Mercier, P., Honeckman, L., Jokela, F. et al. 2023. Using standardized scales to assess fear at veterinary visits: Intra- and inter-rater reliability. Journal of Veterinary Behavior 62, 12–17.

Behavioral, emotional, and physical signs of distress are increasingly recognized in veterinary patients during routine care. Fear and distress are humane and welfare concerns themselves, but such behavior may also result in deferred veterinary care and less thorough veterinary evaluations. The first step in preventing and treating this fear is to routinely assess it. The aim of this study was to assess the inter-rater reliability of two five-point, subjective, standardized, and benchmarked Likert scales evaluating the dog’s behavior during the entry into the exam room and throughout the exam, and one benchmarked, behavioral scale that evaluated 8 behavioral components for entry into the exam room and for each of 10 steps of a standardized physical exam. We also assessed repeatability/intra-rater reliability for these scales for a subset of domestic dogs (Canis familiaris) studied. Dogs (N = 35) were enrolled in a randomized, placebo-controlled, double-blind study testing a novel pharmaceutical intervention for fear associated with veterinary exams. Dogs were videotaped at the baseline or no intervention visit and 2 weeks later at the interventional (compound/placebo) visit. All videos were coded and blinded with respect to visit order and treatment. Videos were never unblinded for treatment for this study. Videos were only unblinded for order, after the videos were assessed and the data with the statistician. Four people evaluated all videos: three behavior-resident and an experienced specialist/researcher. As part of a series of studies on this population of dogs, we report on inter-rater reliability for all scores among the residents and comparing the specialist’s versus residents’ scoring, and the residents’ intra-rater reliability/repeatability. The results show a fair-to-moderate inter-rater and intra-rater agreement. These data suggest that a standardized scale can be used reliably by multiple people and may be an essential component of a valid baseline behavioral database.