This article was downloaded by: [Gazi University] On: 03 February 2015, At: 05:57 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Applied Animal Welfare Science Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/haaw20

Use of Accelerometers to Measure Stress Levels in Shelter Dogs Sarah Jones

a c

b

, Seana Dowling-Guyer , Gary J. Patronek

b

R. Marder , Sheila Segurson D'Arpino

b d

& Emily McCobb

a b

, Amy

a

a

Center for Animals and Public Policy, Cummings School of Veterinary Medicine, Tufts University b

Center for Shelter Dogs, Animal Rescue League of Boston , Massachusetts c

University of California , Davis

d

Click for updates

Maddie's Animal Care Center , Pleasanton , CA Published online: 31 Jan 2014.

To cite this article: Sarah Jones , Seana Dowling-Guyer , Gary J. Patronek , Amy R. Marder , Sheila Segurson D'Arpino & Emily McCobb (2014) Use of Accelerometers to Measure Stress Levels in Shelter Dogs, Journal of Applied Animal Welfare Science, 17:1, 18-28, DOI: 10.1080/10888705.2014.856241 To link to this article: http://dx.doi.org/10.1080/10888705.2014.856241

PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &

Downloaded by [Gazi University] at 05:57 03 February 2015

Conditions of access and use can be found at http://www.tandfonline.com/page/termsand-conditions

JOURNAL OF APPLIED ANIMAL WELFARE SCIENCE, 17:18–28, 2014 Copyright © Taylor & Francis Group, LLC ISSN: 1088-8705 print/1532-7604 online DOI: 10.1080/10888705.2014.856241

Use of Accelerometers to Measure Stress Levels in Shelter Dogs Downloaded by [Gazi University] at 05:57 03 February 2015

Sarah Jones,1 Seana Dowling-Guyer,2 Gary J. Patronek,1;2 Amy R. Marder,2 Sheila Segurson D’Arpino,2 and Emily McCobb1 1

Center for Animals and Public Policy, Cummings School of Veterinary Medicine, Tufts University 2 Center for Shelter Dogs, Animal Rescue League of Boston, Massachusetts

Stress can compromise welfare in any confined group of nonhuman animals, including those in shelters. However, an objective and practical method for assessing the stress levels of individual dogs housed in a shelter does not exist. Such a method would be useful for monitoring animal welfare and would allow shelters to measure the effectiveness of specific interventions for stress reduction. In this pilot study, activity levels were studied in 13 dogs using accelerometers attached to their collars. Behavioral stress scores as well as urinary and salivary cortisol levels were measured to determine if the dogs’ activity levels while confined in the kennel correlated with behavioral and physiological indicators of stress in this population. The results indicated that the accelerometer could be a useful tool to study stress-related activity levels in dogs. Specific findings included a correlation between the salivary cortisol and maximum activity level (r D .62, p D .025) and a correlation between the urine cortisol-to-creatinine ratio and average activity level (r D .61, p D .028) among the study dogs. Further research is needed to better understand the complex relationship between stress and activity level among dogs in a kennel environment. Keywords: dogs, animal shelters, stress, cortisol, activity levels

Dogs may experience significant stress in nonhuman animal shelters. Admittance to a shelter results in exposure to several stressors, including a novel environment and routine, loud noises, and social isolation (Association of Shelter Veterinarians, 2010). The stress experienced by dogs within a shelter is believed to affect their welfare and potentially result in health problems (Beerda, Schilder, van Hooff, de Vries, & Mol, 1998, 1999, 2000; Tuber et al., 1999) and exacerbating anxiety-based behavior problems such as fear or aggression. Objective and Sarah Jones is currently a veterinary student at the University of California at Davis. Sheila Segurson D’Arpino is currently the director of Maddie’s Animal Care Center in Pleasanton, CA. Correspondence should be sent to Emily McCobb, Center for Animals and Public Policy, Cummings School of Veterinary Medicine, Tufts University, 200 Westboro Road, North Grafton, MA 01536. Email: [email protected]

