TY - JOUR
AU - Tucker, C. B.
AB - ABSTRACT Feedlot cattle are monitored for the sickness response, both physiological and behavioral, to detect bovine respiratory disease (BRD), but this method can be inaccurate. Diagnostic accuracy may improve if the BRD sickness response is better understood. We hypothesized that steers around peak BRD would have fever, anorexia, and less grooming than controls. We also expected sickness response magnitude to be greater as clinical and pathological severity increased. Unvaccinated steers were assigned to challenge with 1 of 5 BRD viruses or bacteria (BRD challenge; n = 4/pathogen; 20 total), based on susceptibility as determined by serology. Body weight–matched vaccinated animals were given sterile media (Control; n = 4/pathogen; 20 total) and housed by treatment (5 pens/treatment). Rectal temperature was logged every 5 min between 0100 and 0700 h, and time spent feeding (24 h/d), in contact with a brush (13 h/d), and self-licking (24 h/d) were collected from video recordings. Steers were examined and a clinical score (CS) was assigned daily. Bovine respiratory disease challenge steers were euthanized after 5 to 15 d (timing was pathogen specific) and the proportion of grossly affected lung (%LUNG) was recorded. The day of highest CS (peak; d 0) for each BRD challenge steer and the 2 preceding days were analyzed for all variables except self-licking (d 0 only); analogous days were included for Controls. Penwise mixed models (pen was the experimental unit) were used to determine which sickness response elements differed between treatments before and at peak disease, and regression using individual-steer data was used to describe relationships between disease severity (n = 35 for CS and n = 20 for %LUNG) and fever, anorexia, and grooming. Bovine respiratory disease challenge steers had fever (1.1°C higher; P < 0.01) and anorexia (35% lower feeding time; P = 0.03) but did not differ from healthy Controls for brush contact (P = 0.37) or self-licking (P = 0.15). Higher CS and more %LUNG were associated with increased fever (d 0; P ≤ 0.04) and lower feeding (d 0; P < 0.01), brush contact (d 0; P ≤ 0.03), and self-licking (P ≤ 0.05) duration relative to lower CS and less %LUNG. In conclusion, fever and feeding time are good BRD diagnostic measures around peak CS. Further study is needed to clarify why grooming was not a good measure. The sickness response is greater as BRD severity increases; fever is most closely related to CS and anorexia is most closely related to %LUNG. Regardless of which aspect is monitored, cattle with milder disease may be more difficult to detect than sicker animals. INTRODUCTION Clinical bovine respiratory disease (BRD) affects 16.2% of U.S. cattle (USDA, 2013), causing fever (Timsit et al., 2011), anorexia (Quimby et al., 2001), and lethargy (Theurer et al., 2013). These clinical signs, used by feedlot personnel to detect BRD, are part of the sickness response, a suite of immune-mediated physiological and behavioral changes caused by potentially injurious stimuli (Hart, 1988; Pecchi et al., 2009). However, detection by feedlot personnel is inaccurate compared with postmortem findings (White and Renter, 2009) and it is unclear why. Improved detection is important from animal welfare and economic standpoints (Snowder et al., 2006; Millman, 2007). Although beneficial during sickness (Pecchi et al., 2009), fever and anorexia are metabolically expensive (Adelman and Martin, 2009). Lethargy, including reduced grooming (Hart, 1988), conserves energy (Aubert, 1999). However, grooming is a little-studied BRD indicator, despite evidence that sick cattle groom less (in mastitis [Fogsgaard et al., 2012] and in BRD [Toaff-Rosenstein et al., 2016). The sickness response varies with disease severity. Sickness response expression is greater at increased endotoxin doses (for fever, anorexia, activity, and grooming; Skinner et al., 2009; Harden et al., 2011; Tikhonova et al., 2011) in rodents. Similarly, cattle with increased parasite loads (Szyszka and Kyriazakis, 2013), more severe mastitis (Kemp et al., 2008), and BRD (White et al., 2012) show a greater sickness response than those with mild disease. Therefore, cattle with mild BRD may have more subtle clinical signs and, therefore, escape detection. Our objectives were to 1) assess fever, feeding, and grooming around peak clinical disease as diagnostic measures and 2) describe the relationship between fever, feeding, and grooming and BRD severity. We hypothesized that BRD would result in fever, anorexia, and less grooming and that sickness response expression would increase in proportion to disease severity. MATERIALS AND METHODS Animals and Challenge Procedure This study took place between June and September 2012 at the beef research facility of the University of California, Davis (Davis, CA) after approval by the Institutional Animal Care and Use Committee. Forty Angus–Hereford crossbred steers from the university herd (7–10.5 mo old and 306 kg average BW, with a range of 216–390 kg, at the start of the trial) were enrolled. Before weaning, the entire herd was vaccinated for Histophilus somni (Somubac; Zoetis, Florham Park, NJ) and Moraxella bovis (Addison Biological Laboratory Inc., Fayette, MO). Additionally, a portion of the herd was randomly vaccinated for bovine respiratory syncytial virus (BRSV) and infectious bovine rhinotracheitis (IBR; Bovi-Shield Gold; Zoetis). At weaning, all steers were treated with a topical, broad-spectrum parasiticide (Dectomax; Zoetis). Respiratory disease was induced using 5 different pathogens because the current work was done in conjunction with another study (Gershwin et al., 2015) whose objective included examining RNA-level changes after single-pathogen challenge. Our objective was to understand the BRD sickness response, including physiological and behavioral changes, as related to clinical severity and without regards to any pathogen in particular. Therefore, the most important consideration was that challenged animals were clinically affected. Data were obtained from the period around the peak of clinical signs, which varied by pathogen (described below), to confine conclusions to a biologically relevant time point that would be relatively uniform across pathogens. Clinically healthy steers, chosen from the unvaccinated portion of the herd, were assigned in a nonrandom manner to 1 of 5 sequential, single-pathogen challenge trials (BRD challenge; n = 20; 4/pathogen trial) performed in this order: BRSV, IBR, Mycoplasma bovis, Pasteurella multocida, and Mannheimia haemolytica. Assignment to pathogen was based on susceptibility to challenge pathogen, as determined by serology (low titer, indicative of poor specific immunity to that pathogen), and BW. Serology was performed on only unvaccinated steers (for BRSV and IBR) and on all remaining animals (for P. multocida and M. haemolytica). Mycoplasma bovis assignment was based on vaccination status (only animals unvaccinated for BRD pathogens were assigned to Mycoplasma bovis challenge and vaccinated steers to Mycoplasma bovis control) as serology data were unavailable. In general, heavier steers were assigned to earlier trials to improve uniformity of animal size across the entire study. Within weight groups, steers having the lowest titers were assigned to BRD challenge. The remaining animals were Controls (n = 20; 4/pathogen trial). Ten milliliters of whole blood was collected by jugular venipuncture in silicone-coated clot tubes (Becton, Dickinson and Company, Franklin Lakes, NJ) and stored at 4°C overnight, after which serum was removed and centrifuged at 2,000 rotations/min 500G for 10 minutes at room temperature and the supernatant was submitted to the California Animal Health and Food Safety Laboratory (Davis, CA) for BRSV and IBR titers using indirect immune-fluorescent antibody staining or the laboratory of Dr. A. Confer (Oklahoma State University, Stillwater, OK) for P. multocida and M. haemolytica serological titer determination by ELISA. After they were assigned to treatment, BRD challenge steers received no additional vaccines, to render them more susceptible to the pathogens, whereas Controls, to increase their likelihood of remaining healthy, were also vaccinated for BRSV, IBR, parainfluenza-3, and bovine viral diarrhea virus (Bovi-Shield Gold 5; Zoetis) and M. haemolytica (One Shot; Zoetis). Some serology results arrived after treatment assignment, and 10 steers were switched between treatments based on these findings. As a result, 4 Controls did not receive any additional vaccines, 2 Controls were not vaccinated for M. haemolytica, and 4 BRD challenge steers (2 M. haemolytica and 2 P. multocida) were fully vaccinated. Steers were transported 84 km to the research facility no fewer than 10 d before the study started. Seven days before challenge, unique markings were made with hair dye (Clairol Nice ‘N Easy Borne Blonde; Procter and Gamble, Cincinnati, OH) on the left and right withers and flanks to facilitate video identification. Six days before challenge, a topical insecticide (Cylence; Bayer Animal Health, Shawnee Mission, KS) was applied. Additionally, in the 86% of steers that had clinical signs of mild to moderate ringworm (round, hairless, scaled lesions), affected areas were scrubbed daily with a 2% topical iodine solution. Viruses were aerosolized through an equine face mask (AeroMask Equine System size medium; Trudell Medical International, London, ON, Canada) using a DeVilbiss air compressor/nebulizer system (DeVilbiss Healthcare, Somerset, PA). Bacteria were given through a sterilized nasotracheal tube inserted until the tracheal bifurcation, after application of topical lidocaine to the nasal passages. Control steers were given an equivalent volume of sterile