TY - JOUR
AU - Tucker, C. B.
AB - Abstract Epidemiological studies can be used to identify risk factors for livestock welfare concerns but have not been conducted in the cow–calf sector for this purpose. The objectives of this study were to investigate the relationships of 1) herd-level management, facilities, and producer perspectives with cattle health and behavior and stockperson handling and 2) stockperson handling on cattle behavior at the individual cow level. Cow (n = 3,065) health and behavior and stockperson handling during a routine procedure (e.g., pregnancy checks) were observed on 30 California ranches. Management and producer perspectives were evaluated using an interview, and handling facility features were recorded at the chute. After predictors were screened for univariable associations, multivariable models were built for cattle health (i.e., thin body condition, lameness, abrasions, hairless patches, swelling, blind eyes, and dirtiness) and behavior (i.e., balking, vocalizing, stumbling and falling in the chute and while exiting the restraint, and running out of the restraint) and stockperson handling (i.e., electric prod use, moving aid use, tail twisting, and mis-catching cattle). When producers empathized more toward an animal's pain experience, there was a lower risk of swelling (odds ratio [OR] = 0.7) but a higher risk of lameness (OR = 1.3), which may indicate a lack of awareness of the latter. Training stockpersons using the Beef Quality Assurance program had a protective effect on cow cleanliness and mis-catching in the restraint (OR = 0.2 and OR = 0.5, respectively). Hydraulic chutes increased the risk of vocalizations (OR = 2.7), possibly because these systems can apply greater pressure to the sides of the animal than manual restraints. When a moving aid was used to move an individual cow, it increased the risk of her balking, but when hands, in particular, were used, the risk of balking decreased across the herd (OR = 34.1 and OR = 0.3, respectively). Likewise, individual cows were at a greater risk of balking, vocalizing, stumbling and falling in the chute, and stumbling and running at exit when they were touched with an electric prod (OR = 11.0, OR = 3.3, OR = 1.9, OR = 2.3, OR = 1.8, and OR = 1.7, respectively). Although the implications of using moving aids are unclear, reducing the use of electric prods could improve cattle handling. In conclusion, cattle handling was influenced by a number of facility and stockperson factors: personnel training, facility design, and electric prod use are key areas for future improvements. INTRODUCTION Epidemiological studies provide evaluation of complex relationships among management, facilities, and animal responses and are valuable for addressing livestock welfare concerns, which are often multifaceted (Millman, 2009). The types of concerns include commonly measured welfare outcomes, such as animal health and behavior, which can be evaluated at the herd or individual animal level. For example, the prevalence of lameness in a dairy herd is associated with facility design and farm characteristics (Lombard et al., 2010; Chapinal et al., 2014), and lameness in individual cows is associated with body condition and lying behavior (Dippel et al., 2009b). Additionally, dairy cow udder health is associated with stockperson behavior in the milking parlor at both the herd and cow levels (Ivemeyer et al., 2011). To date, epidemiological welfare studies have primarily focused on specific concerns (e.g., lameness, morbidity; Dippel et al., 2009a; Brenninkmeyer et al., 2013) in intensive housing, but relatively little attention has been given to pasture-based systems. To date, there has been no comprehensive research investigating the associations among ranch characteristics, management, facilities, producer perspectives, stockperson handling, and cattle health and behavior on cow–calf operations at the herd or cow level. Identifying factors associated with animal welfare can inform decisions about how to strategically make improvements on the ranch. We predicted that management would be associated with cattle health, producer perspectives would influence stockmanship, and handling methods and facility design would subsequently be related to cattle behavior during a routine procedure. The objectives of this study were to investigate the relationships of 1) herd-level management, facilities, and producer perspectives with cattle health and behavior and stockperson handling and 2) stockperson handling on cattle behavior at the individual cow level. METHODS Research involving human subjects was exempt and in full compliance with the Code of Federal Regulations (United States Department of Health and Human Services. 2009). Ranch Selection We first connected with University of California Cooperative Extension specialists, county extension farm advisors, the California Cattlemen's Association (Sacramento, CA), and local veterinarians via phone and email to gain support and collect contact information for cow–calf producers in their area. A total of 50 producers were contacted by phone or email and asked if they would be willing to participate in our study; only 1 refused. Through convenience sampling, we recruited the first 30 cow–calf operations that processed their cows or heifers in the chute for a routine procedure (Table 1) on a day when assessors were able to visit. Each ranch was visited for 1 d between May 2014 and June 2015. All animals evaluated had been under the producer's ownership for at least the previous 21 d (the recommended isolation period for bovine respiratory disease; Callan and Garry, 2002). Table 1. Herd-level predictors considered for inclusion in the univariable analysis of variables associated with stockperson handling and cattle health and behavior on 30 California cow–calf operations Predictor  Category/unit  Mean (SD)  Frequency of yes  Model type1  Hyp2  Ranch characteristics and management      Breeding female herd size  No.  1,051 (1,849)    S, B, and H  –      Ranch age (no. of years of continuous operation)  No.  67 (51)    S, B, and H  –      Farming generations in producer's family  No.  3.7 (2.1)    S, B, and H  –      Cow herd handling events per year  No.  3.4 (5.2)    S, B, and H  –      Visual herd health checks per year  No.  195 (149)    H  –      Visual water access point checks per year  No.  185 (137)    H  –      Visual herd health checks in the previous 2 wk  No.  10.2 (7.0)    H  –      Visual water access point checks in the previous 2 wk  No.  9.7 (5.3)    H  –      Vet visits to the ranch in previous year  No.  5.7 (6.4)    H  –      Maintenance of written sick cattle records  Yes vs. no    23  H  –      Maintenance of written treatment records  Yes vs. no    24  H  –      Maintenance of written death records  Yes vs. no    25  H  –      Knowledge of diseases experienced on ranch  Yes vs. no    22  H  –      Knowledge of majority of deaths on ranch  Yes vs. no    24  H  –      Use of Beef Quality Assurance training program  Yes vs. no    18  S and H  –  Producer perspectives      Perceived effectiveness of herd health plan  1 = ineffective and 5 = very effective  3.6 (0.7)    S and H  –      Belief that there is room for improvement in the herd health plan  1 = no and 5 = yes  3.2 (1.2)    S and H  +      Reaction time for sick cattle  1 = treat once the schedule allows and 5 = treat immediately  3.1 (1.2)    S and H  –      Empathy toward an animal's pain experience based on the statement “animals experience physical pain as humans do”  1 = strongly disagree and 5 = strongly agree  2.9 (1.6)    S and H  –  Chute-side factors        Processing time per animal  s/cow  79 (49)    B  +      Type of procedure          Pregnancy checks  Yes vs. no    23  B  +          Vaccination  Yes vs. no    21  B  +          Changing ear tags  Yes vs. no    9  B  +      Type of moving aid          Hands  Yes vs. no    20  B  –          Paddles  Yes vs. no    11  B  +          Flags  Yes vs. no    9  B  +          Sticks  Yes vs. no    25  B  +      Presence of dogs  Yes vs. no    15  B  +      Alleyway length  m  10.5 (6.7)    B  –      Alleyway width  cm  74 (7)    B  –      Hydraulic chute3  Yes vs. no    21  B  –      Presence of louvers on restraint  Yes vs. no    11  B  –      Straight alleyway4  Yes vs. no    18  B  +      Funnel fence crowd pen (i.e., A-pen)5  Yes vs. no    20  B  +      Dirt used as floor material in alleyway6  Yes vs. no    21  B  –      Two solid-sided walls in alleyway  Yes vs. no    10  B  –      Dirt used as floor material at restraint exit7  Yes vs. no    13  B  –  Predictor  Category/unit  Mean (SD)  Frequency of yes  Model type1  Hyp2  Ranch characteristics and management      Breeding female herd size  No.  1,051 (1,849)    S, B, and H  –      Ranch age (no. of years of continuous operation)  No.  67 (51)    S, B, and H  –      Farming generations in producer's family  No.  3.7 (2.1)    S, B, and H  –      Cow herd handling events per year  No.  3.4 (5.2)    S, B, and H  –      Visual herd health checks per year  No.  195 (149)    H  –      Visual water access point checks per year  No.  185 (137)    H  –      Visual herd health checks in the previous 2 wk  No.  10.2 (7.0)    H  –      Visual water access point checks in the previous 2 wk  No.  9.7 (5.3)    H  –      Vet visits to the ranch in previous year  No.  5.7 (6.4)    H  –      Maintenance of written sick cattle records  Yes vs. no    23  H  –      Maintenance of written treatment records  Yes vs. no    24  H  –      Maintenance of written death records  Yes vs. no    25  H  –      Knowledge of diseases experienced on ranch  Yes vs. no    22  H  –      Knowledge of majority of deaths on ranch  Yes vs. no    24  H  –      Use of Beef Quality Assurance training program  Yes vs. no    18  S and H  –  Producer perspectives      Perceived effectiveness of herd health plan  1 = ineffective and 5 = very effective  3.6 (0.7)    S and H  –      Belief that there is room for improvement in the herd health plan  1 = no and 5 = yes  3.2 (1.2)    S and H  +      Reaction time for sick cattle  1 = treat once the schedule allows and 5 = treat immediately  3.1 (1.2)    S and H  –      Empathy toward an animal's pain experience based on the statement “animals experience physical pain as humans do”  1 = strongly disagree and 5 = strongly agree  2.9 (1.6)    S and H  –  Chute-side factors        Processing time per animal  s/cow  79 (49)    B  +      Type of procedure          Pregnancy checks  Yes vs. no    23  B  +          Vaccination  Yes vs. no    21  B  +          Changing ear tags  Yes vs. no    9  B  +      Type of moving aid          Hands  Yes vs. no    20  B  –          Paddles  Yes vs. no    11  B  +          Flags  Yes vs. no    9  B  +          Sticks  Yes vs. no    25  B  +      Presence of dogs  Yes vs. no    15  B  +      Alleyway length  m  10.5 (6.7)    B  –      Alleyway width  cm  74 (7)    B  –      Hydraulic chute3  Yes vs. no    21  B  –      Presence of louvers on restraint  Yes vs. no    11  B  –      Straight alleyway4  Yes vs. no    18  B  +      Funnel fence crowd pen (i.e., A-pen)5  Yes vs. no    20  B  +      Dirt used as floor material in alleyway6  Yes vs. no    21  B  –      Two solid-sided walls in alleyway  Yes vs. no    10  B  –      Dirt used as floor material at restraint exit7  Yes vs. no    13  B  –  1Type of model for which the predictor was used for analysis: B = behavior; H = health; S = stockperson handling. 