18

Downloaded by [Gazi University] at 05:57 03 February 2015

MEASURING STRESS IN SHELTER DOGS

19

practical methods of assessing stress responses exhibited by dogs within the shelter environment are needed to help shelter workers identify and mitigate stress in dogs. Physiological indicators such as heart rate, circulating catecholamine levels, cortisol levels, and indicators of immune status have been used to study and identify stress in dogs and other animal species (Beerda, Schilder, van Hooff, & de Vries, 1997; Beerda et al., 1998, 1999, 2000; Dreschel & Granger, 2009; Hennessy, Davis, Williams, Mellott, & Douglas, 1997; Hennessy et al., 2001; Hennessy, Williams, Miller, Douglas, & Voith, 1998). Cortisol levels are an indicator of stressful stimulation in most mammals, including dogs (Kirschbaum & Hellhammer, 1989). In dogs, variations in cortisol concentrations have been documented in response to social and spatial restriction, human contact, and sudden nonsocial stimuli (Beerda et al., 1998, 1999; Hennessy et al., 1998; Kobelt, Hemsworth, Barnett, & Butler, 2003; Tuber et al., 1996). Because blood collection for plasma cortisol determination can itself be stressful and cause an increase in cortisol, noninvasive methods, such as fecal, urinary, and salivary cortisol collection, are preferred (Beerda, Schilder, Janssen, & Mol, 1996; Dreschel & Granger, 2009; Hennessy et al., 1997; Kirschbaum & Hellhammer, 1989; Rushen, 1991; Schatz & Palme, 2001). Noninvasively collected cortisol samples have been used to assess stress in shelter dogs. For example, in one study, urinary cortisol levels measured in shelter dogs increased from Day 2, reached their peak on Day 17, steadily declined thereafter (Stephen & Ledger, 2006), and in general, were in agreement with findings of an earlier study that used plasma cortisol in shelter dogs (Hennessy et al., 2001). Salivary cortisol has also been validated as a useful method of cortisol assessment in dogs and provides a more immediate measure than urinary cortisol (Beerda et al., 1996; Dreschel & Granger, 2009). It can take up to 4 min to collect a saliva sample from a dog without the effect of handling being reflected in cortisol concentrations (Kobelt et al., 2003). A number of behavioral indicators have been proposed to reflect poor welfare in dogs who are kenneled (Beerda et al., 1997, 1998, 1999, 2000). These indicators include: (a) abnormal behaviors (repetitive behavior, self-mutilation, or coprophagy), (b) behavior indicating frustration (chewing or vocalizing), (c) conflict behaviors (body shaking or paw lifting), and (d) a lowered, fearful posture (Beerda et al., 1997, 1998, 1999, 2000; Hewson, Hiby, & Bradshaw, 2007). In a shelter setting, dogs may cope with their socially isolated and deprived surroundings by developing attention-seeking behavior and hyperactivity, stress behaviors that can also be linked to a rise in cortisol (Beerda et al., 1996, 1997, 1998, 2000). Dogs may be more active and therefore potentially more stressed in smaller cages (Hite, Hanson, Bohidar, Conti, & Mattis, 1997; Hughes, Campbell, & Kenney, 1977), and dogs have been found to be most active when they are socially isolated (Hetts, Clark, Calpin, Arnold, & Mateo, 1992). Although high activity may be a potential indicator of stress in shelter dogs, inactivity may also be associated with high stress levels (Beerda et al., 1997; Hubrecht, Serpell, & Poole, 1992; McCobb, Patronek, Marder, Dinnage, & Stone, 2005). Hubrecht et al. (1992) found that dogs are often inactive when housed solitarily in areas smaller than 3 m2 . The purpose of this study was to examine the relationship between behavioral signs of stress, physiological indicators of stress (salivary and urinary cortisol), and dog activity levels for the first 3 days after intake, measured with an accelerometer fastened to the collar, among dogs housed in an animal shelter.

20

JONES ET AL.