media, using the same route of administration as their respective challenge group. Further details pertaining to the challenge procedure are described in a companion manuscript (Gershwin et al., 2015). Steers were monitored after challenge for a predetermined period ranging from 5 to 15 d, based on anticipated peak clinical disease for their pathogen (7 d for BRSV, 6 d for IBR, 15 d for Mycoplasma bovis, 5 d for M. haemolytica, and 6 d for P. multocida), as determined from pilot studies (Gershwin et al., 2015). At the end of the monitoring period, only BRD challenge steers were necropsied. For analysis, each day was considered to start and end at midnight. Housing and Behavioral Data Collection Steers were acclimated to experimental housing starting 3 d before challenge and grouped by treatment (4/pen) in 1 of 2 identical roofed, concrete-floored pens (7.2 by 9.1 m) separated by 18.3 m. Treatment assignment to each pen alternated by pathogen trial. Pens were enclosed by a solid wall (1 side), a floor-to-ceiling tarp bolstered by straw bales (1 side), and wooden and metal fencing (2 sides), allowing for dim natural lighting and ventilation, and fly traps were hung throughout. Pens were cleaned daily and straw bedding (approximately 25 cm deep) was replenished. In between trials, all animal areas and equipment were cleaned and disinfected and remained vacant for at least 3 d. A total of 8 black-and-white closed-circuit television video cameras (model number WV-BP334; Panasonic Corporation of North America, Secaucus, NJ) and lenses (model number 13VG2812ASII; Tamron USA Inc., Commack, NY) were installed 1.8 to 2.9 m above each pen. Continuous recording at medium quality and 30 frames/s was performed using a digital video recorder with a combo card (model number GV-1120/1240/1480; USA Vision Systems Inc., Irvine, CA) and a computer equipped with digital surveillance software (Surveillance System V8.5; GeoVision Inc., Taipei, Taiwan). To enable video recording at night, red holiday lights suspended from overhead rafters were lit between 1600 and 0800 h. Observers collecting sickness response behavioral data (time spent feeding, in contact with a brush, and self-licking) from video recordings were blind to treatment assignment. Ambient temperature was monitored with a portable weather station (WS-16; NovaLynx Corp., Auburn, CA). The average daily temperature during the trial was 22.9°C (range 16.3–30.7°C), with a minimum of 8.8°C and a maximum of 41.7°C. The average relative humidity during the trial was 49.5% (range 20.1–65.0%) with a minimum of 7.4% and a maximum of 88.2% Clinical Examination and Necropsy A clinical examination, modified from that described by Collie (1992; Table 1), was performed daily by a veterinarian or trained veterinary student who was aware of treatment assignment, starting at 0800 h. The examination included both generalized (part of the sickness response, e.g., rectal temperature [RT], rumen fill) as well as respiratory-specific clinical signs (e.g., nasal discharge, cough). The order of examination (BRD challenge or Control steers first) was balanced over trials. Separate chutes and handling areas were used for BRD challenge and Control steers, and strict biosecurity protocols were followed. On average, each group of 4 steers was away from its pen for clinical examination for 57 ± 19 min/d. Table 1. Points awarded and summed for each of the 21 components of the clinical examination to generate a clinical score for 26 steers on the day of peak clinical disease, by pathogen. The mean (SD; for rectal temperature [RT] and respiratory rate [RR]) or the number of steers that had each component is shown. Point ranges are given for those components in which severity was taken into account. For bovine respiratory disease (BRD) challenge steers, peak clinical disease was the day with the highest clinical score. Controls were weight matched to BRD challenge steers and data from the analogous day used     Bovine respiratory syncytial virus  Infectious bovine rhinotracheitis  Mannheimia haemolytica  Mycoplasma bovis  Pasteurella multocida  Clinical feature  Points (maximum)  BRD challenge  Control  BRD challenge  Control  BRD challenge  Control  BRD challenge  Control  BRD challenge  Control  Lethargy  60  3  0  4  0  2  0  4  0  0  0  RT, °C (SD)  40 × (RT − 39.5)  39.5 (0.67)  38.9 (0.14)  40.3 (0.50)  39 (0.14)  39.4 (0.28)  38.8 (0.14)  40.1 (0.70)  38.8  39.8  38.9 (0.31)  RR (SD)  RR − 40  72 (18)  59 (2)  85 (7)  56 (1)  75 (6)  54 (5)  60 (24)  55  63  52 (11)  Abnormal respiration1  30  3  0  4  0  2  0  3  0  1  0  Apneustic breathing2  30  0  0  0  0  0  0  1  0  0  0  Mouth breathing3  50  0  0  0  0  0  0  1  0  0  0  Spontaneous cough  40  2  0  4  1  2  2  4  1  1  2  Induced cough4  40  2  0  2  0  2  0  3  0  1  0  Lymphadenopathy5  50  0  0  1  0  0  0  1  1  1  0  Dyspnea6  75  1  0  2  0  0  0  2  0  0  0  Conjunctivitis/eye  30  0  0  3  0  1  0  1  0  1  0  Mucoid–purulent discharge/eye  60  2  1  0  1  0  1  3  0  1  0  Mucoid-purulent discharge/naris  60  2  2  3  0  0  1  0  0  1  3  Frothy salivation  20  0  0  4  0  2  0  1  0  0  1  Decreased rumen fill  100  3  0  4  1  0  0  2  0  0  0  Decreased lung volume/side7  40  3  0  2  0  0  0  3  0  1  0  Abnormal lung sounds/side8  30  3  1  3  0  2  0  4  1  1  1  Crackles/side  80  0  0  0  0  1  0  3  0  0  0  Wheezes/side  60  2  0  1  0  0  0  3  0  0  0  Expiratory grunt9  60  0  0  0  0  0  0  2  0  0  0  Biphasic expiration10  60  0  0  0  0  2  0  3  0  0  0  Total clinical score (SD)  1,472  361 (158)  35 (4)  457 (89)  71 (40)  377 (48)  58 (26)  514 (146)  127  385  97 (73)      Bovine respiratory syncytial virus  Infectious bovine rhinotracheitis  Mannheimia haemolytica  Mycoplasma bovis  Pasteurella multocida  Clinical feature  Points (maximum)  BRD challenge  Control  BRD challenge  Control  BRD challenge  Control  BRD challenge  Control  BRD challenge  Control  Lethargy  60  3  0  4  0  2  0  4  0  0  0  RT, °C (SD)  40 × (RT − 39.5)  39.5 (0.67)  38.9 (0.14)  40.3 (0.50)  39 (0.14)  39.4 (0.28)  38.8 (0.14)  40.1 (0.70)  38.8  39.8  38.9 (0.31)  RR (SD)  RR − 40  72 (18)  59 (2)  85 (7)  56 (1)  75 (6)  54 (5)  60 (24)  55  63  52 (11)  Abnormal respiration1  30  3  0  4  0  2  0  3  0  1  0  Apneustic breathing2  30  0  0  0  0  0  0  1  0  0  0  Mouth breathing3  50  0  0  0  0  0  0  1  0  0  0  Spontaneous cough  40  2  0  4  1  2  2  4  1  1  2  Induced cough4  40  2  0  2  0  2  0  3  0  1  0  Lymphadenopathy5  50  0  0  1  0  0  0  1  1  1  0  Dyspnea6  75  1  0  2  0  0  0  2  0  0  0  Conjunctivitis/eye  30  0  0  3  0  1  0  1  0  1  0  Mucoid–purulent discharge/eye  60  2  1  0  1  0  1  3  0  1  0  Mucoid-purulent discharge/naris  60  2  2  3  0  0  1  0  0  1  3  Frothy salivation  20  0  0  4  0  2  0  1  0  0  1  Decreased rumen fill  100  3  0  4  1  0  0  2  0  0  0  Decreased lung volume/side7  40  3  0  2  0  0  0  3  0  1  0  Abnormal lung sounds/side8  30  3  1  3  0  2  0  4  1  1  1  Crackles/side  80  0  0  0  0  1  0  3  0  0  0  Wheezes/side  60  2  0  1  0  0  0  3  0  0  0  Expiratory grunt9  60  0  0  0  0  0  0  2  0  0  0  Biphasic expiration10  60  0  0  0  0  2  0  3  0  0  0  Total clinical score (SD)  1,472  361 (158)  35 (4)  457 (89)  71 (40)  377 (48)  58 (26)  514 (146)  127  385  97 (73)  1Chest wall excursion on inhaling and exhaling is either overly shallow or deep. 2Pattern of inhalation and exhalation is irregular. 3Open-mouth breathing. 4Induced by encircling the upper trachea with 3 fingers and applying firm pressure. 5Mandibular lymph nodes are enlarged. 6Increased respiratory effort, including outstretched neck and abducted front limbs. 7Air flow cannot be uniformly auscultated across lung field. 8Lung sounds either too harsh/loud (indicating narrow and/or obstructed airways) or too quiet (indicating consolidated area of lung) on auscultation. 9Immediately after expiration, abnormal heaving sound auscultated indicative of abdominal effort. 10Expiratory phase of breathing consists of 2 parts. View Large Table 1. Points awarded and summed for each of the 21 components of the clinical examination to generate a clinical score for 26 steers on the day of peak clinical disease, by pathogen. The mean (SD; for rectal temperature [RT] and respiratory rate [RR]) or the number of steers that had each component is shown. Point ranges are given for those components in which severity was taken into account. For bovine respiratory disease (BRD) challenge steers, peak clinical disease was the day with the highest clinical score. Controls were weight matched to BRD challenge steers and data from the analogous day used     Bovine respiratory syncytial virus  Infectious bovine rhinotracheitis  Mannheimia haemolytica  Mycoplasma bovis  Pasteurella multocida  Clinical feature  Points (maximum)  BRD challenge  Control  BRD challenge  Control  BRD challenge  Control  BRD challenge  Control  BRD challenge  Control  Lethargy  60  3  0  4  0  2  0  4  0  0  0  RT, °C (SD)  40 × (RT − 39.5)  39.5 (0.67)  38.9 (0.14)  40.3 (0.50)  39 (0.14)  39.4 (0.28)  38.8 (0.14)  40.1 (0.70)  38.8  39.8  38.9 (0.31)  RR (SD)  RR − 40  72 (18)  59 (2)  85 (7)  56 (1)  75 (6)  54 (5)  60 (24)  55  63  52 (11)  Abnormal respiration1  30  3  0  4  0  2  0  3  0  1  0  Apneustic breathing2  30  0  0  0  0  0  0  1  0  0  0  Mouth breathing3  50  0  0  0  0  0  0  1  0  0  0  Spontaneous cough  40  2  0  4  1  2  2  4  1  1  2  Induced cough4  40  2  0  2  0  2  0  3  0  1  0  Lymphadenopathy5  50  0  0  1  0  0  0  1  1  1  0  Dyspnea6  75  1  0  2  0  0  0  2  0  0  0  Conjunctivitis/eye  30  0  0  3  0  1  0  1  0  1  0  Mucoid–purulent discharge/eye  60  2  1  0  1  0  1  3  0  1  0  Mucoid-purulent discharge/naris  60  2  2  3  0  0  1  0  0  1  3  Frothy salivation  20  0  