Simplifying the Welfare Quality® assessment protocol for broilers

The European Welfare Quality® project developed standardized animal welfare assessment methods for different categories of farm animals, including broiler chickens. One of the key characteristics of the Welfare Quality® assessment protocols is that it places more focus on animal based measures (i.e. injuries or behaviour) than on design or management criteria (i.e. pen size). Dutch stakeholders have expressed their interest in the assessment protocols for different types of farm animals, but have also suggested that a reduction in performance time may improve the practical applicability of the assessment protocol and improve the probability of adoption of the welfare assessment protocol in practice. Therefore, the aim of the current project was to determine whether or not there is scope for simplification of the broiler assessment protocol by reduction in performance time. According to the standard broiler assessment protocol (Welfare Quality®, 2009), data were collected from 180 broiler flocks, of different breeds and housed under different conditions. Slaughter plant visits were performed for 150 flocks in addition to the on-farm measurements as described in the full assessment protocol. Correlations between animal-based measurements (on-farm as well as at the slaughter plant) were analysed to determine if there were possibilities for simplification of the protocol and based on this, it was decided that there were two possible strategies for simplification. For all flocks, end scores were calculated based on the calculations described in the full assessment protocol. Thereafter, possible strategies for simplification were analysed by (a) calculations of predictions of criterion scores for each strategy of simplification from prevalence for individual measures, (b) comparison of the golden standard (full protocol) with simplification strategy 1 and 2 at the level of final flock score, and (c) comparison of the golden standard with simplification strategy 1 and 2 at the level of principles and criteria. Analysis of correlations between animal based measurements showed no high correlations for on-farm measurements (r0.7) were found for foot pad dermatitis measured on- farm and at the slaughter plant. Therefore, a potential second strategy for simplification, i.e. replacing on-farm measures with slaughter plant measures, was analysed. In this simplification strategy, clinical scores (foot pad dermatitis, hock burn and cleanliness) and gait score were predicted from slaughter plant measurements (foot pad dermatitis and hock burn). Analysis of simplification strategies showed that there was in general close agreement on the level of flock score, as well as on the level of principle and criterion scores. In addition, there was generally a high correlation between the golden standard (the full assessment protocol) and the simplified model on principle and criterion level. Both strategies for simplification of the broiler assessment protocol appear promising regarding the potential for reduction in performance time essential for improvement of the probability of acceptance for implementation in practice. It is advised to validate the results of the data-based simplification strategies in a further study, preferably in flocks that are more widely distributed over the different end score categories, before implementation of the simplification strategies in practice. With respect to the assessment of welfare in practice, the results of this project are a major step forward to practical implementation of a welfare assessment protocol for broilers.