Reliability of palpation using three-dimensional keel bone models

Abstract

Keel bone damage (KBD), including both deviations and fractures, is an important and wide-spread animal welfare problem of laying hens especially in non-cage housing systems. Diagnosis of its prevalence is the first step in finding solutions. The aim of this study was to increase the reliability of diagnosis of KBD by palpation using three-dimensional (3D) keel bone models. An online training session was organized about how to use 3D models of 3 different severity levels of keel bone damage as a tagged tactile analogue scale. After the online training took place, the participating groups were asked to score the keel bone damage levels of 20 laying hens at each of their respective facilities with and without using the models (40 hens in total) on a continuous scale. The repeatabilities of the scoring results of 37 people from 11 groups (countries) were computed with the R-package rptR to test whether the use of models increased the reliability of the scores. In general, the repeatabilities of the scores on a continuous and a binary scale were low. Contrary to expectations, the use of 3D models did not increase the reliability in scoring KBD by palpation. In future studies with these models, a more even distribution of KBD among groups and on-site instead of online training might give more reliable results.