0  4  0  2  0  1  0  0  1  Decreased rumen fill  100  3  0  4  1  0  0  2  0  0  0  Decreased lung volume/side7  40  3  0  2  0  0  0  3  0  1  0  Abnormal lung sounds/side8  30  3  1  3  0  2  0  4  1  1  1  Crackles/side  80  0  0  0  0  1  0  3  0  0  0  Wheezes/side  60  2  0  1  0  0  0  3  0  0  0  Expiratory grunt9  60  0  0  0  0  0  0  2  0  0  0  Biphasic expiration10  60  0  0  0  0  2  0  3  0  0  0  Total clinical score (SD)  1,472  361 (158)  35 (4)  457 (89)  71 (40)  377 (48)  58 (26)  514 (146)  127  385  97 (73)      Bovine respiratory syncytial virus  Infectious bovine rhinotracheitis  Mannheimia haemolytica  Mycoplasma bovis  Pasteurella multocida  Clinical feature  Points (maximum)  BRD challenge  Control  BRD challenge  Control  BRD challenge  Control  BRD challenge  Control  BRD challenge  Control  Lethargy  60  3  0  4  0  2  0  4  0  0  0  RT, °C (SD)  40 × (RT − 39.5)  39.5 (0.67)  38.9 (0.14)  40.3 (0.50)  39 (0.14)  39.4 (0.28)  38.8 (0.14)  40.1 (0.70)  38.8  39.8  38.9 (0.31)  RR (SD)  RR − 40  72 (18)  59 (2)  85 (7)  56 (1)  75 (6)  54 (5)  60 (24)  55  63  52 (11)  Abnormal respiration1  30  3  0  4  0  2  0  3  0  1  0  Apneustic breathing2  30  0  0  0  0  0  0  1  0  0  0  Mouth breathing3  50  0  0  0  0  0  0  1  0  0  0  Spontaneous cough  40  2  0  4  1  2  2  4  1  1  2  Induced cough4  40  2  0  2  0  2  0  3  0  1  0  Lymphadenopathy5  50  0  0  1  0  0  0  1  1  1  0  Dyspnea6  75  1  0  2  0  0  0  2  0  0  0  Conjunctivitis/eye  30  0  0  3  0  1  0  1  0  1  0  Mucoid–purulent discharge/eye  60  2  1  0  1  0  1  3  0  1  0  Mucoid-purulent discharge/naris  60  2  2  3  0  0  1  0  0  1  3  Frothy salivation  20  0  0  4  0  2  0  1  0  0  1  Decreased rumen fill  100  3  0  4  1  0  0  2  0  0  0  Decreased lung volume/side7  40  3  0  2  0  0  0  3  0  1  0  Abnormal lung sounds/side8  30  3  1  3  0  2  0  4  1  1  1  Crackles/side  80  0  0  0  0  1  0  3  0  0  0  Wheezes/side  60  2  0  1  0  0  0  3  0  0  0  Expiratory grunt9  60  0  0  0  0  0  0  2  0  0  0  Biphasic expiration10  60  0  0  0  0  2  0  3  0  0  0  Total clinical score (SD)  1,472  361 (158)  35 (4)  457 (89)  71 (40)  377 (48)  58 (26)  514 (146)  127  385  97 (73)  1Chest wall excursion on inhaling and exhaling is either overly shallow or deep. 2Pattern of inhalation and exhalation is irregular. 3Open-mouth breathing. 4Induced by encircling the upper trachea with 3 fingers and applying firm pressure. 5Mandibular lymph nodes are enlarged. 6Increased respiratory effort, including outstretched neck and abducted front limbs. 7Air flow cannot be uniformly auscultated across lung field. 8Lung sounds either too harsh/loud (indicating narrow and/or obstructed airways) or too quiet (indicating consolidated area of lung) on auscultation. 9Immediately after expiration, abnormal heaving sound auscultated indicative of abdominal effort. 10Expiratory phase of breathing consists of 2 parts. View Large A clinical score (CS) was calculated daily for each steer by assigning weights to each component of the examination and then summing them, with a higher score corresponding to more clinically severe BRD (Table 1). Versions of this scoring system have been extensively used by coauthor L. Gershwin in other challenge contexts and shown to successfully capture BRD severity, for example, in terms of both viral burden and extent of respiratory tract pathology (Jordan et al., 2015). The CS was used in determining which days of all available data to include in all analyses on a steer-by-steer basis. Three days of data were chosen from BRD challenge steers: the day of peak clinical signs (d 0), defined as that with the highest CS, and the 2 d preceding (d −1 and −2). This was expected to be the time period including and just before peak sickness response, in preparation for future studies that would focus on earlier time points when disease was less clinically evident. If the highest CS was assigned on the morning of necropsy, then data from the day immediately before were used for d 0. For Control steers, the analogous days were chosen for analysis as their weight-matched BRD challenge steer within their pathogen trial. Only BRD challenge steers were killed (penetrating captive bolt followed by exsanguination) on the morning after the monitoring period elapsed, an average of 3.2 d (range 1–11 d) after peak CS. A single steer challenged with Mycoplasma bovis developed severe respiratory distress and was euthanized 5 d after challenge using an overdose of intravenous phenobarbital. The proportion of grossly affected lung (%LUNG) was estimated using visual observation and palpation by the veterinary pathologist. Rectal Temperature – Physiological Aspect of the Sickness Response An aluminum device containing a temperature logger (TidbiT v2 number UTBI-001; Onset Computer Corp., Pocasset, MA), as previously described (Reuter et al., 2010), recorded RT every 5 min starting 48 h before challenge. The tail and rectum were rinsed with water and assessed daily, and the devices were removed if steers had evidence of irritation. For Mycoplasma bovis, the devices were removed between d 7 and 13 relative to challenge to prevent irritation from prolonged use. To avoid the RT readings being confounded by high daytime ambient temperatures, 73 readings taken between 0100 and 0700 h were averaged for analysis. Average ambient temperature between 0100 and 0700 h was 15.2°C (range 12.5–24.0°C), with a minimum of 9.1°C and maximum of 26.0°C. Feeding Time – Behavioral Aspect of the Sickness Response A starter ration (47.5% flaked corn, 17.2% dried distiller's grains, 13.7% alfalfa hay, 11.7% oat hay, 7.1% molasses, and 1.3% fat on an as-fed basis containing 87.5% DM, 13.3% CP, 27.7% NDF, and 75.8% TDN) was fed in head-gated feed bunks. Each of the 2 feed bunks was 1.35 m long and had 2 head gates, such that all 4 steers could feed simultaneously. The total amount fed/day was 4% of the BW of all steers in the pen (40% fed at 0700–0730 h and 60% at 1630–1700 h). All leftover feed was removed immediately before the morning feeding, and water was provided ad libitum from a self-filling trough. Feeding time (Table 2) was calculated for individual steers using 5-min instantaneous video scan samples with 1/0 scoring, on a 24-h basis. Table 2. Ethogram for recording feeding and grooming behavior (contact with a brush and self-licking) Behavior  Description  Feeding time  Both ears or the head/neck past the head gate bars of the feed bunk  Brush contact  Touching of any body part with the brush. While touching, brush could be either stationary or moving.  Self-licking  Turning the muzzle/head/neck toward any other body part followed by up and down or back and forth movements of the muzzle/head/neck lasting at least 1.0 s. A bout began when repetitive movements commence and ends when the muzzle/head/neck start to turn away from the body part.  Behavior  Description  Feeding time  Both ears or the head/neck past the head gate bars of the feed bunk  Brush contact  Touching of any body part with the brush. While touching, brush could be either stationary or moving.  Self-licking  Turning the muzzle/head/neck toward any other body part followed by up and down or back and forth movements of the muzzle/head/neck lasting at least 1.0 s. A bout began when repetitive movements commence and ends when the muzzle/head/neck start to turn away from the body part.  View Large Table 2. Ethogram for recording feeding and grooming behavior (contact with a brush and self-licking) Behavior  Description  Feeding time  Both ears or the head/neck past the head gate bars of the feed bunk  Brush contact  Touching of any body part with the brush. While touching, brush could be either stationary or moving.  Self-licking  Turning the muzzle/head/neck toward any other body part followed by up and down or back and forth movements of the muzzle/head/neck lasting at least 1.0 s. A bout began when repetitive movements commence and ends when the muzzle/head/neck start to turn away from the body part.  Behavior  Description  Feeding time  Both ears or the head/neck past the head gate bars of the feed bunk  Brush contact  Touching of any body part with the brush. While touching, brush could be either stationary or moving.  Self-licking  Turning the muzzle/head/neck toward any other body part followed by up and down or back and forth movements of the muzzle/head/neck lasting at least 1.0 s. A bout began when repetitive movements commence and ends when the muzzle/head/neck start to turn away from the body part.  