Because abnormal claw shape included hoof overgrowth, differences between herds in the prevalence of this type of lesion can be explained partly by different claw trimming routines. Smilie et al. (1999) ranked 13 herds according to the prevalence of different lesions associated with sub-clinical laminitis, and found no persistent ranking between lesions. In contrast, we found that herds that ranked high with respect to the prevalence of sole haemorrhages also ranked high for sole ulcers, white-line haemorrhages, and double soles. Correspondingly, herds that ranked high for heel-horn erosion also ranked high for erosive and proliferative dermatitis, IH and double sole.

258 T. Manske et al. / Preventive Veterinary Medicine 54 (2002) 247±263 Hind hooves were more affected by lesions than front hooves, which is in agreement with earlier ndings (Andersson and LundstroÈm, 1981; Murray et al., 1996; Bergsten et al., 1998). The difference between hind and front hooves was particularly large for dermatitis and IHÐcon rming Thysen (1987). In housed (especially tethered) cattle, front hooves are less exposed to manure and urine. In our study, there was a tendency for loose-housed animals to have a higher correlation between lesions of the hind and front hooves than tied; the differences were, however, not signi cant. Lesions with a high prevalence typically occurred simultaneously on both left and right hind and front hooves, respectivelyðsupporting the suggestion of Le Fevre et al. (2000), that the contra-lateral hoof is a good indicator of lesions. Andersson and LundstroÈm (1981) and Philipot et al. (1994) reported correlations between different hoof lesions (e.g. between sole haemorrhages and sole ulcer, between ID dermatitis and heel-horn erosion and between sole ulcer, double sole and white-line separation) at hoof level. In our study, we examined the correlation between different types of lesions at three hierarchical levels. Two different hoof lesions would be associated at the hoof level if they shared some common causative mechanisms, if one lesion caused the other, or if separate causative mechanisms coexisted. At the cow levelðand even more so at the herd levelðlesions that re ect different developmental stages of the same underlying disease also would be associated. Due to the possibility of different causes of association at different hierarchical levels, it is unwise to extrapolate the results to another level. 4.3.3. Dermatitis and heel-horn erosion As in the study by Andersson and LundstroÈm (1981), heel-horn erosion was the most common lesion. Enevoldsen et al. (1991b) and Philipot et al. (1994) also reported a high prevalence of heel-horn erosion in dairy cattle. Heel-horn erosion is particularly common in housed cattle (Enevoldsen et al., 1991b)Ðpossibly due to faecal contamination of stalls and alleys (Bergsten and Pettersson, 1992; MuÈlling and Budras, 2002; Hultgren and Bergsten, 2001). Fewer and less-severe lesions are seen during the grazing season (Peterse, 1985) suggesting a curative or preventive effect of improved hygiene or increased exercise. The positive (although not so strong) correlation at the hoof and cow levels between heel-horn erosion and laminitis-associated lesions (sole haemorrhage and sole ulcer) in our study is in agreement with earlier results (Andersson and LundstroÈm, 1981; Toussaint Raven, 1989; Greenough and Vermunt, 1991). Horn produced during and after bouts of laminitis has an impaired quality and is more susceptible to erosion (Greenough and Vermunt, 1991). Conversely, loss of heel-horn might destabilise the weight-bearing of the claw and cause compression of the corium due to overburdening (Toussaint Raven, 1989). Dermatitis is predominantly a problem in an unhygienic hoof environment (Bergsten, 1997), a situation in which softening of the claw capsule occurs (Bergsten and Pettersson, 1992). The lack of association between dermatitis and laminitis-associated lesions in our study indicates that softening of the claw capsule might not be an important risk factor for the development of laminitis lesions. The high correlation between heel-horn erosion and dermatitis is in agreement with Andersson and LundstroÈm (1981) and Philipot et al. (1994)Ðbut not with Enevoldsen et al. (1991b), who only reported an association in primiparous cows. Our results indicate that

T. Manske et al. / Preventive Veterinary Medicine 54 (2002) 247±263 259 the two lesions might be both parts of the same process (as advocated by e.g. Toussaint Raven, 1989) and share common causative mechanisms. We did not make a distinction between ID and digital dermatitis (DD) because ID and early (mild) stages of DD are very similar, and occur in the same area of the hoof (Bergsten, 1997). DD causing lameness, as described from many other countries, is a rare nding in Sweden. On the other hand, we made a distinction between verrucose (proliferative) and erosive dermatitis. This distinction is easily discernible but might be arbitrary; proliferative and erosive lesions have been suggested to be different evolutionary stages of the same disease (DoÈpfer et al., 1997). The latter notion is supported by the recent demonstration of spirochetes in both types of lesions (Manske, T., unpublished observations) and the association between the two types of dermatitis at cow and herd levels found in the present study. Erosive dermatitis had a higher prevalence in our study than the 5±8% reported by Enevoldsen et al. (1991b), but lower than the 83% in zero-grazed cattle (Smits et al., 1992), and the 66% in dairy calves (Frankena et al., 1992). Because several cattle were examined in the autumn (shortly after housing), one reason for the lower prevalence