110 Using qualitative behavior assessment to investigate horse well-being during equine assisted activities

Abstract

The objective of this study was to explore whether horses’ emotional well-being could be determined using qualitative behavior assessment (QBA) and if equine assisted activities (EAA) had a significant impact on their welfare. QBA is a methodology that has gained recent scientific validation and is considered an accurate welfare-monitoring tool. It is the analysis of subtle changes within body language to capture the expressive quality of the whole animal. Implementation of this technique involved recruitment of observers (19 students) to watch video recordings of horses. There was no requirement of prior horse experience. The first part of this project utilized free choice profiling (FCP) in which observers generated their own qualitative descriptors of behavior after watching 6 video clips of 5 min each. The second part required them to quantitatively rate a horse's expression of each term from a 30 min video recording. Video clips for the first part were a subset of the recordings used in the second part and involved the same horses. Ratings were completed on a Visual Analogue Scale (VAS). Videos were collected during weekly sessions over an 8-wk period and consisted of either 30 min of a horse assigned to an adaptive horsemanship lesson with a veteran (EAA, n = 6,20.3 ± 1.9 yrs, mean ± SE) or standing control with no interaction (CON, n = 6, 13.8 ± 1.7 yrs). The data generated were analyzed using a generalized procrustes analysis (GPA) and produced a significant (P = 0.001, T-critical = 18.3, T-value = 109.7) consensus profile configuration explaining 41.2% of the variation between horses. Component 1 was positively correlated with terms such as “stressed” and “anxious” and negatively correlated with terms including “calm,” “patient,” and “comfortable.” A repeated measures ANOVA with fixed effects of treatment and week and random effect of horse was used to analyze horse scores on the first and second components. Statistically different means were separated using Tukey's method. Significance was set at P ≤ 0.05. Scores on component 1 differed significantly (P = 0.0003) between EAA and CON horses with EAA horses being rated as calmer, more patient, and more comfortable. QBA was an effective means in determining horse well-being. EAA resulted in less stress than the standing control.