2Hyp = hypothesized direction of effect; if the predictor applies to the ranch, or increases in number, we hypothesize that the prevalence of the outcome of the associated model type will either increase (+) or decrease (-). 3All other chutes were manually operated. 4All other alleyways were curved using partial or half circle or S-shaped designs. 5All other crowd pens used a sweep-tub design. 6All other alleyways concrete flooring. 7All other restraint exits had concrete or rubber flooring. View Large Table 1. Herd-level predictors considered for inclusion in the univariable analysis of variables associated with stockperson handling and cattle health and behavior on 30 California cow–calf operations Predictor  Category/unit  Mean (SD)  Frequency of yes  Model type1  Hyp2  Ranch characteristics and management      Breeding female herd size  No.  1,051 (1,849)    S, B, and H  –      Ranch age (no. of years of continuous operation)  No.  67 (51)    S, B, and H  –      Farming generations in producer's family  No.  3.7 (2.1)    S, B, and H  –      Cow herd handling events per year  No.  3.4 (5.2)    S, B, and H  –      Visual herd health checks per year  No.  195 (149)    H  –      Visual water access point checks per year  No.  185 (137)    H  –      Visual herd health checks in the previous 2 wk  No.  10.2 (7.0)    H  –      Visual water access point checks in the previous 2 wk  No.  9.7 (5.3)    H  –      Vet visits to the ranch in previous year  No.  5.7 (6.4)    H  –      Maintenance of written sick cattle records  Yes vs. no    23  H  –      Maintenance of written treatment records  Yes vs. no    24  H  –      Maintenance of written death records  Yes vs. no    25  H  –      Knowledge of diseases experienced on ranch  Yes vs. no    22  H  –      Knowledge of majority of deaths on ranch  Yes vs. no    24  H  –      Use of Beef Quality Assurance training program  Yes vs. no    18  S and H  –  Producer perspectives      Perceived effectiveness of herd health plan  1 = ineffective and 5 = very effective  3.6 (0.7)    S and H  –      Belief that there is room for improvement in the herd health plan  1 = no and 5 = yes  3.2 (1.2)    S and H  +      Reaction time for sick cattle  1 = treat once the schedule allows and 5 = treat immediately  3.1 (1.2)    S and H  –      Empathy toward an animal's pain experience based on the statement “animals experience physical pain as humans do”  1 = strongly disagree and 5 = strongly agree  2.9 (1.6)    S and H  –  Chute-side factors        Processing time per animal  s/cow  79 (49)    B  +      Type of procedure          Pregnancy checks  Yes vs. no    23  B  +          Vaccination  Yes vs. no    21  B  +          Changing ear tags  Yes vs. no    9  B  +      Type of moving aid          Hands  Yes vs. no    20  B  –          Paddles  Yes vs. no    11  B  +          Flags  Yes vs. no    9  B  +          Sticks  Yes vs. no    25  B  +      Presence of dogs  Yes vs. no    15  B  +      Alleyway length  m  10.5 (6.7)    B  –      Alleyway width  cm  74 (7)    B  –      Hydraulic chute3  Yes vs. no    21  B  –      Presence of louvers on restraint  Yes vs. no    11  B  –      Straight alleyway4  Yes vs. no    18  B  +      Funnel fence crowd pen (i.e., A-pen)5  Yes vs. no    20  B  +      Dirt used as floor material in alleyway6  Yes vs. no    21  B  –      Two solid-sided walls in alleyway  Yes vs. no    10  B  –      Dirt used as floor material at restraint exit7  Yes vs. no    13  B  –  Predictor  Category/unit  Mean (SD)  Frequency of yes  Model type1  Hyp2  Ranch characteristics and management      Breeding female herd size  No.  1,051 (1,849)    S, B, and H  –      Ranch age (no. of years of continuous operation)  No.  67 (51)    S, B, and H  –      Farming generations in producer's family  No.  3.7 (2.1)    S, B, and H  –      Cow herd handling events per year  No.  3.4 (5.2)    S, B, and H  –      Visual herd health checks per year  No.  195 (149)    H  –      Visual water access point checks per year  No.  185 (137)    H  –      Visual herd health checks in the previous 2 wk  No.  10.2 (7.0)    H  –      Visual water access point checks in the previous 2 wk  No.  9.7 (5.3)    H  –      Vet visits to the ranch in previous year  No.  5.7 (6.4)    H  –      Maintenance of written sick cattle records  Yes vs. no    23  H  –      Maintenance of written treatment records  Yes vs. no    24  H  –      Maintenance of written death records  Yes vs. no    25  H  –      Knowledge of diseases experienced on ranch  Yes vs. no    22  H  –      Knowledge of majority of deaths on ranch  Yes vs. no    24  H  –      Use of Beef Quality Assurance training program  Yes vs. no    18  S and H  –  Producer perspectives      Perceived effectiveness of herd health plan  1 = ineffective and 5 = very effective  3.6 (0.7)    S and H  –      Belief that there is room for improvement in the herd health plan  1 = no and 5 = yes  3.2 (1.2)    S and H  +      Reaction time for sick cattle  1 = treat once the schedule allows and 5 = treat immediately  3.1 (1.2)    S and H  –      Empathy toward an animal's pain experience based on the statement “animals experience physical pain as humans do”  1 = strongly disagree and 5 = strongly agree  2.9 (1.6)    S and H  –  Chute-side factors        Processing time per animal  s/cow  79 (49)    B  +      Type of procedure          Pregnancy checks  Yes vs. no    23  B  +          Vaccination  Yes vs. no    21  B  +          Changing ear tags  Yes vs. no    9  B  +      Type of moving aid          Hands  Yes vs. no    20  B  –          Paddles  Yes vs. no    11  B  +          Flags  Yes vs. no    9  B  +          Sticks  Yes vs. no    25  B  +      Presence of dogs  Yes vs. no    15  B  +      Alleyway length  m  10.5 (6.7)    B  –      Alleyway width  cm  74 (7)    B  –      Hydraulic chute3  Yes vs. no    21  B  –      Presence of louvers on restraint  Yes vs. no    11  B  –      Straight alleyway4  Yes vs. no    18  B  +      Funnel fence crowd pen (i.e., A-pen)5  Yes vs. no    20  B  +      Dirt used as floor material in alleyway6  Yes vs. no    21  B  –      Two solid-sided walls in alleyway  Yes vs. no    10  B  –      Dirt used as floor material at restraint exit7  Yes vs. no    13  B  –  1Type of model for which the predictor was used for analysis: B = behavior; H = health; S = stockperson handling. 2Hyp = hypothesized direction of effect; if the predictor applies to the ranch, or increases in number, we hypothesize that the prevalence of the outcome of the associated model type will either increase (+) or decrease (-). 3All other chutes were manually operated. 4All other alleyways were curved using partial or half circle or S-shaped designs. 5All other crowd pens used a sweep-tub design. 6All other alleyways concrete flooring. 7All other restraint exits had concrete or rubber flooring. View Large Herd-Level Measures Ranch Characteristics and Management. Information about the ranch and its management practices was collected through a questionnaire-guided interview with the producer (i.e., a farm manager) at each ranch. Questions inquired about the overall breeding cow herd size; age of the ranch (i.e., years of continuous operation); number of farming generations in the producer's family; the number of times the cow herd was processed in the chute each year; the number of times the herd and their water access points (e.g., water troughs, streams, ponds) were visually observed in the past year and the previous 2 wk; the average number of yearly veterinarian ranch visits; maintenance of written records for sick cattle, treatments, and death losses; knowledge of diseases experienced on the ranch (assessed based on if the producer was able to give an estimate of the number of cases of each disease experienced in the previous year); knowledge of death losses (assessed based on if the producer said that he or she knew the cause of death for over 50% of mortalities); and the types of stockperson training programs used. For the number of times the herd and their water access points were observed in the past year, we calculated a total for all the sites the herd grazed. For example, 1 ranch pastured their cattle on 1 location year-round and observed them 1 time/wk, totaling 52 checks/yr. Another ranch kept their cattle at location A for 5 mo, observing them 2 times/wk, and location B for 7 mo, observing them 1.5 times/mo, totaling a similar 55 checks/yr. Producer Perspectives. The interview also contained questions that aimed to assess the producer's perspectives toward their cattle on a 5-point Likert scale, where they were asked to rate their response on a spectrum between disagreement (i.e., 1) and agreement (i.e., 5), with the middle value (i.e., 3) denoting neutrality. Producers were asked to rate the perceived effectiveness of their health plan (scale: 1 = not effective and 5 = very effective) and if they thought there was room for improvement (scale: 1 = no and 5 = yes). Producers were also asked how quickly they react when an animal starts to get sick (scale: 1 = treat once the schedule allows and 5 = treat immediately) and, as a measure of empathy, how they felt about the statement “animals experience physical pain as humans do” (scale: 1 = strongly disagree and 5 = strongly agree; adopted from Kielland et al., 2009). Chute-Side Factors. The type and duration of the procedures performed were recorded along with the types of moving aids used and whether the cattle could visually or audibly detect the presence of dogs. Design features of the restraint and alleyway used during the processing event were recorded. Restraint features included the type used (hydraulic vs. manual), presence of louvers, and type of flooring surface directly in front of the restraint. Alleyway features included the shape (straight vs. curved), presence of 2 fully solid-sided walls, width (measured at the height of a randomly selected reference cow's withers), length (distance between the restraint tailgate and crowd pen that allowed for only a single-file line of cattle), floor material, and type of crowd pen (funnel fence vs. sweep tub). Cow-Level Measures For each herd, assessors began evaluating the first animal sent through the chute and continued observing until the at least the minimum sample size was met. A minimum sample size was observed following the recommendations of Welfare Quality on farm cattle assessment protocols (Welfare Quality, 2009). For example, all cattle were observed in herds with 30 or fewer animals and a minimum of 95 cattle were observed in herds with 10,000 animals. Two assessors stood 2 to 4 m from the chute, at the tailgate and head gate on the same side, to observe cattle as they entered and exited the chute. We began observing each focal cow or