METHODS Subjects

Downloaded by [Gazi University] at 05:57 03 February 2015

Study dogs were all dogs entering an urban, planned admission animal shelter during 3 months (May, June, and July) in 2009. Dogs were excluded if they were younger than 6 months, housed with another dog, injured or diagnosed by the shelter veterinarian as sick, or too aggressive to handle. Dogs who entered the shelter afterhours or who were immediately made available for adoption were not able to be enrolled. Finally, dogs receiving medications that would have interfered with their cortisol levels would have been excluded, although this did not apply to any the study dogs. Ethical Considerations The study design was approved by the Center for Animals and Public Policy’s internal review panel. Husbandry and Housing Dogs were housed individually in one of three rooms, and all were cared for according to the same routine shelter protocols. Assessments Activity level. The activity levels of each dog were measured using accelerometers specifically designed for dogs (“SNIF tags,” SNIF Labs, LLC). Accelerometers have been shown to accurately measure the activity of dogs when compared with video recordings, although there is no one standard way of calculating that movement yet (Hansen, Lascelles, Keene, Adams, & Thomson, 2007; Stiles, Palestrini, Beauchamp, & Frank, 2011; Yamada & Tokuriki, 2000). They also have been used in experimental designs to detect the effect of an intervention or treatment (Brown, Boston, & Farrar, 2010; Stiles et al., 2011). In the SNIF tag device, accelerometers are embedded in radiofrequency identification tags that capture and transmit motion data to a central online database. The SNIF tag measures the activity level by continuously recording the total amount of “energy” expended in 5-min intervals so that in one 24-hr period, 288 measurements are recorded. The mean energy expended and maximum energy expended were calculated for each dog using the SNIF tag output and were taken to represent the average and greatest activity level. The SNIF tags included a key ring-type device that was used to attach the device to the ventral location of the collar, which has been shown to be a good location for collecting locomotive data (Hansen et al., 2007). Cortisol measurements. Urine was collected from each dog using a constructed device that enabled the collector to catch urine as the dog naturally voided. A cardboard bowl was used for one dog who was afraid of the constructed device. When dogs urinated in their kennels, the urine was collected via a syringe off of the kennel floor, immediately after they had voided. For

Downloaded by [Gazi University] at 05:57 03 February 2015

MEASURING STRESS IN SHELTER DOGS

21

one sample for one dog, urine was collected from the floor in the morning when the observer arrived. Enrolled dogs without at least one valid urinary cortisol measurement, including dogs for whom samples could not be obtained or whose samples were not adequate for lab analysis, were dropped from the study. After collection, the samples were immediately refrigerated at 41ı F and were stored until they were sent to a commercial diagnostic laboratory (IDEXX Laboratories) where the cortisol-to-creatinine ratio was analyzed. Saliva samples were collected following the recommendations and protocol established by Dreschel and Granger (2009) and Salimetrics LLC. Enrolled dogs without at least one valid salivary cortisol measurement, including those dogs not expected to tolerate an oral swab placement or whose samples were not adequate for lab analysis, were dropped from the study. The cortisol analysis was conducted at a commercial laboratory (Salimetrics LLC) in accordance with previously published guidelines for such analysis (Dreschel & Granger, 2009). After the samples were collected, they were held in a freezer at 20ı F. Behavioral stress scores. Because no formal stress score has been validated in the literature, the stress score used in this study was based on a review of the current literature (Aloff, 2005; Beerda et al., 2000; Handelman, 2008) and consultation with behavior specialists. The score was determined by observations of each dog’s body posture, eyes, mouth, tail position, vocalizations, and ear position, as well as the occurrence of any stress behaviors listed in Table 1. For consistency, all observed behaviors were recorded within a 1-min time period to evaluate each individual dog’s stress and assign a stress score. The rater stood directly outside the dog’s kennel while the dog was loose in the kennel and observed the dog’s behavior for 1 min. No other placement for the rater was possible due to the building design. According to this scoring system, a dog who appeared relaxed (i.e., not exhibiting any of the behaviors listed in Table 1) was assigned a score of 0, whereas a dog who appeared to be highly stressed and was therefore exhibiting significant amounts of stress behaviors was assigned a score of 5 (see Table 2 for the behavioral stress scale).

TABLE 1 Behaviors Used to Develop the Behavioral Stress Assessment Score 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.

Paw lift Lip lick Sneeze Stretch Yawn Scratching (grooming) Cautious/reluctant approach Tail lowered Jumping on cage/walls Whining Panting Avoidance of eye contact

13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24.

Avoidance of interaction Heavy panting Full-body stiffness Ears forced back/pulled tightly to head Staring Tail tucked Heavy panting with drooling Trembling Spinning/circling/weaving Sweaty paws Vocalizing while making constant eye contact Growling

22

JONES ET AL.