View Large Grooming – Behavioral Aspect of the Sickness Response Brush Contact. An automated grooming brush (model number 91526202 swinging cow brush; DeLaval Inc., Kansas City, MO) was wall mounted per manufacturer instructions. The brush hung vertically from a pivoting arm from which it could move in many directions. When an animal pushed the brush off center to a ≥30° angle, it began rotating at a speed of 26 rotations/min. Rotation continued until 10 s after returning to a vertical position, after the steer was no longer pushing the brush. Brush contact was recorded to the nearest second from video using continuous observations (Table 2) for a 13-h period each day: 0100 to 0300, 0700 to 0800, 1100 to 1400, 1500 to 1700, 1800 to 2100, and 2200 to 0000 h. These times were chosen after regression analysis, comparing 24-h durations with various subsets of hours. Based on this analysis (n = 40), the selected 13-h period was an accurate representation of the entire 24 h (R2 = 0.880, with an intercept of 0 [P = 0.970] and slope of 1 [P = 0.113]). Self-Licking. Self-licking was recorded continuously from video to the nearest second for a 24-h period on d 0 (Table 2). Steer positioning and lighting conditions precluded definitive determination of whether a steer was self-licking for varying amounts of time during this 24-h period (between 0.5 to 107 min/steer, with an average of 27 min not visible). Analyses Penwise Mixed Model Analysis: Evaluating the Usefulness of Fever, Feeding and Grooming Around Peak Disease as Diagnostic Measures. Treatment (BRD challenge or Control) was applied at the group level; therefore, pen was considered the experimental unit (hereafter called mixed model analysis). For this analysis, 8 inclusion criteria, derived from the clinical examination, were used to ensure that evaluation of treatment effects involved only animals with BRD clinical signs (BRD challenge) and healthy (Control) steers. This was done to best describe how the sickness response manifests in this population and determine the usefulness of 3 aspects of the sickness response as diagnostic measures. The inclusion criteria included 1) RT > 39.5°C, 2) spontaneous coughing, 3) induced coughing, 4) lethargy, 5) wheezing during auscultation, 6) respiratory rate (RR) > 70/min, 7) purulent nasal discharge, and 8) moderately to severely decreased rumen fill. These components were tallied to create an inclusion criteria score for each of the 3 d intended for analysis, by steer. In cases where a steer had both elevated RR and RT, only a single component was counted to account for possible collinearity between these 2 clinical signs due to factors (e.g., heat stress) unrelated to respiratory disease. Steers were considered to fulfill inclusion criteria on a given day if they scored ≥ 3 (BRD challenge) or ≤ 1 (Control; Table 3). Data were then averaged from included steers, resulting in a pen average for each day. Table 3. Count of bovine respiratory disease (BRD) challenge and Control steers included in mixed model analysis with pen as the experimental unit. Steer numbers are shown by day relative to peak disease, on a per-pathogen basis (maximum possible number of steers/pen = 4). The total number of pens analyzed per day is shown in the last column. The number of animals included in pen average for rectal temperature in cases where additional data were missing due to removal of the measurement device is in parenthesis   Pathogen  Relative day  Bovine respiratory syncytial virus  Infectious bovine rhinotracheitis  Mannheimia haemolytica  Mycoplasma bovis  Pasteurella multocida  Total no. of pens analyzed  BRD challenge      −2  1  3  0  4 (3)  0  3      −1  1  4  2  4 (2)  0  4      0  3  4  2  4 (3)  1  5  Control      −2  3  1  4  1  4  5      −1  3  2  4  1  3  5      0  2  2  4  1  3  5    Pathogen  Relative day  Bovine respiratory syncytial virus  Infectious bovine rhinotracheitis  Mannheimia haemolytica  Mycoplasma bovis  Pasteurella multocida  Total no. of pens analyzed  BRD challenge      −2  1  3  0  4 (3)  0  3      −1  1  4  2  4 (2)  0  4      0  3  4  2  4 (3)  1  5  Control      −2  3  1  4  1  4  5      −1  3  2  4  1  3  5      0  2  2  4  1  3  5  View Large Table 3. Count of bovine respiratory disease (BRD) challenge and Control steers included in mixed model analysis with pen as the experimental unit. Steer numbers are shown by day relative to peak disease, on a per-pathogen basis (maximum possible number of steers/pen = 4). The total number of pens analyzed per day is shown in the last column. The number of animals included in pen average for rectal temperature in cases where additional data were missing due to removal of the measurement device is in parenthesis   Pathogen  Relative day  Bovine respiratory syncytial virus  Infectious bovine rhinotracheitis  Mannheimia haemolytica  Mycoplasma bovis  Pasteurella multocida  Total no. of pens analyzed  BRD challenge      −2  1  3  0  4 (3)  0  3      −1  1  4  2  4 (2)  0  4      0  3  4  2  4 (3)  1  5  Control      −2  3  1  4  1  4  5      −1  3  2  4  1  3  5      0  2  2  4  1  3  5    Pathogen  Relative day  Bovine respiratory syncytial virus  Infectious bovine rhinotracheitis  Mannheimia haemolytica  Mycoplasma bovis  Pasteurella multocida  Total no. of pens analyzed  BRD challenge      −2  1  3  0  4 (3)  0  3      −1  1  4  2  4 (2)  0  4      0  3  4  2  4 (3)  1  5  Control      −2  3  1  4  1  4  5      −1  3  2  4  1  3  5      0  2  2  4  1  3  5  View Large The mixed model analysis was performed with PROC MIXED (version 9.3 of the SAS System for Windows; SAS Inst. Inc., Cary, NC), with a variance components covariate structure, REML method of model fitting, and Satterthwaite method for estimation of degrees of freedom. The model included the fixed effects of treatment (BRD challenge or Control), day relative to peak disease, and their interactions and the random effect of pen, nested within treatment, for each of the sickness response variables (RT, feeding time, brush contact, and self-licking). The model for self-licking did not include the effect of day relative to peak disease as only 24 h of data were collected for each steer. Model fit and assumptions were checked; in the event of nonnormality (self-licking), appropriate transformation (square root) was used and the data were back-transformed before graphing. Regression with Individual Steers Analysis: Understanding How Sickness Response Expression Varies Across the Clinical Spectrum. After examining clinical signs in all animals, it was clear that there was a large degree of variability among pathogen trials, which was captured by 2 continuous variables: CS and %LUNG. This variability within BRD challenge steers was expected given the differing pathogenesis and clinical courses of the various challenge organisms, although variability was also seen in Control steers. For this reason, 2 regression analyses were conducted using all available individual steer data (no inclusion criteria were applied). These analyses used PROC REG in SAS, separately for each of the 3 d analyzed, to evaluate the effect of clinical and pathological severity on sickness response expression: Clinical severity: relationship between daily CS and fever, feeding, and grooming, including individual-level data from all steers Pathological severity: relationship between %LUNG and fever, feeding, and grooming, including individual-level data from all BRD challenge steers These analyses accounted for the fact that differences in the dependent variables—fever, anorexia, and grooming—could also be attributed to BRD severity, which was not captured in the penwise analysis of treatment effects (BRD challenge vs. Control). Finally, additional regression analyses were performed using PROC REG to determine 1) the relationship between the 2 measures of disease severity (CS and %LUNG), in only BRD challenge steers, and 2) the relationship between the 2 grooming measures (brush contact and self-licking duration), using data from all steers. The P-value associated with the coefficient of determination (R2) was determined. In all cases, a P-value of ≤ 0.05 was considered significant and 0.05 < P ≤ 0.09 was considered a tendency. Excluded Animals and Missing Data. Five Controls were excluded from all analyses because they developed clinical signs of spontaneous BRD. A variable amount of data was lost over the course of the study because of video failure (affecting a total of 9 d on trial; range 4–427 min/d), and specifically for self-licking, there were times that individual steers were not visible, as described earlier. To account for the possible effect of missing data, all behavioral data were initially analyzed both as total duration and as proportion of observable time. However, no differences were found between the outcomes of these analyses, and results are, therefore, presented as total duration. Reliability. For feeding time, interobserver reliability for each pair of observers was determined by the kappa coefficient of concordance (Martin and Bateson, 2007). The mean and ranges of reliability scores for pairs of observers was κ = 0.94 (0.83–1.0). For both brush contact and self-licking, inter- and intraobserver reliability was calculated using a regression analysis. For brush contact and self-licking, interobserver reliability had an R2 ≥ 0.97, an intercept of 0 (P ≥ 0.10), and slope of 1 (P ≥ 0.16) and intraobserver reliability had an R2 ≥ 0.92, intercept of 0 (P ≥ 0.18), and slope of 1 (P ≥ 0.09). RESULTS Rectal Temperature and Feeding Time Steers challenged with BRD had higher RT (Fig. 1A; treatment, F(1, 8) = 26.19, P ≤ 0.01) and lower feeding time (Fig. 1B; treatment, F(1, 8) = 6.72, P = 0.03) than Controls. The effect of treatment was consistent during the observed days for RT (treatment × day, F(1, 14) = 0.43, P = 0.66) and feeding time (treatment × day, F(1, 14) = 0.09, P = 0.92). Both RT (day, F(1, 14) = 1.94, P = 0.18) and feeding time (day, F(1, 14) = 0.49, P = 0.62) did not change over the 3 d. Figure 1. View largeDownload slide (A) Average rectal temperature, between 0100 and 0700 h (total of 73 readings, taken every 5 min), and (B) feeding time in pens of bovine respiratory disease (BRD)–challenged (BRD challenge) and Control steers on day of peak clinical disease (d 0) and the 2 preceding days. Values are least squares means; error bars represent SEM. *P < 0.05 (fixed effect of treatment). Figure 1. View largeDownload slide (A) Average rectal temperature, between 0100 and 0700 h (total of 73 readings, taken every 5 min), and (B) feeding time in pens of bovine respiratory disease (BRD)–challenged (BRD challenge) and Control steers on day of peak clinical disease (d 0) and the 2 preceding days. Values are least squares means; error bars represent SEM. *P < 0.05 (fixed effect of treatment). Clinical scores explained a large portion of the variability in RT (Fig. 2A, 2B, and 2C), and feeding time (Fig. 2D, 2E, and 2F) on all days, with the clearest relationship on d 0 for RT and on both d 0 and d −1 for feeding