of dermatitis in our study than in the zero-grazing animals might be the apparently curative effect of grazing (Andersson and LundstroÈm, 1981; Peterse, 1985). Verrucose (proliferative) dermatitis was relatively rare in our study, with an overall prevalence of 2% and a variation in herd-speci c prevalence between 0 and 19%. In a limited study of culled dairy cattle in USA, Brown et al. (2000) found papillomatous DD lesions in 29% of examined dairy cattle. The cow level association between dermatitis and IH and the lack of association between abnormal claw shape and IH indicates that dermatitis might be of relatively more importance in the development of IH than other causes, e.g. splay toes. Herds that ranked high for IH prevalence also ranked high for lesions related to unhygienic environment (dermatitis and heel-horn erosion), offering further support to the association between a chronic irritation to the ID skin and the development of IH. 4.3.4. Haemorrhages of the sole and white-line British researchers have hypothesised that the causative mechanisms differ for haemorrhages in the sole and white-line (Leach et al., 1997) and argued in favour of analysing sole and white-line lesions separately (Le Fevre et al., 2000). In our study, sole and whiteline haemorrhages were strongly associated at all levels, but the correlations between sole and white-line haemorrhages and other hoof lesions differed markedlyðpossibly offering support to the theory of different causative mechanisms proposed by the British researchers. Haemorrhages in the horn of the sole and white-line are thought to re ect earlier episodes of laminitis and (or) hormonally induced alterations in the suspensory apparatus of the pedal bone (Nilsson, 1963; Kempson and Logue, 1993; Bergsten, 1994; Ossent, 1999). The prevalence of such lesions in this study (30% sole, and 14% white-line haemorrhages) is lower than in earlier Swedish studies; Bergsten (1994) reported a prevalence of 79% sole haemorrhages in 11 herds with no treatments of clinical laminitis and Andersson and LundstroÈm (1981) a 60% prevalence of sole haemorrhages in culled dairy cattle. Smilie et al. (1999) reported 62% sole haemorrhages in rst-lactation cows and

260 T. Manske et al. / Preventive Veterinary Medicine 54 (2002) 247±263 Enevoldsen et al. (1991a) 50% in Danish dairy herds. In contrast to other studies, we did not consider mild haemorrhages (``paintbrush-marks'') and in contrast to the studies by Bergsten and Enevoldsen, et al., we recorded sole ulcer as a separate entity. It is also possible that the different housing systems in our study (more tied animals) relative to the Danish and American contributed to a lower prevalence of lesions (Thysen, 1987). The proportion of variance due to between-herd variability was relatively low for haemorrhages of the sole and white-line, indicating a greater importance of individuallevel factors (e.g. stage of lactation, parity, breed) for the presence of these lesions. 4.3.5. White-line fissure and double sole White-line ssures and double soles are caused by disturbances in the horn production following laminitis (Ossent et al., 1997). No abscesses of the white-line (``white-line disease'') were diagnosed in our study. In high-producing dairy cattle, Smilie et al. (1996) found 16% white-line ssures and Philipot et al. (1994) found white-line separation in 20% and double sole in 8% of examined animals. In the study of Smilie, et al., all cattle were housed in cubicles. It is possible that a higher weight-bearing and shearing forces on the wall of the claw capsule in cubicle systems (Bergsten, 2001) account for the higher prevalence, at least in the study by Smilie. The negative association between white-line ssures and abnormal claw shape (overgrowth) in our study was unexpected. The association disappeared when data were strati ed on recorderðpossibly indicating differences in scoring. Because recorder was confounded completely with previous claw trimming routines, it is also possible that claw trimming (in reducing abnormal claw shapes) causes white-line lesions, possibly through an exposure of immature horn in the white-line. The lack of association between white-line ssures and white-line haemorrhages in our study indicate that the two lesions neither share causative mechanisms nor are different developmental stages of the same lesion-type, as suggested by MuÈlling (2002). 4.3.6. Sole ulcer Sole ulcers are caused by circulatory disturbances in the corium (laminitis) exacerbated by compression of the corium between the distal phalanx and the underlying sole horn (Ossent, 1999). Sole ulcers heal slowly even if treated (Lischer, 2000) and are painful (Whay et al., 1998). Due to poor prognosis and premature culling, it is the most costly hoof lesion in dairy cattle (Kossaibati and Esslemont, 1997). The high risk of lameness in cows with sole ulcers in our study is in agreement with an UK study, in which sole ulcers were the most common cause of lameness, with an ulcer found in 28% of 8645 lameness cases (Murray et al., 1996). Other studies (which were also not focused on lame cattle) indicated a somewhat lower prevalence than ours; in zero-grazed Dutch cattle, the prevalence of sole ulcer was 6% (Smits et al., 1992) and in a French study 5% (Philipot et al., 1994). In a slaughterhouse series, Andersson and LundstroÈm (1981) found a 14% prevalence of sole ulcer with a variation between seasons and breedsðindicating that sole ulcers impart a signi cant risk of culling as also shown by Sprecher et al. (1997). The association between sole ulcers, sole haemorrhages, double soles and abnormal claw shapes on all levels supports a common aetiologyðmost likely, laminitis.