TABLE 2 Behavioral Canine Stress Score Scale Score 1. Mild 2. Mild/Moderate

3. Moderate

Downloaded by [Gazi University] at 05:57 03 February 2015

4. Moderate/Severe

5. Severe/Extreme

Description Observe no more than two to three different mild stress behaviors (i.e., Numbers 1–8 in Table 1) no more than once or twice each (Dog is overall pretty calm.) Observe no more than two to three different mild stress behaviors (i.e., Numbers 1–8 in Table 1) more than three times and/or observe no more than one to two more severe stress behaviors (i.e., Numbers 9–14 in Table 1) Observe mild stress behaviors (i.e., Numbers 1–8 in Table 1) continuously and/or observe no more than one to two of more severe stress behaviors (i.e., Numbers 9–14 in Table 1) continuously Observe many of the previously mentioned mild behaviors (i.e., Numbers 1–8 in Table 1) more than three times and/or observe no more than two to three more severe stress behaviors (i.e., Numbers 14–24 in Table 1) Observe more than three of the mild stress behaviors (i.e., Numbers 1–4 in Table 1) continuously/heavily and/or observe the severe stress behaviors (i.e., Numbers 15–24 in Table 1) multiple times

Note. Behaviors were recorded within a 1-min time period and were assigned a stress score based on the frequency of that behavior.

Study Design All measurements were taken by a single investigator (S.J.) who also assigned the behavioral stress scores. On Day 0, dogs were admitted to the shelter and the SNIF tag was applied. On Day 1, dogs were scored for behavior while in their kennels and cortisol samples were collected. Saliva samples were collected in the morning, midday, and afternoon; and urine samples were collected in the morning and afternoon. On Day 2, dogs were scored for behavior stress in the morning and afternoon when the SNIF tags were removed. Urine and saliva samples were collected in the morning. Data Analysis Measurements were averaged for urinary and salivary cortisol across the three sampling times for each dog, creating a mean urinary and mean salivary cortisol score for each dog. A similar calculation was performed for the behavioral stress score across the six measurement periods to determine a mean stress score. Mean activity level was calculated across the entire measurement period for each dog, whereas maximum activity level was defined as the highest activity measurement for each subject during the measurement period. Descriptive statistics were calculated for mean urinary cortisol, mean salivary cortisol, mean stress score, mean activity level, and maximum activity level. Differences in measurement by categorical demographic characteristics were tested using t tests, and relationships between continuous demographic variables (e.g., age, weight) and the measurement variables were examined using Pearson’s correlation coefficients. Scatter plots and Pearson’s correlation coefficients were used to examine relationships between the following pairs of variables: (a) accelerometer readings versus urinary cortisol levels, (b) accelerometer readings versus salivary cortisol levels, (c) accelerometer readings versus subjective stress scores, (d) cortisol levels versus behavioral

MEASURING STRESS IN SHELTER DOGS

23

stress scores, and (e) urinary cortisol levels versus salivary cortisol levels. The p values less than .05 were considered statistically significant. Data were analyzed using PASW (SPSS) Version 18.

RESULTS

Downloaded by [Gazi University] at 05:57 03 February 2015

Subjects Of the 41 dogs who entered the shelter during the study period (May 12, 2009–July 26, 2009), 17 (42%) were eligible to be included in the study. Reasons for exclusion included entering the shelter afterhours (10 dogs), being group-housed (7 dogs), aggression (2 dogs), and injury (3 dogs). In addition, 1 dog whose tag fell off shortly after admission and 1 dog who was immediately placed up for adoption were not able to be enrolled. Twelve of 17 enrolled dogs (71%) were female, and 5 of these were spayed. Five of 17 (29%) of the enrolled dogs were male, and none of them were neutered. Study dogs ranged in age from 6 months to older than 7 years. Dogs were predominantly mixed breeds. Six of 17 (41%) dogs were surrendered by their owners while 10 (59%) were transferred from other shelters, and 1 dog was returned by an adopter. There were more female dogs among the enrolled dogs than the excluded dogs (12 vs. 6) and more male dogs among the excluded group than the enrolled group (X 2 D 8.40, p D .004). There were no other significant differences between enrolled and excluded dogs. Four enrolled dogs were later dropped from the study due to missing salivary cortisol measures, bringing the final sample size to 13 dogs.

