Desenvolvimento, validação e aplicabilidade do "check-list" para avaliação preliminar do grau de bem-estar animal de cães.

Abstract

Human society has reframed the relationships established with animals, replacing the anthropocentric with a defensive approach and, despite this inter-species relationship dating from the beginnings of human history and providing numerous and varied benefits, negative interactions are recurrent. Thus, there is a growing social demand for the authorities to act, in the face of denunciations of criminal actions, carried out against animals, which are based on scientific studies on complex subjects such as Animal Welfare (AW) and Maltreatment. The present work proposed the development and validation of a tool for preliminary assessment of the degree ofAW. To this end, a “check-list” was created that considered the specificities of users and species to formulate their requirements. For validation, psychometric prerogatives were used. Thus they were created and evaluated three scenarios that mimicked the degrees of AW considered Low / Very Low, Regular and High / Very high. Data were collected during April 2020, totaling 85 assessments, of which two were observations made by judges. All variables were compiled in Microsoft Excel® and statistical analyzes conducted in Stata Statistical Software: Release 14®. The internal consistency of each item was estimated using Cronbach's Alpha; nutritional indicator showed the coefficient of 0.839; the comfort indicator of 0.7139; the health indicator of 0.8412 and the behavioral indicator of 0.637. For the equivalence of the instrument, the general kappa presented a value of 0.9034, considering 95% of confidence. The p-value was considered significant. Within the evaluation of the judges or the apparent validation, the judgment of 04 observers was obtained. The overall average of the tool, with regard to objectivity, clarity and relevance, was, respectively: 4.9, 4.78 and 4.84. IVC's observed for the standards of objectivity, clarity and relevance were: 97.6%, 96% and 96%. The validity of the predictive criterion was verified by comparing the proposed tool, the “check-list”, and the Animal Welfare Expertise Protocol. We must highlight that the instrument on screen reached the values of 100% sensitivity and 93% specificity. The positive and negative predictive values were 88 and 100%, respectively. The accuracy presented was 83.61% and the kappa considered the agreement to be moderate (0.55). The positive likelihood ratio found was 14.90. It was noted that the average time to respond to the 3 proposed scenarios was 10 minutes and 51 seconds, with a general average in each scenario being 03 minutes and 33 seconds. This study showed that the "checklist" to evaluate the degree of preliminary AW domestic canids it is a reliable tool and psychometric and statistical qualities that allow us to say whether the animal is assessed in a suggestible state of ill-treatment.