T. Manske et al. / Preventive Veterinary Medicine 54 (2002) 247±263 261 5. Conclusion Hoof lesions are common among dairy cows in today's intensive Swedish dairy production. Because the most-prevalent lesions were not associated with lameness, only records obtained at examination of all animals will give a correct estimate of the lesion prevalence. Sole ulcers had the strongest association with lameness. Hoof lesions are correlated at the hoof, cow and herd levels but the correlation patterns differ between levels. Acknowledgements The authors gratefully acknowledge Henrik Stryhn for statistical advice, the different claw trimmers that made this study possible (especially Anette Svensson, our research technician), and all participating farmers. The quality of this paper was greatly improved by the comments from one of the referees. This study was supported nancially by the Swedish Farmer's Foundation for Research and AGROVAÈ ST. References Altman, D.G., 1991. Practical Statistics for Medical Research. Chapman & Hall, London, pp. 283±284. Andersson, L., LundstroÈm, K., 1981. The influence of breed, age, body weight and season on digital diseases and hoof size in dairy cows. Zbl. Vet. Med. A 28, 141±151. Bargai, U., 2000. Lameness in a dairy herd: an epidemiological model. Comput. Cont. Educ. Pract. 22, 58±67. Barkema, H.W., Westrik, J.D., van Keulen, K.A.S., Schukken, Y.H., Brand, A., 1994. The effects of lameness on reproductive performance, milk production and culling in Dutch dairy farms. Prev. Vet. Med. 20, 249±259. Bergsten, C., 1994. Haemorrhages of the sole horn of dairy cows as a retrospective indicator of laminitis: an epidemiological study. Acta Vet. Scand. 35, 55±66. Bergsten, C., 1997. Infectious diseases of the digits. In: Greenough, P.R., Weaver, A.D. (Eds.), Lameness in Cattle, 3rd Edition. Saunders, Philadelphia, PA, pp. 94±96. Bergsten, C., 2001. Effects of conformation and management system on hoof and leg diseases and lameness in dairy cows. In: Anderson, D. (Ed.), Veterinary Clinics of North America: Food Animal Practice, Vol. 17. Saunders, Philadelphia, PA, pp. 1±23. Bergsten, C., Pettersson, B., 1992. The cleanliness of cows tied in stalls and the health of their hooves as influenced by the use of electric trainers. Prev. Vet. Med. 13, 229±238. Bergsten, C., Hancock, D.D., Gay, J.M., Gay, C.C., Fox, L.K., 1998. Claw diseases: the most common cause of dairy lameness diagnoses, frequencies and risk groups in a university herd. In: Williams (Ed.), Proceedings of the 31st Annual Conference on American Association of Bovine Pract., Spokane, USA, pp. 188±194. Brown, C.C., Kilgo, P.D., Jacobsen, K.L., 2000. Prevalence of papillomatous digital dermatitis among culled adult cattle in the southeastern United States. Am. J. Vet. Res. 61, 928±930. Claris Corp., 1996. FileMaker Pro. Handbook. Studentlitteratur, Lund, Sweden. Clarkson, M.J., Downham, D.Y., Faull, W.B., Hughes, J.W., Manson, F.J., Merritt, J.B., Murray, R.D., Russell, W.B., Sutherst, J.E., Ward, W.R., 1996. Incidence and prevalence of lameness in dairy cattle. Vet. Rec. 138, 563±567. Collick, D.W., Ward, W.R., Dobson, H., 1989. Associations between types of lameness and fertility. Vet. Rec. 125, 103±106. DoÈpfer, D., Koopmans, A., Meijer, F.A., SzakaÂll, I., Schukken, Y.H., Klee, W., Bosma, R.B., Cornelisse, J.L., van Asten, A.J.A.M., ter Huurne, A.A.H.M., 1997. Histological and bacteriological evaluation of digital dermatitis in cattle, with special reference to spirochetes and Campylobacter faecalis. Vet. Rec. 140, 620±623.

