Welfare assessment in broiler farms: Transect walks versus individual scoring

Current scientific approaches to welfare assessment in broilers are based on individual sampling that can be time consuming under field conditions. On the other hand, farmers conduct routine checks based on walks through the house to screen birds' health con- dition. We adapted the walks through following line transect methodology used in wildlife studies to explore their feasibility as a welfare assessment tool. The aim of this study was to compare broiler welfare assessed by individual sampling and transect walks. We evaluated 6 identically managed flocks. For individual sampling, we collected measures on 150 birds, including weight, breast dirtiness, hock and footpad dermatitis, lame- ness, and immobility. Transect observations were con- ducted by slowly walking on randomized paths within each house recording: immobility, lameness, back dirti- ness, sickness, agony, and dead. Transect walks al- lowed detection of small variations (P < 0.003) in the prevalence of most welfare indicators considered with consistency in interobserver reliability (P ≥ 0.05). In addition, assessments across transects were highly con- sistent (P ≥ 0.05). Individual sampling was also sensi- tive to differences across houses (P < 0.01) with the ex- ception of immobility (P = 0.783). No differences were found across sampling locations (P ≥ 0.05). However, both methods differed greatly in the frequency of the incidence of the parameters considered. For example, immobility varied from 0.2 ± 0.02% for transect walks to 4 ± 2.3% for individual sampling, whereas lame- ness varied between 0.8 ± 0.07% and 24.2 ± 4.7% for transect and samplings, respectively. It is possible that the transect approach may have overlooked walking de- ficiencies because a large number of birds were scored, although if this was the case, the consistency obtained in the scoring across observers and transects would be surprising. Differences may also be related to possibly biased individual sampling procedures, where less mo- bile and passive individuals may be more likely to be caught. Furthermore the procedure may cause fatigue and fear reactions reducing mobility. Current study provides new insights into constraints and advantages of broiler on-farm assessment methods, which should be considered for designing on-farm welfare assessment protocols