time. Increased clinical severity was associated with higher RT and lower feeding time relative to healthier animals. Figure 2. View largeDownload slide Relationship between clinical score (x-axis) and, on the y-axis, average rectal temperature (between 0100 and 0700 h; total of 73 readings, taken every 5 min) on d −2 (A), −1 (B), and 0 (C) and feeding time on d −2 (D), −1 (E), and 0 (F), in 35 bovine respiratory disease–challenged and Control steers. Peak disease (d 0) was determined as the day with the highest clinical score. Figure 2. View largeDownload slide Relationship between clinical score (x-axis) and, on the y-axis, average rectal temperature (between 0100 and 0700 h; total of 73 readings, taken every 5 min) on d −2 (A), −1 (B), and 0 (C) and feeding time on d −2 (D), −1 (E), and 0 (F), in 35 bovine respiratory disease–challenged and Control steers. Peak disease (d 0) was determined as the day with the highest clinical score. In BRD challenge steers, the extent of gross lung lesions explained variation in RT (Fig. 3A, 3B, and 3C) and feeding time (Fig. 3D, 3E, and 3F). On d 0, as %LUNG increased, RT was higher and feeding time was lower. Proportion of grossly affected lung tended to explain the variability in feeding time, but not RT, on d −1. On d −2, there was no relationship between %LUNG and RT or feeding time. Figure 3. View largeDownload slide Relationship between percent lung affected by lesions as determined at necropsy (x-axis) and, on the y-axis, average rectal temperature (between 0100 and 0700 h; total of 73 readings, taken every 5 min) on d −2 (A), −1 (B), and 0 (C) and feeding time on d −2 (D), −1 (E), and 0 (F), in 20 bovine respiratory disease–challenged steers. Figure 3. View largeDownload slide Relationship between percent lung affected by lesions as determined at necropsy (x-axis) and, on the y-axis, average rectal temperature (between 0100 and 0700 h; total of 73 readings, taken every 5 min) on d −2 (A), −1 (B), and 0 (C) and feeding time on d −2 (D), −1 (E), and 0 (F), in 20 bovine respiratory disease–challenged steers. Brush Contact and Self-Licking There was no difference in duration of brush contact (Fig. 4; treatment, F(1, 8) = 0.90, P = 0.37) between BRD challenge and Control steers. There was also no difference in duration of self-licking (4.8 vs. 8.4 min/24 h on d 0; back-transformed least squares means) between BRD challenge and Control steers (treatment, F(1, 8) = 2.47, P = 0.15). Additionally, brush contact was stable over the 3 d (day, F(1, 14) = 0.87, P = 0.44), and there were no day-to-day differences in brush contact as a function of treatment (treatment × day, F(1, 14) = 0.55, P = 0.59). Figure 4. View largeDownload slide Average brush contact duration in bovine respiratory disease (BRD)–challenged (BRD challenge) and Control steers on day of peak clinical disease (d 0) and the 2 preceding days. Values are least squares means; error bars represent SEM. Figure 4. View largeDownload slide Average brush contact duration in bovine respiratory disease (BRD)–challenged (BRD challenge) and Control steers on day of peak clinical disease (d 0) and the 2 preceding days. Values are least squares means; error bars represent SEM. On d 0, clinical severity explained a significant portion of the variation in brush contact (Fig. 5C) and self-licking (Fig. 6A) duration. Generally, the higher the CS, the lower the brush contact duration. On both days before peak disease (Fig. 5A and 5B), however, variation in brush contact could not be explained by CS. Figure 5. View largeDownload slide Relationship between clinical score (x-axis) and, on the y-axis, average duration of brush contact on d −2 (A), −1 (B), and 0 (C) in 35 bovine respiratory disease–challenged and Control steers. Peak disease (d 0) was determined as the day with the highest clinical score. Figure 5. View largeDownload slide Relationship between clinical score (x-axis) and, on the y-axis, average duration of brush contact on d −2 (A), −1 (B), and 0 (C) in 35 bovine respiratory disease–challenged and Control steers. Peak disease (d 0) was determined as the day with the highest clinical score. Figure 6. View largeDownload slide Relationship between (A) clinical score (n = 35) and (B) percent lung affected by lesions as determined at necropsy (n = 20; x-axis) and average duration of self-licking (y-axis) on day of peak disease in bovine respiratory disease–challenged and Control steers. Peak disease was determined as the day with the highest clinical score. Figure 6. View largeDownload slide Relationship between (A) clinical score (n = 35) and (B) percent lung affected by lesions as determined at necropsy (n = 20; x-axis) and average duration of self-licking (y-axis) on day of peak disease in bovine respiratory disease–challenged and Control steers. Peak disease was determined as the day with the highest clinical score. On d 0 (and d −1 for brush contact), extent of gross lung lesions explained a significant portion of the variation in brush contact (Fig. 7B and 7C) and self-licking (Fig. 6B) in BRD challenge steers. As %LUNG increased, brush contact and self-licking duration decreased. However, on d −2, %LUNG did not explain a significant portion of the variation in brush contact (Fig. 7A). Figure 7. View largeDownload slide Relationship between percent lung affected by lesions as determined at necropsy (x-axis) and, on the y-axis, brush contact duration on d −2 (A), −1 (B), and 0 (C) in 20 bovine respiratory disease–challenged steers. Figure 7. View largeDownload slide Relationship between percent lung affected by lesions as determined at necropsy (x-axis) and, on the y-axis, brush contact duration on d −2 (A), −1 (B), and 0 (C) in 20 bovine respiratory disease–challenged steers. Relationship between Grooming Variables On d 0, no relationship was observed between the duration of brush contact and self-licking (R2 = 0.008; n = 35; P = 0.61). Relationship between Clinical Score and Percent Lung Affected Among BRD challenge steers on d 0, the higher the steers' CS, the higher the %LUNG (Fig. 8), such that variation in CS explained a significant portion of the variation in %LUNG. Figure 8. View largeDownload slide Relationship between clinical score (x-axis) and percent lung affected by lesions as determined at necropsy (y-axis) on day of peak disease in 20 bovine respiratory disease–challenged steers. Peak disease was determined as the day with the highest clinical score. Figure 8. View largeDownload slide Relationship between clinical score (x-axis) and percent lung affected by lesions as determined at necropsy (y-axis) on day of peak disease in 20 bovine respiratory disease–challenged steers. Peak disease was determined as the day with the highest clinical score. DISCUSSION Bovine respiratory disease challenge steers that developed clinical disease had fever (1.1°C higher) and anorexia (35% lower feeding time) but no difference in brush contact or self-licking compared with healthy Controls. Steers with severe BRD, as indicated by both higher CS and %LUNG, were associated with higher fever, lower feeding time, and less brush contact and self-licking relative to healthier individuals. In general, this relationship was clearer on d 0 than on d −1 or −2 relative to peak BRD. Although CS and %LUNG had a moderately positive relationship with one another, there was, unexpectedly, no relationship between brush contact and self-licking duration. Value of Fever, Anorexia, and Grooming as Diagnostic Measures Earlier studies have often relied solely on visual appraisal of clinical signs, which is considered inaccurate (White and Renter, 2009), to classify cattle as healthy or sick without providing further evidence of health (e.g., Sowell et al., 1999; White et al., 2012). However, evidence of respiratory tract involvement is necessary to confirm BRD as the underlying cause of the sickness response, versus another disease process. Indeed, the sickness response, a type of innate immune mechanism, indicates presence of inflammation but is not diagnostic for a specific disease (Dantzer, 2004). To confine mixed model analysis to those steers that had clinical signs of BRD or, conversely, were clinically healthy, 8 inclusion criteria were applied on a daily basis. Five of these inclusion criteria components (coughing, both spontaneous and induced; purulent nasal discharge; fever; and abnormal respiration, including elevated RR), when evaluated similarly to the methodology used in the current study, have been demonstrated to have high level of accuracy in correctly classifying BRD in dairy calves whose disease status was confirmed with pathogen isolation (Love et al., 2014). Three additional aspects (wheezing, lethargy, and rumen fill) were also considered as part of the inclusion criteria, as they are also clinical signs that are commonly observed in BRD-affected cattle (Curtis et al., 1986; Apley, 2006). The purpose of these inclusion criteria was not to define a commercially relevant diagnostic scheme but rather to evaluate, on a daily basis, the likelihood that an individual had BRD based on clinical signs. The CS was intended to be a comprehensive measure of BRD clinical severity and included respiratory-specific clinical signs such as nasal discharge, coughing, and abnormal respiration and sickness-response changes such as fever, decreased rumen fill (anorexia), and lethargy. Others (e.g., Collie, 1992; Schaefer et al., 2007; Love et al., 2014) have also used a combination of both respiratory-specific factors (e.g., coughing, breathing effort) and sickness response components (e.g., anorexia, fever, lethargy) to diagnose BRD and describe its severity, in addition to postmortem findings (e.g., Amrine et al., 2013). Based on the clear relationship between CS and %LUNG, the clinical measures used in