Restless nights? Nocturnal activity as a useful indicator of adaptability of shelter housed dogs

Shelter dogs face the challenge of adapting to a kennel environment. Individual differences in adaptation are known to exist. Resting patterns might be indicative of adaptability to such a novel environment, but need to be evaluated for its usefulness, like every potentially physiological and behavioural parameter. Here, we evaluated nocturnal activity patterns of dogs as indicators of adaptability to novel environments. We measured nocturnal activity (3-axial accelerometer, the Actical®) and two physiological stress parameters, i.e. urinary cortisol/creatinine ratio (UCCR) and body weight in 29 dogs relinquished to a shelter (SD group) in the first two days after intake at the shelter (n = 29), after a 12-day habituation period in the shelter (n = 28) and >6 weeks post-adoption (n = 17). A control group of 29 pet dogs kept at home (CPD group), matching the SD group characteristics, was also assessed for its nocturnal activity and UCCR. Linear mixed model analysis, t-tests and Friedman tests were used to analyse the data. The main findings are: 1) the SD group exhibited higher nocturnal activity (total activity counts, activity duration and number of rest bouts) the first two nights after intake than on night 12, with decreasing inter-individual variances. Compared to the CPD group they showed higher nocturnal activity on night 1 (all p < 0.001) and night 12 (all p ≤ 0.001) except for total activity counts on night 12. We found no ‘first-night effect’, where sleep is disturbed during the first night; nocturnal activity in the shelter did not significantly differ between nights 1 and 2 in the shelter. 2) In line with literature findings, SD group UCCRs were higher shortly after intake than after a 12-day habituation period and after adoption, and higher than in the CPD group. 3) An interaction was found between weight class and both nocturnal activity and UCCR levels: in their first days in the shelter, smaller dogs showing higher levels than larger dogs. 4) Dogs in the SD group lost, on average, 5% of their body weight between intake and the two-week habituation period. In conclusion, nocturnal activity, as measured by an accelerometer, may be a valid parameter to monitor adaptability of dogs to a kennel environment. Monitoring nocturnal activity in this way can be a useful and cost-effective additional indicator for assessing dog welfare