Transect walks : method sensitivity for on-farm welfare evaluation in turkeys

There is increasing consumer demands for livestock and poultry products that meet minimum animal welfare standards during their production cycle. Additionally, a growing number of farmers are aware about full compliance with the animal welfare standards that could play an important economic role in commercial intensive productions. Indeed, animal welfare assessment protocols have meaningful effects to promote and guarantee the on-farm safeguard animal standards. Transect walks method appeared to provide a practical approach to welfare assessment in broilers farms. Because of the similarities of meat poultry production, this method could be considered a reasonable approach for turkey welfare evaluation in terms of time demands, within costs and feasibility. The aim of this study was to determine the sensitivity of this method at 3 commercial turkey farms with similar facilities (Animal Welfare Indicators project, FP7-KBBE-2010-4). In this study, 10 commercial female-turkey flocks (6 houses with identical management), ranging from 3,212 to 6,000 birds and belonging to the same company were evaluated one week before slaughter. On turkey farms, walking through the house is a routine daily procedure to check the health status of the birds. Two previously trained assessors in performing the transect methodology and properly assessing the selected indicators, evaluated each paired house sequentially and independently within the same day by walking through predefined transect bands (1 to 4) in random order. The animal-based indicators considered were: immobility, lameness, wounds and featherless. The statistical model used was GLM in the GENMOD procedure (SAS V9). The results showed that this welfare assessment method highlights even small variation among houses for the considered variables. In fact, there were significant difference across houses (P=0.0021, P=0.0006, P=0.0081 and P=0.0003) for all the evaluated indicators respectively. Differences across observers were detected for wounds and featherless (P<0.0001). These results may be due to the difficulties in assessing equally these parameters while walking. On the contrary, lameness, probably the most important welfare problem in meat poultry, showed a good concordance between observers (P= 0.361). These preliminary findings suggest that this new approach has potential as a tool for on-farm welfare evaluation, which may be worthwhile to be further developed.