heifer (n = 3,065) once the preceding animal was fully caught in the head gate. For the first animal, or in instances where the focal cow was not yet in the alleyway when the preceding animal was caught, observation began when all 4 of the focal animal's hooves entered the alleyway. Observation ended when the focal cow exited the restraint, and this window was called the “individual animal observation period.” It was used for all measures unless explicitly stated otherwise. In addition to the following measures, we also examined cattle for broken tails and bloated rumens, but these outcomes were rare (0 and 1, respectively) and excluded from further analysis. Cattle Health. Body condition was assessed on a 3-point scale, where 1 = thin, 2 = acceptable, and 3 = obese. Cattle with thin body condition were classified as those with 3 or more ribs being visible, a sharp and prominent tail head and backbone, a tight brisket, and little to no muscling apparent in hindquarters. Obese cattle were those that appeared full and blocky with no distinct bone structure, protruding fat deposits on tail head and pin bones, and a distended brisket. All other animals were considered to have acceptable body condition. Only 2 ranches had some prevalence of obese cattle; therefore, this outcome was not used in further analysis. Locomotion scores (LS) were also collected on a 3-point scale, where 1 = acceptable, 2 = moderately lame, and 3 = severely lame. Cattle with a LS of 2 were those with an easily identifiable affected limb or limbs, that favored 1 or more limbs while standing, and that had shortened strides and a slight limp while walking. Cattle with a LS of 3 were those that were barely able or unable to place weight on 1 or more limbs, walked slowly, and made frequent stops. We observed only 2 cows with a LS of 3 across all 30 ranches. For the analysis, we combined the prevalence of all lame cattle. Cattle were assessed for abrasions, hairless patches, swelling, blind eyes, and dirtiness by examining 1 side of the body, including the head, neck, side, and rump; the outer side of the closest legs; and the inner side of the furthest legs. A cow or heifer was considered to have an abrasion if she had at least 1 area on the body ≥ 2 cm with an open wound or scab that was not a result of branding. Hairless patches were recorded if there was at least 1 area on the body ≥ 2 cm in any dimension where hair was missing for reasons other than brand marks, hairless cysts, or polyps. Swelling was recorded if there was inflammation with a diam. ≥ 2 cm under the skin causing it to appear inflated, bulbous, and tight. An eye that was missing or opaque, discolored, or had an inflamed lens as distinct from the sclera, possibly accompanied by squinting or ocular discharge, was recorded as a blind eye. An assessor confirmed blindness by approaching the animal to evaluate her lack of response to a hand opening and closing 10 cm away from the affected eye. Cattle were recorded as dirty if there were dried, nondiarrheal plaques (a 3-dimensional layer of mud or feces) on the side above the hock and knee measuring at least 2 hands in total. Cattle Behavior in the Chute. A cows or heifer was considered to balk when the route was clear in front of or behind her but she refused to move forward or backward within 4 s after being touched by a moving aid or electric prod. Vocalizations (i.e., any audible tone) by the cow or heifer were recorded only after she was restrained but before the procedure began. Stumbles were recorded when the cow's knee or hock came into contact with the ground. If there was contact between her torso and the ground, it was considered a fall. Stumbles and falls were recorded both in the chute and at exit (scoring for the latter began when the head gate was opened to release the cow and typically ended 3 to 6 strides later). Cows or heifers that took at least 2 strides as a gait faster than a trot once all 4 hooves touched the ground outside of the restraint on exiting were recorded as a run. Exit behavior and LS were assessed during exit of the restraint, and the distance animals were observed varied from 10 to 25 m. Stockperson Handling. Electric prod use was recorded when a prod with an activated charge touched the cow or heifer. Tail twists were recorded when a stockperson grasped the cow's tail in their hand and turned it to either side. Mis-catches were recorded whenever a cow was improperly caught on the first attempt. A proper catch was when a single animal was loaded into the restraint with the head gate closed behind her ears and in front of the shoulders, all 4 hooves touched the floor inside the chute, and the subsequent animal was not caught in the tailgate. Statistical Analysis Data Management. One producer did not respond to the question “how do you feel about the statement ‘animals experience physical pain as humans dO'?” and was assigned a neutral response (i.e., 3) to avoid missing values in the analysis. For the questions regarding the number of times the veterinarian regularly visited the ranch each year, 2 producers were veterinarians themselves and therefore were assigned the greatest value for this measure (i.e., 30). Three producers called their veterinarian only on an as-needed basis and were assigned a value of 0. Three types of floor material were used at the restraint exit (dirt, rubber, and concrete), but the analysis was based on whether or not dirt was used because differences in surface texture (smooth vs. rough rubber or concrete) were not assessed. Training programs were analyzed on the basis of whether or not the Beef Quality Assurance (BQA) program (a resource that offers guidelines for cattle handling, husbandry techniques, management, facility maintenance, and record keeping; BQA, 2010) was used because it was the only common program among ranches. Ordinal (i.e., Likert scale) predictors were analyzed as continuous variables. A total of 4 assessors recorded animal-based measures over the course of the study. Interobserver reliability for each pair of assessors was determined by the kappa coefficient of concordance (Hollenbeck, 1978; Martin and Bateson, 1993). Logistic Regression. In this cross-sectional study, 2 types of analyses were performed. The first considered the experimental unit to be the herd, using herd-level predictors for cattle health and behavior and stockperson handling outcomes (Table 1). The second analysis considered the experimental unit to be the cow, using cow-level stockperson handling predictors for cattle behavior outcomes, with the exception of runs out of the restraint, which also used previous cow behavior outcomes as predictors. Predictors were initially considered for model type in a hypothesis-driven manner. The direction of hypothesized predictor effect on the outcomes is provided in Table 1. Variables (predictors and outcomes) with at least 5 ranches per category were considered for analysis. For both types of analysis, univariable association between the outcomes and predictors were assessed using SAS software, version 9.4 (GLIMMIX, with options DIST = binomial; SAS Inst. Inc., Cary, NC). Predictors with a univariable association of P ≤ 0.2 and having biological relevance, as described in Table 1, were evaluated for use in the multivariable regression models. Qualified predictors were assessed for collinearity to avoid submitting highly correlated variables (|r| ≥ 0.70) to the same model, for which none were removed. Finally, manual backward elimination was used to build multivariable models, which included the ranch as a random effect (using the RANDOM statement, with the TYPE = vc option) for the cow-level analysis. Predictors were removed from the final model if P > 0.05. Variance inflation factors were calculated after each model to confirm lack of multicollinearity. RESULTS The kappa coefficient of concordance for each pair of assessors was ≥ 0.8 for all variables. Ranches varied in average herd size (mean 1,051 [SD 1,849]; range 28 to 10,000 cows and heifers) and location throughout California.Table 1 describes the range of management practices, facility designs, and producer perspectives observed. Table 2 provides outcome information, for which there was variation in stockperson handling and cattle behavior but cattle health problems were rare. Three thousand sixty-five cattle were observed across the 30 ranches, averaging 102 (SD 40) observations/ranch (range 28 to 182 observations/ranch). Table 2. Cow-level health and behavior outcomes as a percentage of the herd in which they were observed on 30 California cow–calf operations Outcome  Mean  SD  Range  Cattle health      Thin  2.3  5.5  0 to 29.4      Lame  1.3  1.5  0 to 7.1      Swelling  2.0  2.6  0 to 11.9      Hairless patch  9.1  8.4  1.1 to 46.0      Abrasion  0.6  0.9  0 to 2.8      Blind eye  0.7  1.2  0 to 4.6      Dirty  1.5  6.3  0 to 34.5  Stockperson handling      Electric prod use  23.5  21.5  0 to 73.0      Moving aid use  67.8  22.3  16.6 to 100.0      Tail twist  7.5  16.4  0 to 71.4      Mis-catch  14.5  14.1  0 to 49.5  Cattle behavior      Balk  22.0  21.9  1.6 to 83.3      Vocalize  5.2  5.1  0 to 20.1      Stumble in the chute  12.3  10.2  0 to 44.3      Fall in the chute  1.2  1.6  0 to 6.3      Stumble on exit  4.7  4.6  0 to 17.9      Fall on exit  0.9  1.4  0 to 5.2      Run  12.8  13.3  0 to 60.4  Outcome  Mean  SD  Range  Cattle health      Thin  2.3  5.5  0 to 29.4      Lame  1.3  1.5  0 to 7.1      Swelling  2.0  2.6  0 to 11.9      Hairless patch  9.1  8.4  1.1 to 46.0      Abrasion  0.6  0.9  0 to 2.8      Blind eye  0.7  1.2  0 to 4.6      Dirty  1.5  6.3  0 to 34.5  Stockperson handling      Electric prod use  23.5  21.5  0 to 73.0      Moving aid use  67.8  22.3  16.6 to 100.0      Tail twist  7.5  16.4  0 to 71.4      Mis-catch  14.5  14.1  0 to 49.5  Cattle behavior      Balk  22.0  21.9  1.6 to 83.3      Vocalize  5.2  5.1  0 to 20.1      Stumble in the chute  12.3  10.2  0 to 44.3      Fall in the chute  1.2  1.6  0 to 6.3      Stumble on exit  4.7  4.6  0 to 17.9      Fall on exit  0.9  1.4  0 to 5.2      Run  12.8  13.3  0 to 60.4  View Large Table 2. Cow-level health and behavior outcomes as a percentage of the herd in which they were observed on 30 California cow–calf operations Outcome  Mean  SD  Range  Cattle health      Thin  2.3  5.5  0 to 29.4      Lame  1.3  1.5  0 to 7.1      Swelling  2.0  2.6  0 to 11.9      Hairless patch  9.1  8.4  1.1 to 46.0      Abrasion  0.6  0.9  0 to 2.8      Blind eye  0.7  1.2  0 to 4.6      Dirty  1.5  6.3  0 to 34.5  Stockperson handling      Electric prod use  23.5  21.5  0 to 73.0      Moving aid use  67.8  22.3  16.6 to 100.0      Tail twist  7.5  16.4  0 to 71.4      Mis-catch  14.5  14.1  0 to 49.5  Cattle behavior      Balk  22.0  21.9  1.6 to 83.3      Vocalize  5.2  5.1  0 to 20.1      Stumble in the chute  12.3  10.2  0 to 44.3      Fall in the chute  1.2  1.6  0 to 6.3      Stumble on exit  4.7  4.6  0 to 17.9      Fall on exit  0.9  1.4  0 to 5.2      Run  12.8  13.3  0 to 60.4  Outcome  Mean  SD  Range  Cattle health      Thin  2.3  5.5  0 to 29.4      