this study appear to have adequately captured the underlying pathological processes, at least in the challenged steers for which these data were available. Our finding that BRD challenge steers that developed clinical signs had both fever and lower feeding time than healthy Controls agrees with previous work that found that cattle with clinical signs of BRD spent less time at the feed bunk (Sowell et al., 1999; Buhman et al., 2000) and had a fever of magnitude similar to that observed here (as reviewed by Schaefer et al., 2007; Adams et al., 2013). It provides further evidence that these are both good diagnostic indicators for this disease, in that consistent differences were observed between clinically affected and healthy steers at and during both days before peak sickness. Validating early diagnostic indicators is important, as early detection improves response to treatment (as reviewed by Griffin, 2014; Wolfger et al., 2015) and attending to BRD-affected animals in a timely manner is also important from an animal welfare perspective (Millman, 2007; Beausoleil and Mellor, 2015). Fever is metabolically expensive, and simultaneous anorexia reduces energy intake (Hart, 1988). An evolutionary perspective suggests that natural selection has shaped the sickness response so that its benefits—promoting recovery from infection—outweigh the costs (as reviewed by Dantzer, 2004; Rauw, 2012). It follows that to achieve energy balance, sick animals may need to limit even those behaviors that are important for long-term fitness if they are not essential for short-term survival (Aubert, 1999). Grooming is also expected to decrease in sickness as part of an energy-conservation strategy (Hart, 1988) and potentially could be a good BRD indicator. Self-grooming has hygienic, thermoregulatory, sensory stimulation, and stress-relieving effects (Spruijt et al., 1992). Cattle self-lick and rub on objects (Simonsen, 1979; Ishiwata et al., 2008), including purpose-designed brushes (Wilson et al., 2002; DeVries et al., 2007; Georg et al., 2007), which they will compete with one another to access (Val-Laillet et al., 2008). Although average durations of brush contact and self-licking in healthy steers were similar to those reported for pastured cattle (tree grooming and self-licking 22 and 7 min/d, respectively; Kohari et al., 2007), a large degree of variability in both grooming behaviors was observed even in Controls, ranging from 0 to 86 min of brush contact/d and from 2 to 36 min self-licking/d. The lack of differences in self-licking between BRD challenge and Controls in the current study was unexpected, as an earlier study found that BRD cattle groom less than healthy animals (Toaff-Rosenstein, 2016). Sick rats (Yirmiya et al., 1994; Tikhonova et al., 2011), dairy cows (Fogsgaard et al., 2012), calves (Borderas et al., 2008), and goats (Takeuchi and Mori, 1995) also self-lick less when sick. Perhaps use of an automated brush is not energetically expensive, unlike our earlier BRD study in which the device did not rotate and required animals to actively rub themselves to groom. Competition for brush access, as has been observed in cows (Val-Laillet et al., 2008), might also require additional physical exertion, but no attempt was made to quantify the energy investment required for grooming in this experiment. Regardless, this does not explain why self-licking did not decrease in BRD challenge steers, which may be explained by changes in the feeding regimen during the transition between pasture grazing and trial conditions. Cattle may perform more oral behaviors, including self-licking, to compensate for time-budget differences when fed a mixture of concentrate and hay (a lower proportion of time spent foraging) compared with pasture grazing (Ishiwata et al., 2008). All cattle in the current study may have increased self-licking behavior as they were fed a concentrate-based diet from a feed bunk. Unexpectedly, there was also no relationship between duration of brush contact and self-licking, but it is unclear why. There are a few additional methodological possibilities as to why no treatment differences were observed in grooming behavior between BRD challenge steers and Controls. First, the inclusion criteria for the penwise analysis involved measures of fever and anorexia (decreased rumen fill) but nothing for grooming behavior (e.g., rough or dirty hair coat). These inclusion criteria may have inadvertently favored finding treatment differences for fever and feeding time but not for grooming, because aspects related to brush contact or self-licking were not used in determining which animals were included in the mixed model analysis. Importantly, applying the inclusion criteria as well as choosing days to analyze based on the CS (which also incorporated aspects of the sickness response) likely increased the magnitude of treatment differences, relative to a population in which these factors were not considered. Additionally, although daily ringworm treatment was provided, intended to alleviate skin irritation, it is unfortunately possible that this condition contributed to grooming variability. In the future, it is of interest to further explore grooming behavior during sickness, recognizing that the energy investment required to groom may be grooming-object and stocking-density dependent and that these may be important factors to consider when interpreting results. Presence of skin conditions may also be confounding factors (e.g., oral grooming increases with greater tick loads in impala; Mooring et al., 1996), although we were not able to explicitly examine the effect of ringworm in the current study as severity was not quantified, daily treatment was provided, and the condition was common (86% of animals). Relationship between Fever, Anorexia, and Grooming and Bovine Respiratory Disease Severity Given that BRD is a multifactorial condition with many causative pathogens, it is highly variable in both its clinical and its postmortem presentation (Fulton et al., 2009; Panciera and Confer, 2010; Gershwin et al., 2015). It involves a complicated interplay of host–pathogen interactions, determined, in part, by the specific microbe or microbes involved (Czuprynski, 2009; Fulton et al., 2009). Even after single-pathogen challenge, clinical responses range from mild to severe and minimal to extensive lung lesions (Reeve-Johnson, 2001; White et al., 2012). Because multiple challenge pathogens were used in the current study, the observed variation in clinical and pathological outcomes was expected. This experimental design potentially complicated interpretation—for example, by resulting in variable presentation and progression of clinical and pathological signs. However, it also provided an opportunity to better understand the relationship between sickness response components and BRD severity and perhaps better mimic the clinical variability observed in spontaneous disease, given that BRD may result from a multitude of pathogens. Two measures, CS and %LUNG, were used to describe BRD severity from a clinical and pathological perspective, respectively. Their relationship to one another was expected, and an earlier study also found that worse lung pathology was correlated with a higher CS (Reeve-Johnson, 2001). It follows that the greater the pulmonary inflammation, which includes the release of mediators responsible for the sickness response (Czuprynski, 2009; N'jai et al., 2013), the more marked the respiratory-specific clinical signs (e.g., RR, coughing, wheezes) will be. Perhaps one of the reasons for the apparently poor specificity of clinical observations in BRD detection, at least when lung pathology is used to diagnose cases (White and Renter, 2009), is that BRD-associated inflammation and clinical signs may result not only from pulmonary lesions but also from involvement of the upper respiratory tract, as in BRSV and IBR (Spilki et al., 2004; Gershwin et al., 2015). Indeed, 2 of the individuals that appeared to be outliers in the regression between CS and %LUNG were, in fact, IBR-challenged steers (both had 5% gross lung lesions but relatively high CS), supporting this idea. Clinical scores explained the variation in some components of the sickness response but not others. Rectal temperature had a clear relationship with CS on all 3 d, especially on the day of peak BRD. However, the relationship between CS and feeding time was moderate and between both brush contact and self-licking was weak. In support of these findings, earlier studies also found body temperature to be a good measure of clinical BRD (Reeve-Johnson, 2001; Schaefer et al., 2007; Adams et al., 2013) and described a relationship between feeding time and clinical severity (White et al., 2012). Similarly, a clear relationship between grooming and CS was also expected, given that calves with worse clinical signs of BRD are more lethargic (moved less distance) relative to healthy animals (White et al., 2012) and assuming that decreases in brush contact and self-licking are, in part, due to energy conservation. However, this expected relationship was not demonstrated in the current study. As suggested for the penwise analysis, the relationship between CS and both RT and feeding time was likely strengthened, in part, because points from fever and rumen fill were incorporated in the CS, but there was no analogous accounting for brush contact or self-licking. Additionally, the relationship between CS and RT in particular might have been strengthened because only readings from between 0100 and 0700 h were used. Clinical examinations were performed every morning, shortly afterward, so that temporally, CS and RT were more closely related than were measures taken over the entire 24-h period. The relationships between %LUNG, a second measure of