Accuracy and reliability of two body weight estimators based on linear measurements in horses

Abstract

Estimation methods are routinely used to determine the body weight of a horse when a scale is not available. However, it is important to evaluate the accuracy and reliability of these predictions since nutritional management and drug dosage, among others, require knowledge of the animal€™s weight. The objective of this study was to evaluate the accuracy and reliability of estimated weights of horses using two prediction models, and to discuss the effect of the predictions on animal management. Seventy-one adult male animals (40 mules and 31 horses) were weighed on a precision scale (control weight). Chest circumference and body length were measured and used to estimate body weight using the two models. Accuracy was evaluated by analysis of fit indices, comparison of squared prediction errors, delta Akaike€™s information criterion, and decomposition of the mean squared error of prediction. For economic evaluation, three production scenarios were established: simple (worming + forage), traditional (worming + forage + concentrate mixture), and traditional with supplementation (worming + forage + concentrate mixture + supplementation). Economic values were collected through a pricing survey carried out in the Midwest region of Brazil and converted to US dollars. Percent differences in costs were evaluated using Fisher€™s exact test. The most suitable model for the morphometric profile of the animals was that using a combination of chest circumference and body length. The differences in economic values between production scenarios were significant (P<0.05). Losses using the inadequate model are 10% higher compared to control; thus, in a facility with 20 equids that uses this model, the losses are ‰ˆ US$ 10,000/year. When investment in a scale is not possible, it is fundamental to evaluate which model is more compatible with the body biotype of the herd since all types of management that depend on the body weight of the animals may result in significant economic losses.