Lame  1.3  1.5  0 to 7.1      Swelling  2.0  2.6  0 to 11.9      Hairless patch  9.1  8.4  1.1 to 46.0      Abrasion  0.6  0.9  0 to 2.8      Blind eye  0.7  1.2  0 to 4.6      Dirty  1.5  6.3  0 to 34.5  Stockperson handling      Electric prod use  23.5  21.5  0 to 73.0      Moving aid use  67.8  22.3  16.6 to 100.0      Tail twist  7.5  16.4  0 to 71.4      Mis-catch  14.5  14.1  0 to 49.5  Cattle behavior      Balk  22.0  21.9  1.6 to 83.3      Vocalize  5.2  5.1  0 to 20.1      Stumble in the chute  12.3  10.2  0 to 44.3      Fall in the chute  1.2  1.6  0 to 6.3      Stumble on exit  4.7  4.6  0 to 17.9      Fall on exit  0.9  1.4  0 to 5.2      Run  12.8  13.3  0 to 60.4  View Large Cattle Health Cattle health outcomes were correlated with ranch characteristics, management, and producer perspectives (Table 3). The risk of thin cattle increased by 17% for every 10 yr increase in ranch age and by 29% for each additional water access point check in the previous 2 wk (P < 0.001). A 1-unit increase in producer empathy toward an animal's pain experience increased the risk of lameness by 31% (P = 0.024) but decreased the risk of swelling by 22% (P = 0.013). A 1-unit increase in producer belief that there was room for improvement in their herd health management plan was correlated with a greater risk of swelling (odds ratio = 1.4, P = 0.026) and hairless patches (odds ratio = 1.3, P = 0.001). The risk of hairless patches decreased by 12 (P = 0.008) and 5% (P < 0.001) for every additional family farming generation and veterinarian visit per year, respectively. The risk of hairless patches increased by 3% (P < 0.001) for every 10 additional water access point checks per year. Abrasion risk decreased by 6% (P = 0.017) for every 10 additional herd health checks performed per year. The risk of dirty cattle increased by 31% (P < 0.001) for each additional herd handling event per year and decreased by 13 (P = 0.049) and 78% (P = 0.038) for each additional herd health check performed in the previous 2 wk and with use of the BQA training program, respectively. Blind eyes were at 37% lower risk with every additional family farming generation (P = 0.019) but at a 127% greater risk with every 1-unit increase in producer reaction time to sick cattle (P = 0.022). Table 3. Multivariable associations of cattle health outcomes with herd-level predictors on 30 California cow–calf operations Variable  Estimate  SE  Odds ratio1  95% CI2  P-value  Thin      Older ranch age (per 10 yr)  0.015  0.002  1.166  1.111 to 1.223  <0.001      More water access point checks in the previous 2 wk  0.251  0.047  1.285  1.167 to 1.415  <0.001  Lame      More empathy toward animal's pain experience  0.267  0.112  1.306  1.039 to 1.641  0.024  Swelling      Greater belief that there is room for improvement in the herd health plan  0.304  0.129  1.355  1.039 to 1.767  0.026      More empathy toward animal's pain experience  −0.243  0.092  0.785  0.650 to 0.947  0.013  Hairless patch      More water access point checks per year (per 10 checks)  0.003  0.001  1.028  1.017 to 1.039  <0.001      More farming generations in producer's family  −0.131  0.045  0.877  0.799 to 0.964  0.008      More vet visits in the past year  −0.054  0.009  0.948  0.929 to 0.966  <0.001      Greater belief that there is room for improvement in the herd health plan  0.240  0.065  1.272  1.112 to 1.454  0.001  Abrasion      More herd health checks per year (per 10 checks)  −0.006  0.002  0.944  0.901 to 0.989  0.017  Blind eye      More farming generations in producer's family  −0.467  0.187  0.627  0.427 to 0.921  0.019      Faster reaction time for sick cattle  0.818  0.337  2.266  1.136 to 4.519  0.022  Dirty      More cow herd handling events per year  0.267  0.058  1.306  1.159 to 1.471  <0.001      More herd health checks in the previous 2 wk  −0.141  0.068  0.869  0.755 to 0.999  0.049      Use of Beef Quality Assurance training program (yes vs. no)  −1.522  0.695  0.218  0.052 to 0.911  0.038  Variable  Estimate  SE  Odds ratio1  95% CI2  P-value  Thin      Older ranch age (per 10 yr)  0.015  0.002  1.166  1.111 to 1.223  <0.001      More water access point checks in the previous 2 wk  0.251  0.047  1.285  1.167 to 1.415  <0.001  Lame      More empathy toward animal's pain experience  0.267  0.112  1.306  1.039 to 1.641  0.024  Swelling      Greater belief that there is room for improvement in the herd health plan  0.304  0.129  1.355  1.039 to 1.767  0.026      More empathy toward animal's pain experience  −0.243  0.092  0.785  0.650 to 0.947  0.013  Hairless patch      More water access point checks per year (per 10 checks)  0.003  0.001  1.028  1.017 to 1.039  <0.001      More farming generations in producer's family  −0.131  0.045  0.877  0.799 to 0.964  0.008      More vet visits in the past year  −0.054  0.009  0.948  0.929 to 0.966  <0.001      Greater belief that there is room for improvement in the herd health plan  0.240  0.065  1.272  1.112 to 1.454  0.001  Abrasion      More herd health checks per year (per 10 checks)  −0.006  0.002  0.944  0.901 to 0.989  0.017  Blind eye      More farming generations in producer's family  −0.467  0.187  0.627  0.427 to 0.921  0.019      Faster reaction time for sick cattle  0.818  0.337  2.266  1.136 to 4.519  0.022  Dirty      More cow herd handling events per year  0.267  0.058  1.306  1.159 to 1.471  <0.001      More herd health checks in the previous 2 wk  −0.141  0.068  0.869  0.755 to 0.999  0.049      Use of Beef Quality Assurance training program (yes vs. no)  −1.522  0.695  0.218  0.052 to 0.911  0.038  1Per 1-unit increase, unless otherwise noted. 2CI = confidence interval. View Large Table 3. Multivariable associations of cattle health outcomes with herd-level predictors on 30 California cow–calf operations Variable  Estimate  SE  Odds ratio1  95% CI2  P-value  Thin      Older ranch age (per 10 yr)  0.015  0.002  1.166  1.111 to 1.223  <0.001      More water access point checks in the previous 2 wk  0.251  0.047  1.285  1.167 to 1.415  <0.001  Lame      More empathy toward animal's pain experience  0.267  0.112  1.306  1.039 to 1.641  0.024  Swelling      Greater belief that there is room for improvement in the herd health plan  0.304  0.129  1.355  1.039 to 1.767  0.026      More empathy toward animal's pain experience  −0.243  0.092  0.785  0.650 to 0.947  0.013  Hairless patch      More water access point checks per year (per 10 checks)  0.003  0.001  1.028  1.017 to 1.039  <0.001      More farming generations in producer's family  −0.131  0.045  0.877  0.799 to 0.964  0.008      More vet visits in the past year  −0.054  0.009  0.948  0.929 to 0.966  <0.001      Greater belief that there is room for improvement in the herd health plan  0.240  0.065  1.272  1.112 to 1.454  0.001  Abrasion      More herd health checks per year (per 10 checks)  −0.006  0.002  0.944  0.901 to 0.989  0.017  Blind eye      More farming generations in producer's family  −0.467  0.187  0.627  0.427 to 0.921  0.019      Faster reaction time for sick cattle  0.818  0.337  2.266  1.136 to 4.519  0.022  Dirty      More cow herd handling events per year  0.267  0.058  1.306  1.159 to 1.471  <0.001      More herd health checks in the previous 2 wk  −0.141  0.068  0.869  0.755 to 0.999  0.049      Use of Beef Quality Assurance training program (yes vs. no)  −1.522  0.695  0.218  0.052 to 0.911  0.038  Variable  Estimate  SE  Odds ratio1  95% CI2  P-value  Thin      Older ranch age (per 10 yr)  0.015  0.002  1.166  1.111 to 1.223  <0.001      More water access point checks in the previous 2 wk  0.251  0.047  1.285  1.167 to 1.415  <0.001  Lame      More empathy toward animal's pain experience  0.267  0.112  1.306  1.039 to 1.641  0.024  Swelling      Greater belief that there is room for improvement in the herd health plan  0.304  0.129  1.355  1.039 to 1.767  0.026      More empathy toward animal's pain experience  −0.243  0.092  0.785  0.650 to 0.947  0.013  Hairless patch      More water access point checks per year (per 10 checks)  0.003  0.001  1.028  1.017 to 1.039  <0.001      More farming generations in producer's family  −0.131  0.045  0.877  0.799 to 0.964  0.008      More vet visits in the past year  −0.054  0.009  0.948  0.929 to 0.966  <0.001      Greater belief that there is room for improvement in the herd health plan  0.240  0.065  1.272  1.112 to 1.454  0.001  Abrasion      More herd health checks per year (per 10 checks)  −0.006  0.002  0.944  0.901 to 0.989  0.017  Blind eye      More farming generations in producer's family  −0.467  0.187  0.627  0.427 to 0.921  0.019      Faster reaction time for sick cattle  0.818  0.337  2.266  1.136 to 4.519  0.022  Dirty      More cow herd handling events per year  0.267  0.058  1.306  1.159 to 1.471  <0.001      More herd health checks in the previous 2 wk  −0.141  0.068  0.869  0.755 to 0.999  0.049      Use of Beef Quality Assurance training program (yes vs. no)  −1.522  0.695  0.218  0.052 to 0.911  0.038  1Per 1-unit increase, unless otherwise noted. 2CI = confidence interval. View Large Stockperson Handling Ranch characteristics and use of the BQA training program were associated with stockpersons handling (Table 4). For each animal handled, the risk of using electric prods increased by 1% for every 100 more breeding females in the herd (P < 0.001) but decreased by 5% for every 10 yr increase in ranch age (P = 0.001) and by 16% for every additional family farming generation (P < 0.001). Each additional farming generation was also associated with a 20% reduced risk of touching cattle with a moving aid (P < 0.001). The risk of mis-catching cattle decreased by 45% on ranches that used the BQA training program (P < 0.001). None of the herd-level predictors were correlated with tail twisting at the P < 0.2 significance level in the initial screening, so no multivariable model was constructed. Table 4. Multivariable associations of stockperson handling outcomes with herd-level predictors on 30 California cow–calf operations Variable  Estimate  SE  Odds ratio1  95% CI2  P-value  Electric prod use      Larger breeding female herd size (per 100 animals)  <0.001  <0.001  1.011  1.007 to 1.015  <0.001      Older ranch age (per 10 yr)  −0.005  0.001  0.952  0.926 to 0.979  0.001      More farming generations in producer's family  −0.168  0.036  0.845  0.785 to 0.910  <0.001  Moving aid use      More farming generations in producer's family  −0.218  0.023  0.805  0.767 to 0.843  <0.001  Mis-catch      Use of Beef Quality Assurance training program (yes vs. no)  −0.602  0.099  0.548  0.447 to 0.671  <0.001      Tail twist3  –  –  –  –  –  Variable  Estimate  SE  Odds ratio1  95% CI2  P-value  Electric prod use      Larger breeding female herd size (per 100 animals)  <0.001  <0.001  1.011  1.007 to 1.015  <0.001      Older ranch age (per 10 yr)  −0.005  0.001  0.952  0.926 to 0.979  0.001      More farming generations in producer's family  −0.168  0.036  0.845  0.785 to 0.910  <0.001  Moving aid use      More farming generations in producer's family  −0.218  0.023  0.805  0.767 to 0.843  <0.001  Mis-catch      Use of Beef Quality Assurance training program (yes vs. no)  −0.602  0.099  0.548  0.447 to 0.671  <0.001      Tail twist3  –  –  –  –  –  1Per 1-unit increase, unless otherwise noted. 