Development and Validation of a Test for the Classification of Horses as Broken or Unbroken

SIMPLE SUMMARY: Transportation is a stressful event for all animal species, but some species may be subjected to worse welfare consequences than others due to their ethological characteristics and specific coping strategies. Among equines, horses with a low level of tameness are at higher risk for transport-related disease and injury. For this reason, in Europe, regulations for the protection of animals during transport (Regulation EC 1/2005) are stricter for unbroken (untamed) vs. broken (tamed) horses. However, in practice, official veterinarians cannot verify regulatory compliance as there is no valid tool for the classification of horses as broken or unbroken. This study proposes the Broken/Unbroken Test (BUT) for assessment and scoring of horse behaviour during approach, haltering, and handling. After a validation process, our study has shown that the BUT is a reliable, valid, and feasible tool for determining whether a horse is broken or unbroken. The use of this tool would allow simple verification of compliance with Regulation 1/2005, and would help to ensure that transport procedures for unbroken horses are more respectful of their ethological and physiological characteristics. This may reduce the incidence of adverse welfare consequences for horses during transportation. ABSTRACT: Regulation EC 1/2005 has stricter rules for transportation of unbroken (untamed) vs. broken (tamed) horses, but does not provide adequate tools for their identification. This study aimed to develop and validate such a tool. A behavioural test (Broken/Unbroken Test (BUT)) based on approaching, haltering, and leading was applied to 100 horses. Physiological and additional behavioural data were also collected, and the horses’ status (broken/unbroken) was assessed by the expert who administered the BUT. Each horse’s behaviour during the BUT was scored by four trained observers blinded to the horse’s history. The BUT score showed excellent inter-observer, intra-observer, and test–retest reliability (all intraclass correlation coefficients (ICCs) > 0.75). It was also negatively associated with respiratory rate, avoidance distance, and time needed to approach, halter, and lead the horse (p < 0.05 for all). The optimal BUT score cut-off for discrimination between broken and unbroken horses (gold standard: expert judgment) showed 97.8% sensitivity and 97.3% specificity. There was almost perfect agreement between BUT-based and expert classification of horses (ICC = 0.940). These findings confirm the BUT’s construct and criterion validity. The BUT could provide officials with a feasible, reliable, and valid tool to identify a horse’s broken/unbroken status and, consequently, direct stakeholders towards correct transport procedures.