Insights into shelter dog outcomes using behavioral and biological stress markers

Finding optimal tools to assess stress in animal shelters could protect welfare by identifying animals at risk of long stays and euthanasia. The goal of this observational study was to explore the utility of in-kennel behavior, fecal cortisol and immunoglobulin A (IgA) concentrations as indicators of shelter outcomes and length of stay. Between October 2023 and July 2024, 224 dogs were enrolled from three animal shelters in Pennsylvania. Fecal samples and behavioral videos were collected at two timepoints after shelter entry (0-48 h, 7-9 days), and analyzed using ELISAs and BORIS behavioral software, respectively. At shelter entry, canine behavior was largely associated with non-adoption outcomes. For instance, dogs who were returned to their owners showed increased panting (X²=13.38, p < 0.01), while euthanized and transferred dogs spent less time sitting than adopted dogs (X²=13.98, p < 0.01). For dogs who remained in the shelter at one week, shelter outcomes were largely associated with human-directed behaviors. Dogs who were adopted or transferred spent more time standing with their front paws on the cage (X²=7.95, p = 0.02) and wagging their tails than euthanized dogs (X²=6.66, p = 0.04). Adopted dogs had significantly higher cortisol concentrations than transferred dogs at baseline, but not at one week (F(2,78)=4.05, p= 0.02). IgA was not significantly related to shelter outcomes. Our findings suggest in-kennel behavioral observations may provide the most useful information for shelters to assess stress. Stress biomarkers did not offer definitive insights about shelter outcomes but highlighted a need for further research into adaptive coping styles in the shelter.