An investigation using different data gathering methods into the prevalence of behavioral problems in shelter dogs – a pilot study

Behavioral problems, especially stereotypies, in shelter dogs may be an indication of poor management practices within the shelters. To facilitate the inclusion of behavioral problems in an official standardized protocol for shelter assessment, it is necessary to identify a method to gather prevalence data, which is both valid and feasible in terms of time and resources. The aim of this pilot study was to investigate the efficacy under these constraints, of different data gathering methods in assessing the prevalence of problem behavior in shelter dogs. The methods employed were: a proxy interview (N = 428 dogs), an external inspection of the pens conducted either by non expert observers (n° 281, 292, 315, 237 dogs, respectively) or by a veterinary behaviorist (N = 303 dogs) in three shelters. Behavioral observation and tests were conducted and counselling was given on a subsample. The interviewed staff members reported a 58% overall prevalence of behavioral problems, with only 4.44% of these reported to have started during the dog’s stay at the shelter. Agreement between what was reported by the shelter staff members and what was recorded by observers inspecting the pens was not high, sensitivity ranging from 0.52 to 0.60 and specificity from 0.64 to 0.74. Stereotypies were underestimated by staff and by observers in comparison with behavioral observation results. Further studies are needed to find a feasible and reliable method to assess the prevalence of behavioral problems in shelter dogs if this variable is to be included in an official shelter quality/standard assessment protocol.