BRD severity, and the various sickness response components varied from those observed for CS, in that feeding time, brush contact, and self-licking duration all showed a higher R2 with %LUNG than they did with CS. Feeding time had a particularly clear relationship with %LUNG, especially on the day of peak BRD, but it is not known why extent of gross lung lesions was able to explain the greatest amount of variation for this sickness response component relative to the others. In contrast to the other 3 measures, the relationship between RT and %LUNG was less clear than it was for CS, even on the day of peak BRD. In one study, cattle with more pulmonary pathology at necropsy had a higher fever than those with milder lesions (Reeve-Johnson, 2001), but the current work did not support these findings. Perhaps one explanation is that although, on the one hand, a higher fever can result in better pathogen clearance and, therefore, fewer lung lesions, on the other hand, it also can cause more lung damage due to widespread recruitment of white blood cells and release of oxygen radicals (Evans et al., 2015; Harden et al., 2015). Finally, calves with a higher proportion of consolidated lung tissue are more lethargic relative to healthier animals (White et al., 2012), supporting the moderate R2 value observed between %LUNG and brush contact. It is not clear, however, why a similar R2 was not observed for self-licking. Taken together, the individual steer regression findings suggest that when there are milder clinical signs and less extensive gross lung lesions, fever, anorexia, and less grooming may be more difficult to discern than in cattle with worse clinical and pathological abnormalities. Others have also found that accuracy of clinical scoring improves with more extensive pulmonary lesions, providing evidence that BRD clinical signs are clearer in sicker cattle (Amrine et al., 2013). It is still unclear, however, at exactly which point in the continuum of clinical and pulmonary severity there is an observable sickness response. Indeed, BRD is primarily detected, both in experimental work as well as in the field, based on visually assessed clinical impression and severity is determined retrospectively based on relapse rate or mortality (Wolfger et al., 2015), neither of which elucidates actual extent of disease at the time of initial diagnosis. Additionally, postmortem confirmation is often available only months later and usually in treated animals (Wolfger et al., 2015), such that the remaining pathological lesions may not be representative of the initial ones. In conclusion, both fever and feeding time appear to be good diagnostic measures at the peak of BRD clinical signs and the 2 d before, with consistent differences observed between BRD challenge steers and Controls. There was also a clear relationship between both fever and feeding to clinical signs and lung lesions. This provides clinical and pathological evidence that they are sensitive to changes in BRD severity. Brush contact is moderately correlated to lung lesions, but despite this relationship, grooming generally appears less promising as a diagnostic measure. Regardless of which component is monitored, cattle with milder disease may be more difficult to detect than sicker animals. LITERATURE CITED Adams A. E. Olea-Popelka F. J. Roman-Muniz I. N. 2013. Using temperature-sensing reticular boluses to aid in the detection of production diseases in dairy cows. J. Dairy Sci.  96: 1549– 1555. doi: https://doi.org/10.3168/jds.2012-5822 Google Scholar CrossRef Search ADS PubMed  Adelman J. S. Martin L. B. 2009. Vertebrate sickness behaviors: Adaptive and integrated neuroendocrine immune responses. Integr. Comp. Biol.  49: 202– 214. doi: https://doi.org/10.1093/icb/icp028 Google Scholar CrossRef Search ADS PubMed  Amrine D. E. White B. J. Larson R. Anderson D. E. Mosier D. A. Cernicchiaro N. 2013. Precision and accuracy of clinical illness scores, compared with pulmonary consolidation scores, in Holstein calves with experimentally induced Mycoplasma bovis pneumonia. Am. J. Vet. Res.  74: 310– 315. doi: https://doi.org/10.2460/ajvr.74.2.310 Google Scholar CrossRef Search ADS PubMed  Apley M. 2006. Bovine respiratory disease: Pathogenesis, clinical signs, and treatment in lightweight calves. Vet. Clin. North Am. Food Anim. Pract.  22: 399– 411. doi: https://doi.org/10.1016/j.cvfa.2006.03.009 Google Scholar CrossRef Search ADS PubMed  Aubert A. 1999. Sickness and behaviour in animals: A motivational perspective. Neurosci. Biobehav. Rev.  23: 1029– 1036. doi: https://doi.org/10.1016/S0149-7634(99)00034-2 Google Scholar CrossRef Search ADS PubMed  Beausoleil N. J. Mellor D. J. 2015. Introducing breathlessness as a significant animal welfare issue. N. Z. Vet. J.  63: 44– 51. doi: https://doi.org/10.1080/00480169.2014.940410 Google Scholar CrossRef Search ADS PubMed  Borderas T. F. de Passillé A. M. Rushen J. 2008. Behavior of dairy calves after a low dose of bacterial endotoxin. J. Anim. Sci.  86: 2920– 2927. doi: https://doi.org/10.2527/jas.2008-0926 Google Scholar CrossRef Search ADS PubMed  Buhman M. J. Perino L. J. Galyean M. L. Wittum T. E. Montgomery T. H. Swingle R. S. 2000. Association between changes in eating and drinking behaviors and respiratory tract disease in newly arrived calves at a feedlot. Am. J. Vet. Res.  61: 1163– 1168. doi: https://doi.org/10.2460/ajvr.2000.61.1163 Google Scholar CrossRef Search ADS PubMed  Collie D. D. 1992. Pulmonary function changes and clinical findings associated with chronic respiratory disease in calves. Br. Vet. J.  148: 33– 40. doi: https://doi.org/10.1016/0007-1935(92)90064-8 Google Scholar CrossRef Search ADS PubMed  Curtis R. A. Viel L. McGuirk S. M. Radostits O. M. Harris F. W. 1986. Lung sounds in cattle, horses, sheep and goats. Can. Vet. J.  27: 170– 172. Google Scholar PubMed  Czuprynski C. J. 2009. Host response to bovine respiratory pathogens. Anim. Health Res. Rev.  10: 141– 143. doi: https://doi.org/10.1017/S1466252309990181 Google Scholar CrossRef Search ADS PubMed  Dantzer R. 2004. Cytokine-induced sickness behaviour: A neuroimmune response to activation of innate immunity. Eur. J. Pharmacol.  500: 399– 411. doi: https://doi.org/10.1016/j.ejphar.2004.07.040 Google Scholar CrossRef Search ADS PubMed  DeVries T. J. Vankova M. Veira D. M. von Keyserlingk M. A. G. 2007. Short communication: Usage of mechanical brushes by lactating dairy cows. J. Dairy Sci.  90: 2241– 2245. doi: https://doi.org/10.3168/jds.2006-648 Google Scholar CrossRef Search ADS PubMed  Evans S. S. Repasky E. A. Fisher D. T. 2015. Fever and the thermal regulation of immunity: The immune system feels the heat. Nat. Rev. Immunol.  15: 335– 349. doi: https://doi.org/10.1038/nri3843 Google Scholar CrossRef Search ADS PubMed  Fogsgaard K. K. Røntved C. M. Sørensen P. Herskin M. S. 2012. Sickness behavior in dairy cows during Escherichia coli mastitis. J. Dairy Sci.  95: 630– 638. doi: https://doi.org/10.3168/jds.2011-4350 Google Scholar CrossRef Search ADS PubMed  Fulton R. W. Blood K. S. Panciera R. J. Payton M. E. Ridpath J. F. Confer A. W. Saliki J. T. Burge L. T. Welsh R. D. Johnson B. J. Reck A. 2009. Lung pathology and infectious agents in fatal feedlot pneumonias and relationship with mortality, disease onset, and treatments. J. Vet. Diagn. Invest.  21: 464– 477. doi: https://doi.org/10.1177/104063870902100407 Google Scholar CrossRef Search ADS PubMed  Georg H. Jahn-Falk D. Ude G. 2007. Technology against boredom – Is an automatic grooming brush accepted by calves? Landtechnik  62: 166– 167. Gershwin L. J. Van Eenennaam A. L. Anderson M. L. McEligot H. A. Shao M. X. Toaff-Rosenstein R. Taylor J. F. Neibergs H. L. Womack J and the Bovine Respiratory Disease Complex Coordinated Agricultural Project Research Team 2015. Single pathogen challenge with agents of the bovine respiratory disease complex. PLoS One  10: e0142479. doi: https://doi.org/10.1371/journal.pone.0142479 Google Scholar CrossRef Search ADS PubMed  Griffin D. 2014. The monster we don't see: Subclinical BRD in beef cattle. Anim. Health Res. Rev.  15: 138– 141. doi: https://doi.org/10.1017/S1466252314000255 Google Scholar CrossRef Search ADS PubMed  Harden L. M. du Plessis I. Roth J. Loram L. C. Poole S. Laburn H. P. 2011. Differences in the relative involvement of peripherally released interleukin (IL)-6, brain IL-1β, and prostanoids in mediating lipopolysaccharide-induced fever and sickness behavior. Psychoneuroendocrinology  36: 608– 622. doi: https://doi.org/10.1016/j.psyneuen.2010.09.003 Google Scholar CrossRef Search ADS PubMed  Harden L. M. Kent S. Pittman Q. J. Roth J. 2015. Fever and sickness behavior: Friend or foe? Brain Behav. Immun.  50: 322– 333. doi: https://doi.org/10.1016/j.bbi.2015.07.012 Google Scholar CrossRef Search ADS PubMed  Hart B. L. 1988. Biological basis of the behavior of sick animals. Neurosci. Biobehav. Rev.  12: 123– 137. doi: https://doi.org/10.1016/S0149-7634(88)80004-6 Google Scholar CrossRef Search ADS PubMed  Ishiwata T. Uetake K. Kilgour R. J. Eguchi Y. Tanaka T. 2008. Comparison of time budget of behaviors between penned and ranged young cattle focused on general and oral behaviors. Anim. Sci. J.  79: 518– 525. doi: https://doi.org/10.1111/j.1740-0929.2008.00558.x Google Scholar CrossRef Search ADS   Jordan R. Shao M. Mackman R. L. Perron M. Cihlar T. Lewis S. A. Eisenberg E. J. Carey A. Strickley R. G. Chien J. W. Anderson M. L. McEligot H. A. Behrens N. E. Gershwin L. J. 2015. Antiviral efficacy of a respiratory syncytial virus (RSV) fusion inhibitor in a bovine model of RSV infection. Antimicrob. Agents Chemother.  