2CI = confidence interval. 3A multivariable model was not constructed because no predictors were correlated at the P < 0.2 significance level in the initial screening. View Large Table 4. Multivariable associations of stockperson handling outcomes with herd-level predictors on 30 California cow–calf operations Variable  Estimate  SE  Odds ratio1  95% CI2  P-value  Electric prod use      Larger breeding female herd size (per 100 animals)  <0.001  <0.001  1.011  1.007 to 1.015  <0.001      Older ranch age (per 10 yr)  −0.005  0.001  0.952  0.926 to 0.979  0.001      More farming generations in producer's family  −0.168  0.036  0.845  0.785 to 0.910  <0.001  Moving aid use      More farming generations in producer's family  −0.218  0.023  0.805  0.767 to 0.843  <0.001  Mis-catch      Use of Beef Quality Assurance training program (yes vs. no)  −0.602  0.099  0.548  0.447 to 0.671  <0.001      Tail twist3  –  –  –  –  –  Variable  Estimate  SE  Odds ratio1  95% CI2  P-value  Electric prod use      Larger breeding female herd size (per 100 animals)  <0.001  <0.001  1.011  1.007 to 1.015  <0.001      Older ranch age (per 10 yr)  −0.005  0.001  0.952  0.926 to 0.979  0.001      More farming generations in producer's family  −0.168  0.036  0.845  0.785 to 0.910  <0.001  Moving aid use      More farming generations in producer's family  −0.218  0.023  0.805  0.767 to 0.843  <0.001  Mis-catch      Use of Beef Quality Assurance training program (yes vs. no)  −0.602  0.099  0.548  0.447 to 0.671  <0.001      Tail twist3  –  –  –  –  –  1Per 1-unit increase, unless otherwise noted. 2CI = confidence interval. 3A multivariable model was not constructed because no predictors were correlated at the P < 0.2 significance level in the initial screening. View Large Cattle Behavior in the Chute At the herd level, cattle behavior was correlated with ranch characteristics, management, and chute-side factors (Table 5). For every 10 s increase in processing time per animal during the handling event, the risk of stumbles and runs at exit decreased 10 and 7% (P ≤ 0.006), respectively, for each animal handled but balking risk increased by 11% (P < 0.001). When a hydraulic chute was used, the risk of balking and falls in the chute decreased 56 and 68% (P ≤ 0.004), respectively, but risk of vocalization increased by 166% (P = 0.005). The risk of balking decreased by 71% when the ranch used hands as a moving aid (P < 0.001) and risk of vocalization increased by 3% for every 1 m increase in alleyway length (P = 0.001) and more than doubled with the use of louvers (P = 0.001). The risk of vocalizations decreased by 42% for every additional handling event per year (P < 0.001), which also decreased stumbles in the chute and at exit by 22 and 31% (P ≤ 0.001), respectively. The risk of cattle balking, stumbling and falling on exit, and running out of the restraint decreased by 3, 4, 10, and 4%, respectively, for every 1 cm increase in alleyway width (P ≤ 0.001). For every 100 more breeding females in the herd, the risk of both stumbles and falls at exit increased by about 1% (P ≤ 0.040). Finally, the risk of cattle running out of the restraint increased 121% when a straight alleyway was used (P < 0.001) but decreased 29% when hands were used as moving aids (P = 0.006). Table 5. Multivariable associations of cattle behavioral outcomes in the chute with herd-level predictors on 30 California cow–calf operations Variable  Estimate  SE  Odds ratio1  95% CI2  P-value  Balk      Longer processing time per animal (per 10 s)  0.010  0.002  1.111  1.075 to 1.147  <0.001      Wider alleyway (cm)  −0.032  0.006  0.969  0.956 to 0.982  <0.001      Use of hydraulic chute (yes vs. no)  −0.827  0.144  0.437  0.325 to 0.589  <0.001      Use of hands as a moving aid (yes vs. no)  −1.227  0.113  0.293  0.232 to 0.370  <0.001  Vocalize      More cow herd handling events per year  −0.537  0.089  0.585  0.486 to 0.703  <0.001      Longer alleyway, m  0.033  0.008  1.034  1.016 to 1.052  0.001      Use of hydraulic chute (yes vs. no)  0.977  0.320  2.656  1.373 to 5.138  0.005      Louvers present on restraint (yes vs. no)  0.727  0.190  2.069  1.400 to 3.058  0.001  Stumble in chute      More cow herd handling events per year  −0.254  0.053  0.776  0.696 to 0.865  <0.001  Fall in chute      Use of hydraulic chute (yes vs. no)  −1.125  0.355  0.325  0.157 to 0.672  0.004  Stumble at exit      Larger breeding female herd size (per 100 animals)  <0.001  <0.001  1.007  1.000 to 1.013  0.040      More cow herd handling events per year  −0.373  0.103  0.689  0.557 to 0.852  0.001      Longer processing time per animal (per 10 s)  −0.010  0.003  0.904  0.845 to 0.969  0.006      Wider alleyway (cm)  −0.043  0.011  0.958  0.936 to 0.981  0.001  Fall at exit      Larger breeding female herd size (per 100 animal increase)  <0.001  <0.001  1.011  1.001 to 1.021  0.038      Wider alleyway (cm)  −0.105  0.023  0.900  0.858 to 0.944  <0.001  Run out of restraint      Longer processing time per animal (per 10 s)  −0.007  0.002  0.932  0.899 to 0.967  0.001      Wider alleyway (cm)  −0.045  0.007  0.956  0.943 to 0.970  <0.001      Use of a straight alleyway (yes vs. no)  0.795  0.137  2.214  1.668 to 2.938  <0.001      Use of hands as a moving aid (yes vs. no)  −0.342  0.113  0.710  0.563 to 0.897  0.006  Variable  Estimate  SE  Odds ratio1  95% CI2  P-value  Balk      Longer processing time per animal (per 10 s)  0.010  0.002  1.111  1.075 to 1.147  <0.001      Wider alleyway (cm)  −0.032  0.006  0.969  0.956 to 0.982  <0.001      Use of hydraulic chute (yes vs. no)  −0.827  0.144  0.437  0.325 to 0.589  <0.001      Use of hands as a moving aid (yes vs. no)  −1.227  0.113  0.293  0.232 to 0.370  <0.001  Vocalize      More cow herd handling events per year  −0.537  0.089  0.585  0.486 to 0.703  <0.001      Longer alleyway, m  0.033  0.008  1.034  1.016 to 1.052  0.001      Use of hydraulic chute (yes vs. no)  0.977  0.320  2.656  1.373 to 5.138  0.005      Louvers present on restraint (yes vs. no)  0.727  0.190  2.069  1.400 to 3.058  0.001  Stumble in chute      More cow herd handling events per year  −0.254  0.053  0.776  0.696 to 0.865  <0.001  Fall in chute      Use of hydraulic chute (yes vs. no)  −1.125  0.355  0.325  0.157 to 0.672  0.004  Stumble at exit      Larger breeding female herd size (per 100 animals)  <0.001  <0.001  1.007  1.000 to 1.013  0.040      More cow herd handling events per year  −0.373  0.103  0.689  0.557 to 0.852  0.001      Longer processing time per animal (per 10 s)  −0.010  0.003  0.904  0.845 to 0.969  0.006      Wider alleyway (cm)  −0.043  0.011  0.958  0.936 to 0.981  0.001  Fall at exit      Larger breeding female herd size (per 100 animal increase)  <0.001  <0.001  1.011  1.001 to 1.021  0.038      Wider alleyway (cm)  −0.105  0.023  0.900  0.858 to 0.944  <0.001  Run out of restraint      Longer processing time per animal (per 10 s)  −0.007  0.002  0.932  0.899 to 0.967  0.001      Wider alleyway (cm)  −0.045  0.007  0.956  0.943 to 0.970  <0.001      Use of a straight alleyway (yes vs. no)  0.795  0.137  2.214  1.668 to 2.938  <0.001      Use of hands as a moving aid (yes vs. no)  −0.342  0.113  0.710  0.563 to 0.897  0.006  1Per 1-unit increase, unless otherwise noted. 2CI = confidence interval. View Large Table 5. Multivariable associations of cattle behavioral outcomes in the chute with herd-level predictors on 30 California cow–calf operations Variable  Estimate  SE  Odds ratio1  95% CI2  P-value  Balk      Longer processing time per animal (per 10 s)  0.010  0.002  1.111  1.075 to 1.147  <0.001      Wider alleyway (cm)  −0.032  0.006  0.969  0.956 to 0.982  <0.001      Use of hydraulic chute (yes vs. no)  −0.827  0.144  0.437  0.325 to 0.589  <0.001      Use of hands as a moving aid (yes vs. no)  −1.227  0.113  0.293  0.232 to 0.370  <0.001  Vocalize      More cow herd handling events per year  −0.537  0.089  0.585  0.486 to 0.703  <0.001      Longer alleyway, m  0.033  0.008  1.034  1.016 to 1.052  0.001      Use of hydraulic chute (yes vs. no)  0.977  0.320  2.656  1.373 to 5.138  0.005      Louvers present on restraint (yes vs. no)  0.727  0.190  2.069  1.400 to 3.058  0.001  Stumble in chute      More cow herd handling events per year  −0.254  0.053  0.776  0.696 to 0.865  <0.001  Fall in chute      Use of hydraulic chute (yes vs. no)  −1.125  0.355  0.325  0.157 to 0.672  0.004  Stumble at exit      Larger breeding female herd size (per 100 animals)  <0.001  <0.001  1.007  1.000 to 1.013  0.040      More cow herd handling events per year  −0.373  0.103  0.689  0.557 to 0.852  0.001      Longer processing time per animal (per 10 s)  −0.010  0.003  0.904  0.845 to 0.969  0.006      Wider alleyway (cm)  −0.043  0.011  0.958  0.936 to 0.981  0.001  Fall at exit      Larger breeding female herd size (per 100 animal increase)  <0.001  <0.001  1.011  1.001 to 1.021  0.038      Wider alleyway (cm)  −0.105  0.023  0.900  0.858 to 0.944  <0.001  Run out of restraint      Longer processing time per animal (per 10 s)  −0.007  0.002  0.932  0.899 to 0.967  0.001      Wider alleyway (cm)  −0.045  0.007  0.956  0.943 to 0.970  <0.001      Use of a straight alleyway (yes vs. no)  0.795  0.137  2.214  1.668 to 2.938  <0.001      Use of hands as a moving aid (yes vs. no)  −0.342  0.113  0.710  0.563 to 0.897  0.006  Variable  Estimate  SE  Odds ratio1  95% CI2  P-value  Balk      Longer processing time per animal (per 10 s)  0.010  0.002  1.111  1.075 to 1.147  <0.001      Wider alleyway (cm)  −0.032  0.006  0.969  0.956 to 0.982  <0.001      Use of hydraulic chute (yes vs. no)  −0.827  0.144  0.437  0.325 to 0.589  <0.001      Use of hands as a moving aid (yes vs. no)  −1.227  0.113  0.293  0.232 to 0.370  <0.001  Vocalize      More cow herd handling events per year  −0.537  0.089  0.585  0.486 to 0.703  <0.001      Longer alleyway, m  0.033  0.008  1.034  1.016 to 1.052  0.001      Use of hydraulic chute (yes vs. no)  0.977  0.320  2.656  1.373 to 5.138  0.005      Louvers present on restraint (yes vs. no)  0.727  0.190  2.069  1.400 to 3.058  0.001  Stumble in chute      More cow herd handling events per year  −0.254  0.053  0.776  0.696 to 0.865  <0.001  Fall in chute      Use of hydraulic chute (yes vs. no)  −1.125  0.355  0.325  0.157 to 0.672  0.004  Stumble at exit      Larger breeding female herd size (per 100 animals)  <0.001  <0.001  1.007  1.000 to 1.013  0.040      More cow herd handling events per year  −0.373  0.103  0.689  0.557 to 0.852  0.001      Longer processing time per animal (per 10 s)  −0.010  0.003  0.904  0.845 to 0.969  0.006      Wider alleyway (cm)  −0.043  0.011  0.958  0.936 to 0.981  0.001  Fall at exit      Larger breeding female herd size (per 100 animal increase)  <0.001  <0.001  1.011  1.001 to 1.021  0.038      Wider alleyway (cm)  −0.105  0.023  0.900  0.858 to 0.944  <0.001  Run out of restraint      Longer processing time per animal (per 10 s)  −0.007  0.002  0.932  0.899 to 0.967  0.001      Wider alleyway (cm)  −0.045  0.007  0.956  0.943 to 0.970  <0.001      Use of a straight alleyway (yes vs. no)  0.795  0.137  2.214  1.668 to 2.938  <0.001      Use of hands as a moving aid (yes vs. no)  −0.342  0.113  0.710  0.563 to 0.897  0.006  1Per 1-unit increase, unless otherwise noted. 