Stress Measurements Descriptive statistics for the physiologic and behavioral stress assessments are shown in Table 3. Values for the urinary cortisol-to-creatinine ratio as reported by the reference laboratory ranged from 3.00 to 25.00, with a mean of 9.87. The mean correlation for urinary cortisol ratios across the three time periods was .88. Salivary cortisol ranged from 0.07 to 0.86 g/dL, with a mean of 0.31 g/dL. The mean correlation for salivary cortisol across the three time periods was .42. Behavioral stress scores ranged from 2 to 5, with a mean score of 3.5 and a mean correlation of .46. Mean urinary cortisol-to-creatinine ratios and mean salivary cortisol levels were significantly correlated (r D .61, p D .026). The mean stress score was not significantly correlated with either the urinary cortisol-to-creatinine ratios (r D .44, p D .136) or the salivary cortisol levels (r D .44, p D .130). Male dogs had significantly higher behavioral stress scores than did female dogs (Xmales D 4.0, Xfemales D 3.2, t D 3.60, p D .004), but there were no significant differences by sex in either measure of cortisol (p > .05). The correlation of age to each of the stress measurements was not significant (p > .05). There was a significant negative correlation between weight and the mean urinary cortisol-to-creatinine ratio (r D .76, p D .003) and between weight and salivary cortisol levels (r D .56, p D .046). There was no relationship between the weight of the dogs and the behavioral stress scores (p > .05).

24

JONES ET AL.

TABLE 3 Physiological and Behavioral Measures of Stress in 13 Shelter Dogs

Downloaded by [Gazi University] at 05:57 03 February 2015

Measure Urinary Cortisol (Cortisol-to-Creatinine ratio) a.m. Day 1 p.m. Day 1 a.m. Day 2 Overall Salivary Cortisol (g/dL) a.m. Day 1 Noon Day 1 p.m. Day 1 Overall Stress Score (1–5) (1 D minimally stressed, 5 D highly stressed) At time of intake a.m. Day 1 Noon Day 1 p.m. Day 1 a.m. Day 2 Noon Day 2 Overall

Mean

Median

SD

9.25 12.55 9.25 9.87

9.00 11.00 9.00 8.67

3.62 6.95 4.11 4.63

0.37 0.27 0.28 0.31

0.29 0.20 0.24 0.27

0.25 0.15 0.12 0.14

3.9 3.9 3.4 3.2 3.7 3.2 3.5

4.0 4.0 3.0 3.0 4.0 3.0 3.5

0.8 1.0 0.7 0.7 0.5 0.7 0.6

Activity Level The results for activity level (mean and maximum levels) are given in Table 4. Mean activity level and maximum activity level were not significantly correlated (p > .05). Although maximum activity level did not significantly correlate with weight (p > .05), mean activity level was significantly negatively correlated with weight (r D .58, p D .037). Activity levels did not differ significantly between dogs of different sexes, nor was a significant correlation found between activity level and age (p > .05).

TABLE 4 Activity Level for 13 Shelter Dogs by Time of Day for All Dogs Measured by Accelerometer Activity Level (Activity Units)

Mean

Median

SD

Time 1 (12 a.m.–8 a.m.) Time 2 (8 a.m.–4 p.m.) Time 3 (12 a.m.–8 a.m.) Overall (Day 0–Day 2) Maximum

0.01 0.04 0.01 0.02 0.28

0.005 0.04 0.01 0.02 0.27

0.008 0.03 0.01 0.01 0.09

Note. The activity levels were calculated by averaging the total activity within the given time periods.

MEASURING STRESS IN SHELTER DOGS

25

Relation Between Activity Level and Stress Measurements There was a significant correlation between maximum activity level and salivary cortisol (r D .62, p D .025), although there was not a significant correlation between maximum activity level and urinary cortisol (r D .18, p > .05). There was a significant correlation between mean activity level and mean urinary cortisol (r D .61, p D .028) but not between mean activity level and mean salivary cortisol (p > .05). There were no significant correlations between the mean behavioral stress score and the mean or maximum activity level (p > .05).