59: 4889– 4900. doi: https://doi.org/10.1128/AAC.00487-15 Google Scholar CrossRef Search ADS PubMed  Kemp M. H. Nolan A. M. Cripps P. J. Fitzpatrick J. L. 2008. Animal-based measurements of the severity of mastitis in dairy cows. Vet. Rec.  163: 175– 179. doi: https://doi.org/10.1136/vr.163.6.175 Google Scholar CrossRef Search ADS PubMed  Kohari D. Kosako T. Fukasawa M. Tsukada H. 2007. Effect of environmental enrichment by providing trees as rubbing objects in grassland: Grazing cattle need tree-grooming. Anim. Sci. J.  78: 413– 416. doi: https://doi.org/10.1111/j.1740-0929.2007.00455.x Google Scholar CrossRef Search ADS   Love W. J. Lehenbauer T. W. Kass P. H. Van Eenennaam A. L. Aly S. S. 2014. Development of a novel clinical scoring system for on-farm diagnosis of bovine respiratory disease in pre-weaned dairy calves. PeerJ  2: e238. doi: https://doi.org/10.7717/peerj.238 Google Scholar CrossRef Search ADS PubMed  Martin P. Bateson P. 2007. Measuring behaviour – An introductory guide. 3rd ed. Cambridge Univ. Press, Cambridge, UK. doi: https://doi.org/10.1017/CBO9780511810893 Millman S. T. 2007. Sickness behaviour and its relevance to animal welfare assessment at the group level. Anim. Welf.  16: 123– 125. Mooring M. S. McKenzie A. A. Hart B. L. 1996. Grooming in impala: Role of oral grooming in removal of ticks and effects of ticks in increasing grooming rate. Physiol. Behav.  59: 965– 971. doi: https://doi.org/10.1016/0031-9384(95)02186-8 Google Scholar CrossRef Search ADS PubMed  N'jai A. U. Rivera J. Atapattu D. N. Owusu-Ofori K. Czuprynski C. J. 2013. Gene expression profiling of bovine bronchial epithelial cells exposed in vitro to bovine herpesvirus 1 and Mannheimia haemolytica. Vet. Immunol. Immunopathol.  155: 182– 189. doi: https://doi.org/10.1016/j.vetimm.2013.06.012 Google Scholar CrossRef Search ADS PubMed  Panciera R. J. Confer A. W. 2010. Pathogenesis and pathology of bovine pneumonia. Vet. Clin. North Am. Food Anim. Pract.  26: 191– 214. doi: https://doi.org/10.1016/j.cvfa.2010.04.001 Google Scholar CrossRef Search ADS PubMed  Pecchi E. Dallaporta M. Jean A. Thirion S. Troadec J.-D. 2009. Prostaglandins and sickness behavior: Old story, new insights. Physiol. Behav.  97: 279– 292. doi: https://doi.org/10.1016/j.physbeh.2009.02.040 Google Scholar CrossRef Search ADS PubMed  Quimby W. F. Sowell B. F. Bowman J. G. P. Branine M. E. Hubbert M. E. Sherwood H. W. 2001. Application of feeding behaviour to predict morbidity of newly received calves in a commercial feedlot. Can. J. Anim. Sci.  81: 315– 320. doi: https://doi.org/10.4141/A00-098 Google Scholar CrossRef Search ADS   Rauw W. M. 2012. Immune response from a resource allocation perspective. Front. Genet.  3: 1– 14. doi: https://doi.org/10.3389/fgene.2012.00267 Google Scholar CrossRef Search ADS PubMed  Reeve-Johnson L. 2001. Relationships between clinical and pathological signs of disease in calves infected with Mannheimia (Pasteurella) haemolytica type A1. Vet. Rec.  149: 549– 552. doi: https://doi.org/10.1136/vr.149.18.549 Google Scholar CrossRef Search ADS PubMed  Reuter R. R. Carroll J. A. Hulbert L. E. Dailey J. W. Galyean M. L. 2010. Technical note: Development of a self-contained, indwelling rectal temperature probe for cattle research. J. Anim. Sci.  88: 3291– 3295. doi: https://doi.org/10.2527/jas.2010-3093 Google Scholar CrossRef Search ADS PubMed  Schaefer A. L. Cook N. J. Church J. S. Basarab J. Perry B. Miller C. Tong A. K. W. 2007. The use of infrared thermography as an early indicator of bovine respiratory disease complex in calves. Res. Vet. Sci.  83: 376– 384. doi: https://doi.org/10.1016/j.rvsc.2007.01.008 Google Scholar CrossRef Search ADS PubMed  Simonsen H. B. 1979. Grooming behaviour of domestic cattle. Nord. Vet. Med.  31: 1– 5. Google Scholar PubMed  Skinner G. W. Mitchell D. Harden L. M. 2009. Avoidance of physical activity is a sensitive indicator of illness. Physiol. Behav.  96: 421– 427. doi: https://doi.org/10.1016/j.physbeh.2008.11.004 Google Scholar CrossRef Search ADS PubMed  Snowder G. D. Van Vleck L. D. Cundiff L. V. Bennett G. L. 2006. Bovine respiratory disease in feedlot cattle: Environmental, genetic and economic factors. J. Anim. Sci.  84: 1999– 2008. doi: https://doi.org/10.2527/jas.2006-046 Google Scholar CrossRef Search ADS PubMed  Sowell B. F. Branine M. E. Bowman J. G. Hubbert M. E. Sherwood H. E. Quimby W. 1999. Feeding and watering behavior of healthy and morbid steers in a commercial feedlot. J. Anim. Sci.  77: 1105– 1112. Google Scholar CrossRef Search ADS PubMed  Spilki F. R. Esteves P. A. de Lima M. Franco A. C. Chiminazzo C. Flores E. F. Weiblen R. Driemeier D. Roehe P. M. 2004. Comparative pathogenicity of bovine herpesvirus 1 (BHV-1) subtypes 1 (BHV-1.1) and 2a (BHV-1.2a). Pesqui. Vet. Bras.  24: 43– 49. doi: https://doi.org/10.1590/S0100-736X2004000100010 Google Scholar CrossRef Search ADS   Spruijt B. M. van Hooff R. A. M. Gispen W. H. 1992. Ethology and neurobiology of grooming behavior. Physiol. Rev.  72: 825– 852. Google Scholar CrossRef Search ADS PubMed  Szyszka O. Kyriazakis I. 2013. What is the relationship between level of infection and ‘sickness behaviour’ in cattle? Appl. Anim. Behav. Sci.  147: 1– 10. doi: https://doi.org/10.1016/j.applanim.2013.05.007 Google Scholar CrossRef Search ADS   Takeuchi Y. T. K. Mori Y. 1995. Changes in the behavioral parameters following the lipopolysaccharide administration in goats. J. Vet. Med. Sci.  57: 1041– 1044. doi: https://doi.org/10.1292/jvms.57.1041 Google Scholar CrossRef Search ADS PubMed  Theurer M. E. Anderson D. E. White B. J. Miesner M. D. Mosier D. A. Coetzee J. F. Lakritz J. Amrine D. E. 2013. Effect of Mannheimia haemolytica pneumonia on behavior and physiologic responses of calves during high ambient environmental temperatures. J. Anim. Sci.  91: 3917– 3929. doi: https://doi.org/10.2527/jas.2012-5823 Google Scholar CrossRef Search ADS PubMed  Tikhonova M. A. Kulikov V. A. Kulikov A. V. 2011. Effect of LPS and serotonergic drugs on hygienic behavior in mice. Pharmacol. Biochem. Behav.  98: 392– 397. doi: https://doi.org/10.1016/j.pbb.2011.02.005 Google Scholar CrossRef Search ADS PubMed  Timsit E. Assié S. Quiniou R. Seegers H. Bareille N. 2011. Early detection of bovine respiratory disease in young bulls using recticulo-rumen temperature boluses. Vet. J.  190: 136– 142. doi: https://doi.org/10.1016/j.tvjl.2010.09.012 Google Scholar CrossRef Search ADS PubMed  Toaff-Rosenstein R. Gershwin L. J. Zanella A. J. Tucker C. B. 2016. The sickness response in steers with induced bovine respiratory disease before and after treatment with a non-steroidal anti-inflammatory drug. Appl. Anim. Behav. Sci.  181: 49– 62. http://dx.doi.org/10.1016/j.applanim.2016.05.016 Google Scholar CrossRef Search ADS   USDA 2013. Types and costs of respiratory disease treatments in U.S. feedlots. USDA-Animal and Plant Health Inspection Service-Veterinary Services-Centers for Epidemiology and Animal Health (USDA-APHIS-VS-CEAH), Fort Collins, CO. Val-Laillet D. Veira D. M. von Keyserlingk M. A. G. 2008. Short communication: Dominance in free-stall-housed dairy cattle is dependent upon resource. J. Dairy Sci.  91: 3922– 3926. doi: https://doi.org/10.3168/jds.2008-1332 Google Scholar CrossRef Search ADS PubMed  White B. J. Anderson D. E. Renter D. G. Larson R. L. Mosier D. A. Kelly L. L. Theurer M. E. Robért B. D. Walz M. L. 2012. Clinical, behavioral, and pulmonary changes in calves following inoculation with Mycoplasma bovis. Am. J. Vet. Res.  73: 490– 497. doi: https://doi.org/10.2460/ajvr.73.4.490 Google Scholar CrossRef Search ADS PubMed  White B. J. Renter D. G. 2009. Bayesian estimation of the performance of using clinical observations and harvest lung lesions for diagnosing bovine respiratory disease in post-weaned beef calves. J. Vet. Diagn. Invest.  21: 446– 453. doi: https://doi.org/10.1177/104063870902100405 Google Scholar CrossRef Search ADS PubMed  Wilson S. C. Mitlöhner F. M. Morrow-Tesch J. Dailey J. W. McGlone J. J. 2002. An assessment of several potential enrichment devices for feedlot cattle. Appl. Anim. Behav. Sci.  76: 259– 265. doi: https://doi.org/10.1016/S0168-1591(02)00019-9 Google Scholar CrossRef Search ADS   Wolfger B. Timsit E. White B. J. Orsel K. 2015. A systematic review of bovine respiratory disease diagnosis focused on diagnostic confirmation, early detection, and prediction of unfavorable outcomes in feedlot cattle. Vet. Clin. North Am. Food Anim. Pract.  31: 351– 365. doi: https://doi.org/10.1016/j.cvfa.2015.05.005 Google Scholar CrossRef Search ADS PubMed  Yirmiya R. Rosen H. Donchin O. Ovadia H. 1994. Behavioral effects of lipopolysaccharide in rats: Involvement of endogenous opioids. Brain Res.  648: 80– 86. doi: https://doi.org/10.1016/0006-8993(94)91908-9 Google Scholar CrossRef Search ADS PubMed  Footnotes 1 This project was sponsored by National Research Initiative competitive grant number 2011-68004-30367 from the USDA National Institute of Food and Agriculture and a research fellowship from the American Association of Bovine Practitioners. The brushes were donated by DeLaval USA. We gratefully acknowledge the infrastructure and support of the Department of Animal Science, College of Agricultural and Environmental Sciences, and the California Agricultural Experiment Station of the University of California-Davis. Thank you to Mark Anderson, veterinary pathologist, who collected postmortem data; the late Matt Shao and Heather McEligot, who provided invaluable laboratory assistance, farm crew, and feedlot personnel; and the many undergraduate students who assisted with this work. American Society of Animal Science
TI - Fever, feeding, and grooming behavior around peak clinical signs in bovine respiratory disease
JF - Journal of Animal Science
DO - 10.2527/jas.2016-0346
DA - 2016-09-01
UR - https://www.deepdyve.com/lp/oxford-university-press/fever-feeding-and-grooming-behavior-around-peak-clinical-signs-in-SsCS4fYyYk
SP - 3918
EP - 3932
VL - 94
IS - 9
DP - DeepDyve
ER -