Development of a user-friendly protocol and webtool for monitoring and benchmarking broiler chicken welfare during the pre-slaughter phase

The pre-slaughter phase (i.e. the phase at the end of the production period from the catching & loading of the animals until slaughter) is a critical period for broiler welfare and farmer income. We aimed to develop a pre-slaughter broiler chicken welfare monitoring protocol which meets several criteria: user-friendly, reliable, valid, discriminative, allows benchmarking of individual measures as well as an overall integrated welfare score (IWS), and based on expert opinion. We identified, from literature, 16 potential animal-based welfare measures that can be assessed at the slaughter plant. We collected data on these 16 measures for 81 transportations of commercial Belgian flocks. Three measures were removed because they were not affected by the pre-slaughter phase in our sample of transportations. In order to downscale the protocol further, another 4 measures were removed by considering the amount of time that could be saved and the relative weights that were allocated to the various measures by a panel of 9 poultry experts. The final protocol includes measures of fractures, bruising, panting/huddling, splay-legged birds, crowding, supine birds, stuck body parts, mortality and carcass rejections. Inter-observer reliability of the measures that could be tested was high (>73%). It takes circa 50-60 min for an experienced assessor per flock. In order to calculate the IWS, data were aggregated in three steps based on expert opinion (n = 17 or 19). First, observed prevalences were converted to measure scores (from 0 to 100) reflecting expert acceptability for broiler welfare. Then, measure scores were multiplied by weights reflecting the experts’ ranking of the importance of these measures for broiler welfare, and by a compensation reduction factor based on the measure score rank, and then summed. The IWS gives more weight to low measure scores, which corresponds with how experts scored overall welfare of hypothetical flocks. Application on 53 flocks showed a good spread of IWS and good sensitivity to extremely low scores. The protocol describing the precise methodology and the accompanying webtool for calculating the IWS and for benchmarking are freely available (http://survey.ilvo.vlaanderen.be:3838/WQ/). It can be used in quality assurance schemes and for identifying best practices, potential preventative actions and risk factors for broiler welfare during the pre-slaughter phase.