Downloaded by [Gazi University] at 05:57 03 February 2015

DISCUSSION A validated behavioral score for stress would facilitate an accurate assessment of stress in shelter dogs. However, up to this point, it has been difficult to demonstrate any association between behavioral and physiological measures of stress (Hennessy et al., 2001; Rooney, Gaines, & Bradshaw, 2007). To our knowledge, this is the first study to demonstrate the utility of activity level, measured with an accelerometer, as a potential indicator of physiological and behavioral stress in a group of shelter dogs. The results demonstrated that the collar accelerometer was a feasible tool for measuring activity level in the shelter environment. Mean salivary cortisol levels measured in this study (0.31 g/dL) were similar to values found among a population of shelter dogs (0.252–0.409 g/dL Coppola, Grandin, & Enns, 2006) but were higher than those reported in a control sample of dogs who were companions (0.170 g/dL; Dreschel & Granger, 2009). Urinary cortisol-to-creatinine ratios reported here are higher than those reported by Beerda et al. (2000) in four groups of dogs living in varying kennel conditions for longer than a year, and they are higher than the mean ratio reported by van Vonderen, Kooistra, and Rijnberk (1997) in companion dogs. The assumption that shelter dogs are more stressed than dogs residing in homes is thus supported by our findings. These findings may also suggest that dogs entering a kenneled environment, in this case a shelter, are more stressed than are dogs who have adapted to a kennel environment. In this preliminary study, some relationships between physiologic indexes of stress (cortisol) and behavioral indexes (activity) were found. Relationships between activity level and cortisol levels suggest that the activity level may serve as a behavioral indicator for stress. The fact that salivary cortisol levels and maximum activity were related in this group of dogs supports the theory that these parameters might be more useful for the assessment of acute stress in dogs. Salivary cortisol has been previously shown to reflect acute stress in dogs (Hekman, Karas, & Dreschel, 2012). On the other hand, the fact that urinary cortisol, the more cumulative measure of stress, was correlated with the mean activity level suggests that these two parameters (urinary cortisol and mean activity level) might reflect the average state of the dog and thus be useful to assess more chronic stress levels. If these relationships are validated, shelters would be able to monitor maximum activity as an indicator of acute stress and mean activity level as an indicator of the average stress level of their dogs. Further studies with a larger group of dogs would be needed to confirm these apparent trends. An important consideration is that significant increases in activity as well as significant decreases in activity are a sign of stress, and thus, these effects would cancel each other out

Downloaded by [Gazi University] at 05:57 03 February 2015

26

JONES ET AL.

when assessing mean activity level even in a highly stressed population of dogs. In this study, the sample size was too small to allow further analysis of differences in expression of activity (hypoactivity and hyperactivity) in response to stress. One difficulty in using activity levels as a marker for stress is that activity level can be both a sign of stress and a mechanism for coping with the stress (Beerda et al., 1997, 2000), in which case one might expect highly active dogs to have lower cortisol levels. In addition, we cannot rule out observer effects as a factor because direct, in-person assessments were used for behavioral scoring. However, even if the observer presence increased stress levels, one could still expect a correlation between highly stressed and active dogs. Overall, in this study, higher levels of cortisol were correlated with higher activity. Behavioral stress scores, but not cortisol or activity levels, were higher among male dogs. However, male dogs in general were underrepresented among our study population, so it is difficult to draw conclusions from this result. Future studies should explore differences in stress level among male and female dogs, which may allow shelters to better tailor behavioral interventions. The negative correlation found between weight and cortisol levels was unexpected. In standard clinical laboratory practice, no adjustment is made for dog weight when reporting urine cortisol-to-creatinine ratios. Therefore, it is possible that this finding is an artifact of the small sample size. However, the negative correlation between mean activity level (which significantly correlated with urinary cortisol) and weight provides additional support for the notion that smaller dogs may be more stressed in the shelter environment compared with larger dogs, although it is important to note that the accelerometer results can be affected by weight of the dog for some activities (Brown, Michel, Love, & Dow, 2010). Based on these results, we suggest that weight and possibly body condition should be taken into account in all studies of stress, specifically those using cortisol. We were unable to demonstrate a relationship between our behavioral stress score and the animals’ cortisol levels. On the other hand, the correlations that we were able to demonstrate between activity level and physiologic measurement may suggest that activity is a more useful indication of behavioral stress than a simple score in this species. Behavior is complex and may change in response to external stimuli. Rooney et al. (2007) speculated that one explanation for such a lack of correlation is the possibility that some of the changes in behavior observed during kenneling might be interpreted as reflecting learned responses of some dogs to the novel environment, rather than being indicators of welfare status. Moreover, there are individual coping strategies and differences in the behavioral expression of stress.

