A central problem in animal welfare of pet dogs is abandonment by their owners. Abandoned dogs are commonly brought to animal shelters, and they suffer many kinds of psychogenic stressors there (1). In particular, the stress level of dogs as assessed by plasma cortisol concentration was reported to be highest during the first 3 days in the shelter (2). As physiological indicators of stress, hypothalamic–pituitary–adrenal measures using blood, saliva, urinary, fecal, and hair glucocorticoids, and their metabolites are also common, as reviewed by Hennessy (1).

In addition to physiological stress responses, behavioral changes indicate that stress was observed when dogs were kenneled at the shelter. As behavioral indicators that give signs of stress in kenneled dogs, changes in activity levels as well as the expression of stereotypic behaviors and fear that dogs display at the shelter have been reported, as reviewed by Protopopova (3).

In Japan, the human population is aging rapidly, and the abandonment of dogs by the elderly persons who have died or been hospitalized has become one of the main reasons. These dogs are often old, too. It is hypothesized that the elderly dogs have difficulty in adapting themselves to the novel circumstances in the shelter. For example, Mongillo et al. (4) reported that aged dogs (7 years or older) cope less efficiently with emotional distress caused by a change of caretaker.

Therefore, the objective of this study was to assess the stress level and demonstrate stress responses of elderly dogs that were admitted to an animal shelter. As stress indicators, fecal corticosterone level and changes in their ethograms were investigated on a daily basis, especially focusing on the first week of admittance. Fecal corticosterone monitoring was applied because of its non-invasiveness to subject animals (5).

The effect of day in shelter on fecal corticosterone level was significant (χ² = 23.19, df = 5, P < 0.001). Fecal corticosterone levels gently fell from the day after admittance (mean ± SE: 16650.1 ± 3769.7 ng/g) to the seventh day (12178.4 ± 2524.4 ng/g; Figure 1). However, there was no significantly different combination of days in the shelter in post hoc multiple comparisons.

The proportion of behavioral expressions was significantly different between days in shelter (χ² = 120.36, df = 30, P < 0.001). Behavioral activities that showed a big change during the first 6 days of admittance were stereotypies and sleeping behavior. The former gradually decreased from 35.7% on the first day to 2.6% on the sixth day, and the latter increased from 0.0 to 42.7% (Figure 2).

As far as we saw a change in behavioral expression of the elderly dogs, they seem to habituate themselves to the novel circumstances of an animal shelter in the first week of residence. A decrease in the expression of stereotypies or repetitive behavior is a good indication of environmental adaptation. In this study, behavioral observations were conducted in a time zone (1400–1600 hours), where caretakers did not exist nearby. Thus, increase in sleeping behavior in this time would be a good indication, too.

On the other hand, their fecal corticosterone levels were still high even on the seventh day of residence (12178.4 ± 2524.4 ng/g). It was more than 10-fold the mean baseline level of dogs that had regular contact with humans at the same shelter (e.g., dogs used for dog training classes: 1035.9 ± 179.9 ng/g) (7). It is reported that the stress level of dogs admitted to a shelter is the highest in the first 3 days in the shelter, with a gradual waning thereafter (2). We cannot simply compare stress levels across studies because besides the age of the dogs many factors differed (e.g., measurement in the plasma vs. in the feces, different shelters, different dog breeds, etc.). However, when we synthesize these behavioral and physiological indicators, it seems that the elderly dogs need more than 1 week to habituate themselves adequately to the novel circumstances in the shelter.

As a practical way of reducing the stress of dogs in shelters, it is recommended that shelter staff members spend time interacting with the resident dogs (8). However, Mongillo et al. (4) have reported that aged dogs cope less efficiently with the emotional distress caused by a change of caretaker. For this reason, they need a longer time to develop trust for unfamiliar staff members and relax their guard. Gaultier et al. (9) reported that an alternative way to reduce stress linked to social separation is using a dog-appeasing pheromone.

In conclusion, the elderly dogs newly admitted to an animal shelter seem to behaviorally adapt themselves to the novel circumstances but might be stressed even on the seventh day of residence. Further studies are needed to clarify how long elderly or aged dogs need to overcome the stress caused by abandonment by their owners and to treat it adequately.

KU, CY, and AE designed the study through discussion, and TT gave advice about the experimental design. CY and AE performed the data collection, and KU and CY performed statistical analyses. All the authors interpreted data. KU wrote the manuscript, and CY, AE, and TT contributed to the manuscript. All the authors approved the final version of the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.