2CI = confidence interval. View Large Cattle behavior was associated with stockperson handing at the individual cow level (Table 6). All behaviors were performed more frequently when an electric prod was used (Fig. 1). For example, 59% of cattle that balked were touched with an electric prod. In comparison, of those that did not balk, only 18% were prodded. With the exception of falls at exit, electric prod use was positively associated with negative cattle behavioral outcomes (P < 0.001 for balking, vocalizing, stumbling in the chute, and running out of the restraint; P = 0.001 for stumbling at exit; and P = 0.034 for falling in the chute) and increased the risk of occurrence by at least 71%. The risk of cattle falling on exit was increased by greater than 3-fold when they were touched with moving aids (P = 0.018). Balking was also associated with tail twists (P < 0.001) and being touched with a moving aid (P < 0.001). Risk of stumbles in the chute increased by 80% when cattle were mis-caught (P < 0.001). The risk of cattle running out of the restraint increased by at least 133% if they had previously fallen in the chute (P = 0.048) or stumbled while exiting the restraint (P < 0.001). Table 6. Multivariable associations of cattle behavior outcomes in the chute (balking, vocalizing, stumbling, and falling) with cow-level stockperson handling and, for running out of the restraint, both stockperson handling and previous cow behavior predictors on 30 California cow–calf operations Variable  Estimate  SE  Odds ratio1  95% CI2  P-value  Balk (n = 2,952)      Electric prod use  2.401  0.166  11.029  7.959 to 15.283  <0.001      Tail twist  2.166  0.292  8.719  4.921 to 15.449  <0.001      Moving aid use  3.530  0.356  34.125  16.984 to 68.567  <0.001  Vocalize (n = 2,950)      Electric prod use  1.183  0.178  3.265  2.304 to 4.627  <0.001  Stumble in chute (n = 2,952)      Electric prod use  0.618  0.138  1.856  1.415 to 2.434  <0.001      Mis-catch  0.586  0.160  1.797  1.314 to 2.458  <0.001  Fall in chute (n = 2,954)      Electric prod use  0.820  0.386  2.270  1.065 to 4.841  0.034  Stumble at exit (n = 2,944)      Electric prod use  0.585  0.179  1.794  1.263 to 2.548  0.001  Fall at exit (n = 2,944)      Moving aid use  1.471  0.623  4.355  1.283 to 14.779  0.018  Run out of restraint (n = 2,932)      Previously fell in chute  0.847  0.427  2.332  1.009 to 5.390  0.048      Previously stumbled at exit  1.040  0.192  2.828  1.942 to 4.117  <0.001      Electric prod use  0.539  0.129  1.714  1.332 to 2.205  <0.001  Variable  Estimate  SE  Odds ratio1  95% CI2  P-value  Balk (n = 2,952)      Electric prod use  2.401  0.166  11.029  7.959 to 15.283  <0.001      Tail twist  2.166  0.292  8.719  4.921 to 15.449  <0.001      Moving aid use  3.530  0.356  34.125  16.984 to 68.567  <0.001  Vocalize (n = 2,950)      Electric prod use  1.183  0.178  3.265  2.304 to 4.627  <0.001  Stumble in chute (n = 2,952)      Electric prod use  0.618  0.138  1.856  1.415 to 2.434  <0.001      Mis-catch  0.586  0.160  1.797  1.314 to 2.458  <0.001  Fall in chute (n = 2,954)      Electric prod use  0.820  0.386  2.270  1.065 to 4.841  0.034  Stumble at exit (n = 2,944)      Electric prod use  0.585  0.179  1.794  1.263 to 2.548  0.001  Fall at exit (n = 2,944)      Moving aid use  1.471  0.623  4.355  1.283 to 14.779  0.018  Run out of restraint (n = 2,932)      Previously fell in chute  0.847  0.427  2.332  1.009 to 5.390  0.048      Previously stumbled at exit  1.040  0.192  2.828  1.942 to 4.117  <0.001      Electric prod use  0.539  0.129  1.714  1.332 to 2.205  <0.001  1Yes vs. no. 2CI = confidence interval. View Large Table 6. Multivariable associations of cattle behavior outcomes in the chute (balking, vocalizing, stumbling, and falling) with cow-level stockperson handling and, for running out of the restraint, both stockperson handling and previous cow behavior predictors on 30 California cow–calf operations Variable  Estimate  SE  Odds ratio1  95% CI2  P-value  Balk (n = 2,952)      Electric prod use  2.401  0.166  11.029  7.959 to 15.283  <0.001      Tail twist  2.166  0.292  8.719  4.921 to 15.449  <0.001      Moving aid use  3.530  0.356  34.125  16.984 to 68.567  <0.001  Vocalize (n = 2,950)      Electric prod use  1.183  0.178  3.265  2.304 to 4.627  <0.001  Stumble in chute (n = 2,952)      Electric prod use  0.618  0.138  1.856  1.415 to 2.434  <0.001      Mis-catch  0.586  0.160  1.797  1.314 to 2.458  <0.001  Fall in chute (n = 2,954)      Electric prod use  0.820  0.386  2.270  1.065 to 4.841  0.034  Stumble at exit (n = 2,944)      Electric prod use  0.585  0.179  1.794  1.263 to 2.548  0.001  Fall at exit (n = 2,944)      Moving aid use  1.471  0.623  4.355  1.283 to 14.779  0.018  Run out of restraint (n = 2,932)      Previously fell in chute  0.847  0.427  2.332  1.009 to 5.390  0.048      Previously stumbled at exit  1.040  0.192  2.828  1.942 to 4.117  <0.001      Electric prod use  0.539  0.129  1.714  1.332 to 2.205  <0.001  Variable  Estimate  SE  Odds ratio1  95% CI2  P-value  Balk (n = 2,952)      Electric prod use  2.401  0.166  11.029  7.959 to 15.283  <0.001      Tail twist  2.166  0.292  8.719  4.921 to 15.449  <0.001      Moving aid use  3.530  0.356  34.125  16.984 to 68.567  <0.001  Vocalize (n = 2,950)      Electric prod use  1.183  0.178  3.265  2.304 to 4.627  <0.001  Stumble in chute (n = 2,952)      Electric prod use  0.618  0.138  1.856  1.415 to 2.434  <0.001      Mis-catch  0.586  0.160  1.797  1.314 to 2.458  <0.001  Fall in chute (n = 2,954)      Electric prod use  0.820  0.386  2.270  1.065 to 4.841  0.034  Stumble at exit (n = 2,944)      Electric prod use  0.585  0.179  1.794  1.263 to 2.548  0.001  Fall at exit (n = 2,944)      Moving aid use  1.471  0.623  4.355  1.283 to 14.779  0.018  Run out of restraint (n = 2,932)      Previously fell in chute  0.847  0.427  2.332  1.009 to 5.390  0.048      Previously stumbled at exit  1.040  0.192  2.828  1.942 to 4.117  <0.001      Electric prod use  0.539  0.129  1.714  1.332 to 2.205  <0.001  1Yes vs. no. 2CI = confidence interval. View Large Figure 1. View largeDownload slide Percentage of cattle that were touched and not touched with an electric prod that did or did not balk, vocalize, stumble, fall, or run out of the restraint during a routine procedure on cow–calf ranches. Numbers denote the sample size within each behavioral category. Figure 1. View largeDownload slide Percentage of cattle that were touched and not touched with an electric prod that did or did not balk, vocalize, stumble, fall, or run out of the restraint during a routine procedure on cow–calf ranches. Numbers denote the sample size within each behavioral category. DISCUSSION Cattle health outcomes are multifaceted and associated with ranch characteristics, management practices, and producer perspectives. Stockperson handling may be associated with level of experience and training on the ranch. Various aspects of the facility design were associated with how cattle behaved during handling, and cattle that were handled more often had better outcomes. At the cow level, the implications of electric prod use were clear, as it was associated with a greater risk of balking, vocalizing, stumbling and falling in the chute, stumbling at exit, and running out of the restraint, which agrees with the existing scientific literature. Conversely, the interpretation of touching cattle with a moving aid was unclear, as it was protective for negative behavioral outcomes at the herd level but increased risk at the cow level. Cattle Health The prevalence of thin cattle ranged from 0 to 29.4%, averaging 2.3% (SD 5.5). There was a greater risk of having thin cattle on older ranches and those that checked their water access points more frequently in the previous 2 wk. This disagrees with our original hypothesis, as we predicted that the older ranches would have more experience in cattle husbandry and therefore better health outcomes. However, these outcomes could also be influence by the timing of the ranch visit in relation to the herd's calving season. We would expect cows to have a lower body condition following parturition, but information about this time frame was not collected. We would also expect that ranches that were checking their water access points more often may have also been concurrently checking feed levels, which may happen more frequently if forage was sparse. As we conducted this study over the course of a 13 consecutive month period, we speculate that cattle body condition could also be associated with seasonal forage availability within the region. However, directly measuring feed levels was not feasible during the assessment. In addition, as operations often graze their cattle on many locations throughout the state (e.g., winter grazing in California's Central Valley vs. summer grazing in the northern mountains), it is difficult to assess if there is a relationship between ranch age and location. Because there is considerable variability in forage levels, evaluating the animal-based outcome of body condition is a more useful way to assess welfare. It is not clear what ranch age or the frequency of water access point checks mean in terms of forage levels. The prevalence of swelling (mean 2.0% [SD 2.6]) and lameness (mean 1.3% [SD 1.5]) was low across ranches. Cattle had a decreased risk of swelling when producers were more empathetic with animal's pain experience. This agrees with Kielland et al. (2009), who observed a lower prevalence of leg and hock injuries when producers were more empathic toward their cattle. However, we found contrasting results with lameness, as it was positively associated with producer empathy. Recognition of lameness and knowledge of intervention methods is a challenge in the cattle industry (Whay et al., 2003; Webster, 2005; Tremetsberger and Winckler, 2015), but less is known about the awareness of this relatively low incidence disease (averaging 1.3 vs. 30% on California dairies; von Keyserlingk et al., 2012) in the cow–calf sector. Cattle were at greater risk of having swelling and hairless patches when the producer thought that there was room for improvement in their herd health management plan. This might indicate that producers can recognize health problems in the herd and identify that they have opportunities to improve, which agreed with our predictions. We also hypothesized that there may be an association between the location where the ranch administers injections and the area of the body where swelling occurs, because it was often seen in the form of a prior injection site abscess when pilot testing the assessment. However, there was not enough variation in injection site location to be included in the analysis. The risk of hairless patches decreased when the veterinarian visited the ranch more