CONCLUSION In summary, our findings support the use of activity as a tool for stress assessment in shelter dogs; however, the relationship between activity and other stress indicators is complicated. We believe that shelters should monitor the activity levels among dogs, and individual dogs who appear unusually active or inactive should be flagged for further examination and possible behavioral interventions. When designing enrichments for dogs in shelters, activity level should be taken into account and individual differences should be expected. Further research is needed to better understand the relationship between body size, activity, behavioral stress signs, and cortisol levels among dogs.

MEASURING STRESS IN SHELTER DOGS

27

FUNDING This study was supported by a grant from the Stanton Foundation.

Downloaded by [Gazi University] at 05:57 03 February 2015

REFERENCES Aloff, B. (2005). Canine body language: A photographic guide interpreting the native language of the domestic dog. Wenatchee, WA: Dogwise. Association of Shelter Veterinarians. (2010). Guidelines for standards of care in animal shelters. Retrieved from http:// www.sheltervet.org/wp-content/uploads/2011/08/Shelter-Standards-Oct2011-wForward.pdf Beerda, B., Schilder, M. B. H., Janssen, N. S. C. R. M., & Mol, J. A. (1996). The use of saliva cortisol, urinary cortisol, and catecholamine measurements for a noninvasive assessment of stress responses in dogs. Hormones and Behavior, 30, 272–279. Beerda, B., Schilder, M. B. H., van Hooff, J. A. R. A. M., & de Vries, H. W. (1997). Manifestations of chronic and acute stress in dogs. Applied Animal Behaviour Science, 52, 307–319. Beerda, B., Schilder, M. B. H., van Hooff, J. A. R. A. M., de Vries, H. W., & Mol, J. A. (1998). Behavioural, saliva cortisol and heart rate responses to different types of stimuli in dogs. Applied Animal Behaviour Science, 58, 365–381. Beerda, B., Schilder, M. B. H., van Hooff, J. A. R. A. M., de Vries, H. W., & Mol, J. A. (1999). Chronic stress in dogs subjected to social and spatial restriction. I. Behavioral responses. Physiology & Behavior, 66, 233–242. Beerda, B., Schilder, M. B. H., van Hooff, J. A. R. A. M., de Vries, H. W., & Mol, J. A. (2000). Behavioural and hormonal indicators of enduring environmental stress in dogs. Animal Welfare, 9, 49–62. Brown, D. C., Boston, R. C., & Farrar, J. T. (2010). Use of an activity monitor to detect response to treatment in dogs with osteoarthritis. Journal of the American Veterinary Medical Association, 237, 66–70. Brown, D. C., Michel, K. E., Love, M., & Dow, C. (2010). Evaluation of the effect of signalment and body conformation on activity monitoring in companion dogs. American Journal of Veterinary Research, 71, 322–325. Coppola, C. L., Grandin, T., & Enns, R. M. (2006). Human interaction and cortisol: Can human contact reduce stress for shelter dogs? Physiology & Behavior, 87, 537–541. Dreschel, N. A., & Granger, D. A. (2009). Methods of collection for salivary cortisol in dogs. Hormones and Behavior, 55, 163–168. Handelman, B. (2008). Canine behavior: A photo illustrated handbook. Wenatchee, WA: Dogwise. Hansen, B. D., Lascelles, D. X., Keene, B. W., Adams, A. K., & Thomson, A. E. (2007). Evaluation of an accelerometer for at-home monitoring of spontaneous activity in dogs. American Journal of Veterinary Research, 68, 468–475. Hekman, J. P., Karas, A. Z., & Dreschel, N. A. (2012). Salivary cortisol concentrations and behavior in a population of healthy dogs hospitalized for elective procedures. Applied Animal Behaviour Science, 141, 149–157. Hennessy, M. B., Davis, H. N., Williams, M. T., Mellott, C., & Douglas, C. W. (1997). Plasma cortisol levels of dogs at a county animal shelter. Physiology & Behavior, 62, 485–490. Hennessy, M. B., Voith, V., Mazzei, S., Buttram, J., Miller, D., & Linden, F. (2001). Behavior and cortisol levels of dogs in a public animal shelter, and an exploration of the ability of the measure to predict problem behavior after adoption. Applied Animal Behaviour Science, 73, 217–233. Hennessy, M. B., Williams, M. T., Miller, D. D., Douglas, C. W., & Voith, V. (1998). Influence of male and female petters on plasma cortisol and behaviour: Can human interaction reduce the stress of dogs in a public animal shelter? Applied Animal Behaviour Science, 61, 63–67. Hetts, S., Clark, D., Calpin, J., Arnold, C. E., & Mateo, J. M. (1992). Influence of housing conditions on beagle behaviour. Applied Animal Behaviour Science, 34, 137–155. Hewson, C. J., Hiby, E. F., & Bradshaw, J. W. S. (2007). Assessing quality of life in companion and kenneled dogs: A critical review. Animal Welfare, 16, 89–95. Hite, M., Hanson, H. M., Bohidar, N. R., Conti, P. A., & Mattis, P. A. (1977). Effect of cage size on patterns of activity and health of beagle dogs. Laboratory Animal Science, 27, 60–64. Hubrecht, R. C., Serpell, J. A., & Poole, T. B. (1992). Correlates of pen size and housing conditions on the behaviour of kenneled dogs. Applied Animal Behaviour Science, 34, 365–383.