262 T. Manske et al. / Preventive Veterinary Medicine 54 (2002) 247±263 Enevoldsen, C., GroÈhn, Y.T., Thysen, I., 1991a. Sole ulcers in dairy cattle: associations with season, cow characteristics, disease and production. J. Dairy Sci. 74, 1284±1298. Enevoldsen, C., GroÈhn, Y.T., Thysen, I., 1991b. Heel horn erosion and other interdigital disorders in dairy cows: associations with season, cow characteristics, disease and production. J. Dairy Sci. 74, 1299±1309. Frankena, K., van Keulen, K.A.S., Noordhuizen, J.P., Noordhuizen-Stassen, E.N., Gundelach, J., de Jong, D.-J., Saedt, I., 1992. A cross-sectional study into prevalence and risk indicators of digital haemorrhages in female dairy calves. Prev. Vet. Med. 14, 1±12. Greenough, P.R., Vermunt, J.J., 1991. Evaluation of sublinical laminitis in a dairy herd and observations on associated nutritional and management factors. Vet. Rec. 128, 11±17. Hultgren, J., Bergsten, C., 2001. Effects of a rubber-slatted flooring system on cleanliness and foot health in tied dairy cows. Prev. Vet. Med. 52, 75±89. Hultgren, J., Bergsten, C., Manske, T., 1998. Claw trimmings routines in relation to herd size and housing: a survey in Swedish dairy herds. In: Lischer, C.J., Ossent, P. (Eds.), Proceedings of the 10th International Symposium Lameness in Ruminants, Lucerne, Switzerland, September 7±10, 1998, pp. 26±27. Kempson, S.A., Logue, D.N., 1993. Ultrastructural observations of hoof horn from dairy cows: changes in the white line during the first lactation. Vet. Rec. 132, 524±527. Kossaibati, M.A., Esslemont, R.J., 1997. The costs of production diseases in dairy herds in England. Vet. J. 154, 41±51. Le Fevre, A.M., Logue, D.N., McKendrick, I., Gettinby, G., 2000. Interrelationships of the claw horn lesions of dairy cattle among claws and between the sole and white line revealed using logistic regression. In: Mortellaro, C.M., De Vecchis, L., Brizzi, A. (Eds.), Proceedings of the XI International Symposium Disorders of the Ruminant Digit, Parma, Italy, September 3±7, 2000, pp. 199±201. Leach, K.A., Logue, D.N., Kempson, S.A., Offer, J.E., Ternent, H.E., Randall, J.M., 1997. Claw lesions in dairy cattle: development of sole and white line haemorrhages during the first lactation. Vet. J. 154, 215±225. Lischer, C.J., 2000. SohlengeschwuÈre beim Rind: Pathogenese und Heilungsverlauf. Dissertation. VeterinaÈr Chirurgische Klinik UniversitaÈt ZuÈrich, Switzerland, pp. 40±55. Logue, D.N., Offer, J.E., Hyslop, J.J., 1994. Relationship of diet, hoof type and locomotion score with lesions of the sole and white line in dairy cattle. Anim. Prod. 59, 173±181. Mill, J.M., Ward, W.R., 1994. Lameness in cattle and farmers' knowledge, training and awareness. Vet. Rec. 134, 162±164. Mills, L.L., Leach, D.H., Smart, M.E., Greenough, P.R., 1986. A system for the recording of clinical data as an aid in the diagnosis of bovine digital disease. Can. Vet. J. 27, 293±300. MuÈlling, C., 2002. Theories on the pathogenesis of white line diseaseðan anatomical perspective. In: Shearer, J.K. (Ed.), Proceedings of the 12th International Symposium Lameness in Ruminants, Orlando, FL, January 9±13, 2002, pp. 90±98. MuÈlling, C., Budras, K.D., 1998. Influence of environmental factors on horn quality of the bovine hoof. In: Lischer, C.J., Ossent, P. (Eds.), Proceedings of the 10th International Symposium Lameness in Ruminants, Lucerne, Switzerland, September 7±10, 1998, pp. 214±215. Murray, R.D., Downham, D.Y., Merritt, J.R., Russell, W.B., Manson, F.J., 1994. Observer variation in field data describing foot shape in dairy cattle. Res. Vet. Sci. 56, 265±269. Murray, R.D., Downham, D.Y., Clarkson, M.J., Faull, W.B., Hughes, J.W., Manson, F.J., Merritt, J.B., Russell, W.B., Sutherst, J.E., Ward, W.R., 1996. Epidemiology of lameness in dairy cattle: description and analysis of foot lesions. Vet. Rec. 138, 586±591. Nilsson, S.A., 1963. Clinical, morphological and experimental studies of laminitis in cattle. Acta Vet. Scand. 