frequently. Veterinarians can provide producers with information on the cause of integument changes, and therefore, more frequent visits to the ranch may increase the likelihood that the problem is treated. The risk of hairless patches also increased with more water access point checks per year, disagreeing with our predictions. We expected that operations that are more attentive to the nutrients provided for the cattle, such as water, might also be concerned about other aspects of health. Conversely, the risk of abrasions decreased with more herd health checks per year. This agrees with our hypothesized direction of the effect, as we predicted that while checking the herd, producers might also be surveying the environment for abrasive features, such as sharp edges in the corrals or handling facilities. As predicted, there was a positive association between prevalence of blindness in herds and faster producer reaction times to sick cattle. Some eye problems, such as pink eye, are easily transmitted among cattle (Brown et al., 1998), so producers that are aware of this issue in the herd might be more apt to react quickly to prevent spread of the disease. The negative associations of hairless patches and blind eyes with an increasing number of family farming generations agrees with our hypotheses, as we predicted that experience in cattle husbandry could lead to better health outcomes. In contrast to our predictions, the risk of dirty cattle increased when cattle were handled more frequently each year. However, in agreement, it decreased when the ranch used the BQA training program and when the herd had been checked for health problems more often in the previous 2 wk. The prevalence of dirty cattle may also be associated with weather conditions (e.g., if it rained the night before or during the ranch visit) and the substrate and duration spent in the holding pens before processing (e.g., if cattle were gathered and held in a muddy corral the night before). However, these factors were not evaluated in this study and additional research is needed to assess other risk factors for the relatively low incidence of dirty cattle (1.5%). Indeed, it might be that dirtiness is a seasonal issue or one that is faced primarily in the feedlot, because by the time they are slaughtered, 74% of beef cattle have mud or manure on their bodies (Garcia et al., 2008). Maintaining written records for sick cattle, treatment, or death losses or having knowledge of the diseases experienced or the cause of the majority of deaths on the ranch was not a predictor for any health outcome measures. However, there are a number of common health problems, such as bovine respiratory disease, mineral deficiencies, or dystocia, that were not measured in our assessment and may be more directly associated with these factors. Stockperson Handling We found that older ranches, and those with more family farming generations in the producer's family, had lower odds of using an electric prod. Interestingly, there were also lower odds of touching cattle with a moving aid on ranches with more farming generations. The level of experience that is associated with businesses that have been operating longer, and the cattle handling knowledge that is passed down through farming generations, might explain this relationship. However, the risk of electric prod use increased on operations that had more breeding cows. Producers may perceive electric prods as a tool to increase cattle flow rate and might be more apt to use them when they have more cattle to process. Ranches that used the BQA training program had lower risk of mis-catching cattle in the restraint. This could possibly be explained because a portion of this program concentrates on proper cattle handling techniques and provides a self-assessment, so producers can evaluate their stockmanship performance (BQA, 2010). The interaction between stockperson attributes and the subsequent cattle handling techniques is complex, with the most promising predictors being the individual's perspectives toward animals (Hemsworth, 2003), at least in other research. However, we did not find such a relationship in our study, perhaps because we did not assess the perspectives of all the stockpersons who handled cattle on each ranch. Cattle Behavior in the Chute When considering predictors at the herd level, cattle that were handled more frequently each year had a lower risk of vocalizing and stumbling, both in the chute and on exit. Others have found also fewer signs of agitation in the chute when cattle are handled more frequently (handled 2 to 3 times weekly for 4 wk vs. left undisturbed on pasture; Cooke et al., 2009; Francisco et al., 2012). Cattle had greater odds of running out of the chute when a straight alleyway was used. This agrees with the speculation that livestock animals remain calm during handling when their vision is blocked through the use of curved alleyways, solid side walls, and louvers (Grandin, 2008). However, hydraulic restraints and the presence of louvers (all of which were installed on hydraulic systems) increased the risk of vocalization in our study. Grandin (1998) found that cattle begin to vocalize when chute pressure increases to levels that are more easily attainable when the chute is hydraulically operated. Hydraulic restraints, however, were protective in the cases of balking and falling in the chute. Adjusting the pressure applied in these systems may be a way to decrease vocalizations and maintain the additional protective design benefits. The risk of vocalizations also increased with longer alleyways, which disagreed with our hypothesis, as increasing the length of an alleyway (e.g., extending a curved system to make a complete half circle) is often encouraged; alleyway design is thought to be the most important factor influencing cattle handling outcomes (Grandin, 2008). However, the recommended optimal alleyway length is between 12 and 15 m because at excessive lengths, cattle may become restless from standing too long (Grandin, 1997b). Therefore, the association between longer alleyways and increased vocalizations may be due to accrued agitation while standing in the handling facilities. Additionally, wider alleyways were protective for balking, stumbling and falling at exit, and running out of the restraint. However, the way a cow behaves in a wider alleyway might be relative to its size, as it is recommended that alleyways be designed to allow 2 cm of empty space on either side of the largest cow being handled to improve flow (Grandin, 1997b). Information about cow frame size was not collected in our study. Also, information on whether the sides of the alleyway were constructed perpendicular to the ground or V-shaped was not collected but may also be associated with the relationship between alleyway width and cattle behavior. Balking was also associated with longer processing time per cow, a relationship that would be expected, as balking was recorded after 4 s of delay time between when the cow was touched by a handler and when she moved in any direction. We predicted that longer processing times may allow cattle to move slower in the chute, become less agitated, and have a better behavioral outcome. Indeed, this decreased the risk of cattle stumbling on exit and running out of the restraint. We predicted that operations with larger herds would invest in better handling facilities, allowing for fluid cattle movement and improved responses in the chute. However, larger herd size was associated with an increased risk of stumbles and falls at exit. The association here may be explained by the fact that ranches with larger herd sizes were also at greater risk of using electric prods, a handling technique that increased the risk of stumbles and falls. The use of moving aids was protective for balking and running out of the restraint at the herd level but was a risk factor for balking and falling at exit at the cow level. Simon et al. (2016) discuss the difficulty in interpreting what touching cattle with a moving aid means in terms of welfare (i.e., moving aid use ranged from a gentle touch with a hand to an aggressive poke with a stick in a sensitive area), and the results of this study highlight this problem. For example, by definition, all cattle that balked were previously touched with a moving aid or electric prod, so we would expect to see this risk factor association at the cow level. However, it is less clear why using hands as a moving aid decreases the risk of balking when analyzed at the herd level. In future studies, recording methods could be improved in terms of how (e.g., gentle vs. aggressive touches), where (e.g., sensitive vs. nonsensitive areas), when (e.g., only once when cattle balk vs. repetitively), and what type (e.g., hand vs. stick) of moving aid is being used to understand this relationship. When considering cattle behavior at the cow level, the risk of balking, vocalizing, stumbling and falling in the chute, stumbling at exit, and running out of the restraint increased when electric prods were used. This agrees with literature showing that cattle find electric prods aversive (Pajor et al., 2000) and their use is associated with increased vocalization (Grandin, 1998) and exit speed (Stewart et al., 2008). Therefore, reducing the use of electric prods could improve behavioral responses, both initially and subsequently. For example, reducing the prevalence of previous falls and stumbles, which are positively associated with electric prod use, had a protective effect on cattle running out of the restraint. Of all the cattle behaviors exhibited during handling, balking corresponded with the most prod use (in 59% of these cases). Although we did not record the order of electric prod use in this study (i.e., before or after behaviors were performed), if balking prevalence was reduced, we would also expect to see a decrease in prodding, as others have observed at slaughter plants (Grandin, 2001). We would also expect a decrease in tail twists and use of a moving aid with less balking. Although balking appears to be a strategic area for improvement, electric prod use remains a more important area of concern. Cattle that were mis-caught in the restraint had a greater risk of stumbling in the chute. Again, recording the order of events, in terms of when the cattle stumbled, would be beneficial to understand this relationship, because the restraint was often readjusted if a cow went down to her knees during the procedure, which constituted a mis-catch. There were several factors that may have influenced the cattle behavioral results that were not assessed in this study. We suggest that aversive behaviors, such as balking, should be measured at multiple points during the handling event, including at the entrance of the single file alley and the crowd pen because Grandin (1997a) suggests that cattle exhibit more passive behavior once they enter the alleyway. The design of our study, however, did not include observations taken before the single file alleyway because this would have required a third observer. Also, various