Downloaded by [Gazi University] at 05:57 03 February 2015

28

JONES ET AL.

Hughes, H. C., Campbell, S., & Kenney, C. (1977). The effects of cage size and pair housing on exercise of beagle dogs. Laboratory Animal Science, 27, 60–64. Kirschbaum, C., & Hellhammer, D. H. (1989). Salivary cortisol in psychobiological research: An overview. Neuropsychobiology, 22, 150–169. Kobelt, A. J., Hemsworth, P. H., Barnett, J. L., & Butler, K. L. (2003). Sources of sampling variation in saliva cortisol in dogs. Research in Veterinary Science, 75, 157–161. McCobb, E., Patronek, G. J., Marder, A., Dinnage, J. D., & Stone, M. S. (2005). Assessment of stress levels among cats in four animal shelters. Journal of the American Veterinary Medical Association, 226, 548–555. Rooney, N. J., Gaines, S., & Bradshaw, J. (2007). Behavioural and glucocorticoid responses of dogs (Canis familiaris) to kenneling: Investigating mitigation of stress by prior habituation. Physiology & Behavior, 92, 847–854. Rushen, J. (1991). Problems associated with the interpretation of physiological data in the assessment of animal welfare. Applied Animal Behaviour Science, 28, 381–386. Schatz, S., & Palme, R. (2001). Measurement of faecal cortisol metabolites in cats and dogs: A non-invasive method for evaluating adrenocortical function. Veterinary Research Communications, 25, 271–287. Stephen, J. M., & Ledger, R. A. (2006). A longitudinal evaluation of urinary cortisol in kenneled dogs, Canis familiaris. Physiology & Behavior, 87, 911–916. Stiles, E. K., Palestrini, C., Beauchamp, G., & Frank, D. (2011). Physiological and behavioral effects of dextroamphetamine on beagle dogs. Journal of Veterinary Behavior: Clinical Applications and Research, 6, 328–336. Tuber, D. S., Miller, D. D., Caris, K. A., Halter, R., Linden, F., & Hennessy, M. B. (1999). Dogs in animal shelters: Problems, suggestions and needed expertise. Psychological Science, 10, 379–386. van Vonderen, I. K., Kooistra, H. S., & Rijnberk, A. (1997). Intra- and inter-individual variation in urine osmolarity and urine specific gravity in healthy pet dogs of various ages. Journal of Veterinary Internal Medicine, 11, 30–35. Yamada, M., & Tokuriki, M. (2000). Spontaneous activities measured continuously by an accelerometer in beagle dogs housed in a cage. Journal of Veterinary Medical Science, 62, 443–447.

Use of accelerometers to measure stress levels in shelter dogs.

Stress can compromise welfare in any confined group of nonhuman animals, including those in shelters. However, an objective and practical method for a...
173KB Sizes 0 Downloads 0 Views