4 (Suppl. 1). Ossent, P., 1999. Subclinical bovine laminitis. Cattle Pract. 7, 193±195. Ossent, P., Greenough, P.R., Vermunt, J.J., 1997. Laminitis. In: Greenough, P.R., Weaver, A.D. (Eds.), Lameness in Cattle, 3rd Edition. Saunders, Philadelphia, PA, pp. 277±292. Peterse, D.J., 1985. Laminitis and interdigital dermatitis and heel horn erosion. A European perspective. In: Ramanauskas, D., Ferguson, J.G. (Eds.), Veterinary Clinics of North America: Food Animal Practice, Bovine Lameness and Ortopedics, Vol. 1. Saunders, Philadelphia, PA, pp. 83±91. Philipot, J.M., Pluvinage, P., Luquet, F., 1994. Clinical characterization of a syndrome by ecopathology methods: an example of dairy cow lameness. Vet. Res. 25, 239±243.

T. Manske et al. / Preventive Veterinary Medicine 54 (2002) 247±263 263 Ral, G., Philipsson, J., Berglund, B., Manske, T., Hultgren, J., 2001. Selektionsindex foèr kloèvhaèlsa Report. Dept. Anim. Breed Gen., Swed. Univ. Agric. Sci., Uppsala, Sweden. SAS Institute Inc., 2000. JMP Statistical Discovery Software. Statistics and Graphics Guide. Smilie, R.H., Hoblet, K.H., Weiss, W.P., Eastridge, M.L., Rings, D.M., Schnitkey, G.L., 1996. Prevalence of lesions associated with subclinical laminitis in first-lactation cows from herds with high milk production. J. Am. Vet. Med. Assoc. 208, 1445±1451. Smilie, R.H., Hoblet, K.H., Eastridge, M.L., Weiss, W.P., Schnitkey, G.L., Moeschberger, M.L., 1999. Subclinical laminitis in dairy cows: use of severity of hoof lesions to rank and evaluate herds. Vet. Rec. 144, 17±21. Smits, M.C.J., Frankena, K., Metz, J.H.M., Noordhuizen, J.P.T.M., 1992. Prevalence of digital disorders in zerograzing dairy cows. Livest. Prod. Sci. 32, 231±244. Sprecher, D.J., Hostetler, D.E., Kaneene, J.B., 1997. A lameness scoring system that uses posture and gait to predict dairy cattle reproductive performance. Theriogenology 47, 1179±1187. Swedish Dairy Association, 2000. DjurhaÈlsovaÊrd 1999/2000 Report. Swedish Dairy Association, Eskilstuna, Sweden. Thysen, I., 1987. Foot and leg disorders in dairy cattle in different housing systems. In: Wierenga, H.K., Peterse, D.J. (Eds.), Cattle Housing Systems, Lameness and Behaviour, Fed. Rep. Martinus Nijhoff, Dordrecht, pp. 166±178. Toussaint Raven, E., 1989. Cattle Foot Care and Claw Trimming. Farming Press, Ipswich, p. 13. Vaarst, M., Hindhede, J., Enevoldsen, C., 1998. Sole disorders in conventionally managed and organic dairy herds using different housing systems. J. Dairy Res. 65, 175±186. Wells, S.J., Trent, A.M., Marsh, W.E., Robinson, R.A., 1993. Prevalence and severity of lameness in lactating dairy cows in a sample of Minnesota and Wisconsin herds. JAVMA 202, 78±82. Whay, H.R., Waterman, A.E., Webster, A.J., O'Brien, J.K., 1998. The influence of lesion type on the duration of hyperalgesia associated with hind limb lameness in dairy cattle. Vet. J. 156, 23±29. Whitaker, D.A., Kelly, J.M., Smith, S., 2000. Disposal and disease rates in 340 British dairy herds. Vet. Rec. 146, 363±367.