external noises can affect how cattle react in handling situations (Waynert et al., 1999), but our study did not measure these noises. Additionally, we hypothesized that the shape of the head gate stanchion, the floor material in the restraint, type of crowd pen, presence and type of anti-back gate, and treatment of aggressive cows would influence cattle behavior. Due to a lack of variation in these measures, they could not be included in our analysis. The type of procedure performed on the cattle, whether it was a pregnancy check, vaccination, or changing of ear tag, was not a predictor for any of the cattle behavioral outcomes. We asked to observe less invasive procedures to increase comparability of our results; however, we predict that extreme events, such as branding, may have a greater effect on cattle behavior. Furthermore, although we are not aware of any systematic bias in animal management or handling, the fact that all the producers in the study were chosen based on their willingness to participate may have influenced the results. Additionally, we did not find an association between cattle behavior and the presence of dogs. Although none of the dogs in this study touched the cattle (e.g., through biting) or were actively working to move cattle thought the alley, their proximity to the cattle may be a better predictor for these results but was not systematically recorded. Likewise, we predicted that having an open-sided wall in the alleyway or a funnel fence crowd pen would increase adverse cattle behaviors measured, but as we saw no association with these predictors, the way stockpersons use the facilities (e.g., body position or the number of cattle loaded into the crowd pen at a time) may have a greater influence on the outcomes. Finally, the type of floor material in the alleyway and restraint were not predictors for any cattle behavioral model, possibly due to the variation within each material type. For example, dirt floors were dry and course in some cases and wet, muddy, and slick in others. Conclusion Cattle health problems are associated with various ranch characteristics (e.g., ranch age), aspects of management (e.g., frequency of visual herd health checks), and producer perspectives (e.g., empathy toward an animal's pain experience). Cattle handling was influenced by a number of facility and stockperson factors; personnel training, facility design (e.g., alleyway design), and excessive use of electric prods are key areas for future improvements. Footnotes 1 This project was financially supported by the Russell L. Rustici Rangeland and Cattle Research Endowment. The authors thank the University of California Cooperative Extension and the California Cattleman's Association for their assistance in recruiting cow–calf participants and the infrastructure support of the Department of Animal Science, the College of Agricultural and Environmental Sciences, and the California Agricultural Experiment Station of the University of California Davis. For more information about the cow-calf assessment and how it can be applied to a ranch, visit www.ucdcowcalfassessment.com. LITERATURE CITED Beef Quality Assurance (BQA) 2010. Assessor's guide to a Beef Quality Assurance cow-calf assessment. http://www.bqa.org/Media/BQA/Docs/assessors_guide_cow-calf.pdf (Accessed August 25, 2014.) Brenninkmeyer C. Dippel S. Brinkmann J. March S. Winckler C. Knierim U. 2013. Hock lesion epidemiology in cubicle housed dairy cows across two breeds, farming systems and countries. Prev. Vet. Med.  109: 236– 245. doi: https://doi.org/10.1016/j.prevetmed.2012.10.014 Google Scholar CrossRef Search ADS PubMed  Brown M. H. Brightman A. H. Fenwick B. W. Rider M. A. 1998. Infectious bovine keratoconjunctivitis: A review. J. Vet. Intern. Med.  12: 259– 266. doi: https://doi.org/10.1111/j.1939-1676.1998.tb02120.x Google Scholar CrossRef Search ADS PubMed  Callan R. J. Garry F. B. 2002. Biosecurity and bovine respiratory disease. Vet. Clin. North Am. Food Anim. Pract.  18: 57– 77. doi: https://doi.org/10.1016/S0749-0720(02)00004-X Google Scholar CrossRef Search ADS PubMed  Chapinal N. Liang Y. Weary D. M. Wang Y. von Keyserlingk M. A. G. 2014. Risk factors for lameness and hock injuries in Holstein herds in China. J. Dairy Sci.  97: 4309– 4316. doi: https://doi.org/10.3168/jds.2014-8089 Google Scholar CrossRef Search ADS PubMed  Cooke R. F. Arthington J. D. Austin B. R. Yelich J. V. 2009. Effects of acclimation to handling on performance, reproductive, and physiological responses of Brahman-crossbred heifers. J. Anim. Sci.  87: 3403– 3412. doi: https://doi.org/10.2527/jas.2009-1910 Google Scholar CrossRef Search ADS PubMed  Dippel S. Dolezal M. Brenninkmeyer C. Brinkmann J. March S. Knierim U. Winckler C. 2009a. Risk factors for lameness in cubicle housed Austrian Simmental dairy cows. Prev. Vet. Med.  90: 102– 112. doi: https://doi.org/10.1016/j.prevetmed.2009.03.014 Google Scholar CrossRef Search ADS   Dippel S. Dolezal M. Brenninkmeyer C. Brinkmann J. March S. Knierim U. Winckler C. 2009b. Risk factors for lameness in freestall-housed dairy cows across two breeds, farming systems, and countries. J. Dairy Sci.  92: 5476– 5486. doi: https://doi.org/10.3168/jds.2009-2288 Google Scholar CrossRef Search ADS   Francisco C. L. Cooke R. F. Marques R. S. Mills R. R. Bohnert D. W. 2012. Effects of temperament and acclimation to handling on feedlot performance of feeder cattle originated from a rangeland-based cow–calf system. J. Anim. Sci.  90: 5067– 5077. doi: https://doi.org/10.2527/jas.2012-5447 Google Scholar CrossRef Search ADS PubMed  Garcia L. G. Nicholson K. L. Hoffman T. W. Lawrence T. E. Hale D. S. Griffin D. B. Savell J. W. Van Overbeke D. L. Morgan J. B. Belk K. E. 2008. National Beef Quality Audit – 2005: Survey of targeted cattle and carcass characteristics related to quality, quantity, and value of fed steers and heifers. J. Anim. Sci.  86: 3533– 3543. doi: https://doi.org/10.2527/jas.2007-0782 Google Scholar CrossRef Search ADS PubMed  Grandin T 1997a. Assessment of stress during handling and transport. J. Anim. Sci.  75: 249– 257. Google Scholar CrossRef Search ADS   Grandin T 1997b. The design and construction of facilities for handling cattle. Livest. Prod. Sci.  49: 103– 119. doi: https://doi.org/10.1016/S0301-6226(97)00008-0 Google Scholar CrossRef Search ADS   Grandin T 1998. The feasibility of using vocalization scoring as an indicator of poor welfare during cattle slaughter. Appl. Anim. Behav. Sci.  56: 121– 128. doi: https://doi.org/10.1016/S0168-1591(97)00102-0 Google Scholar CrossRef Search ADS   Grandin T 2001. Cattle vocalizations are associated with handling and equipment problems at beef slaughter plants. Appl. Anim. Behav. Sci.  71: 191– 201. doi: https://doi.org/10.1016/S0168-1591(00)00179-9 Google Scholar CrossRef Search ADS PubMed  Grandin T 2008. Humane livestock handling. Storey Publishing, North Adams, MA. Hemsworth P. H 2003. Human-animal interactions in livestock production. Appl. Anim. Behav. Sci.  81: 185– 198. doi: https://doi.org/10.1016/S0168-1591(02)00280-0 Google Scholar CrossRef Search ADS   Hollenbeck A. R 1978. Problems of reliability in observational research. In: Sackett G. P. editor, Observing behavior: Data collection and analysis methods no. 2.  University Park Press, Baltimore, MD. p. 79– 98. Ivemeyer S. Knierim U. Waiblinger S. 2011. Effect of human-animal relationship and management on udder health in Swiss dairy herds. J. Dairy Sci.  94: 5890– 5902. doi: https://doi.org/10.3168/jds.2010-4048 Google Scholar CrossRef Search ADS PubMed  Kielland C. Ruud L. E. Zanella A. J. Østerås O. 2009. Prevalence and risk factors for skin lesions on legs of dairy cattle housed in freestalls in Norway. J. Dairy Sci.  92: 5487– 5496. doi: https://doi.org/10.3168/jds.2009-2293 Google Scholar CrossRef Search ADS PubMed  Lombard J. E. Tucker C. B. von Keyserlingk M. A. G. Kopral C. A. Weary D. M. 2010. Associations between cow hygiene, hock injuries, and free stall usage on US dairy farms. J. Dairy Sci.  93: 4668– 4676. doi: https://doi.org/10.3168/jds.2010-3225 Google Scholar CrossRef Search ADS PubMed  Martin P. Bateson P. 1993. Measuring behaviour: An introductory guide. 2nd ed. Cambridge Univ. Press, Cambridge, UK. Millman S. T 2009. Animal welfare – Scientific approaches to the issues. J. Appl. Anim. Welf. Sci.  12: 88– 96. doi: https://doi.org/10.1080/10888700902719591 Google Scholar CrossRef Search ADS PubMed  Pajor E. A. Rushen J. de Passillé A. M. B. 2000. Aversion learning techniques to evaluate dairy cattle handling practices. Appl. Anim. Behav. Sci.  69: 89– 102. doi: https://doi.org/10.1016/S0168-1591(00)00119-2 Google Scholar CrossRef Search ADS PubMed  Simon G. E. Hoar B. R. Tucker C. B. 2016. Assessing cow–calf welfare. Part 1: Benchmarking beef cow health and behavior, handling; and management, facilities, and producer perspectives. doi: https://doi.org/10.2527/jas2016-0308 Stewart M. Schaefer A. L. Haley D. B. Colyn J. Cook N. J. Stafford K. J. Webster J. R. 2008. Infrared thermography as a non-invasive method for detecting fear-related responses of cattle to handling procedures. Anim. Welf.  17: 387– 393. Tremetsberger L. Winckler C. 2015. Effectiveness of animal health and welfare planning in dairy herds: A review. Anim. Welf.  24: 55– 67. doi: https://doi.org/10.7120/09627286.24.1.055 Google Scholar CrossRef Search ADS   United States Department of Health and Human Services 2009. Code of Federal Regulations. Title 45 Part 46. Washington, DC. von Keyserlingk M. A. G. Barrientos A. Ito K. Galo E. Weary D. M. 2012. Benchmarking cow comfort on North American freestall dairies: Lameness, leg injuries, lying time, facility design, and management for high-producing Holstein dairy cows. J. Dairy Sci.  95: 7399– 7408. doi: https://doi.org/10.3168/jds.2012-5807 Google Scholar CrossRef Search ADS PubMed  Waynert D. F. Stookey J. M. Schwartzkopf-Genswein K. S. Watts J. M. Waltz C. S. 1999. The response of beef cattle to noise during handling. Appl. Anim. Behav. Sci.  62: 27– 42. doi: https://doi.org/10.1016/S0168-1591(98)00211-1 Google Scholar CrossRef Search ADS   Webster J 2005. The assessment and implementation of animal welfare: Theory into practice. Rev. Sci. Tech.  24: 723– 734. Google Scholar CrossRef Search ADS PubMed  Welfare Quality 2009. Welfare Quality® assessment protocol for cattle. Welfare Quality Consortium, Lelystad, the Netherlands. Whay H. R. Main D. C. Green L. E. Webster A. J. 2003. Assessment of the welfare of dairy cattle using animal-based measurements: Direct observations and investigation of farm records. Vet. Rec.  153: 197– 202. doi: https://doi.org/10.1136/vr.153.7.197 Google Scholar CrossRef Search ADS PubMed  American Society of Animal Science
TI - Assessing cow–calf welfare. Part 2: Risk factors for beef cow health and behavior and stockperson handling
JF - Journal of Animal Science
DO - 10.2527/jas.2016-0309
DA - 2016-08-01
UR - https://www.deepdyve.com/lp/oxford-university-press/assessing-cow-calf-welfare-part-2-risk-factors-for-beef-cow-health-and-akYaHW0bwr
SP - 3488
EP - 3500
VL - 94
IS - 8
DP - DeepDyve
ER -