文摘
At a slaughter plant, cattle are sometimes exposed to rough handling which may reduce animal welfare (AW). In an observational study at four Swedish commercial slaughter plants, AW-related behaviours of cattle and actions of abattoir stockpersons handling the same animals were recorded simultaneously. The objective was to estimate the occurrence of different behaviours and actions related to negative AW during driving and stunning at large-scale cattle abattoirs, assess associations between such behaviours and actions, and analyse differences between plants and animal categories (dairy cows, beef cows, adult bulls and heifers/bullocks). Direct continuous observations of focal animals were made using laptops either in a section of the driving race to the stun box (132 animals) or in the stun box (313 animals), generating a total of 14.5 h of observations. The animals were stunned using a penetrating captive bolt gun or a rifle. Counts per animal of 14 behaviours and 16 stockperson actions were calculated. Sixteen percent of the observed animals displayed total behaviour counts >5 in the driving race, and 2% did so in the stun box; 32 and 8% of the observed animals received total counts >5 of stockperson actions in the race and stun box, respectively. We estimated that two-thirds of the animals were processed without displaying/receiving any of the behaviours/actions associated with severely negative AW. AW scores were acquired by adding together all observed behaviour counts (and action counts, separately) weighted by expert-assessed ratings denoting the degree of impaired AW. Spearman rank correlation was used to analyse associations between behaviour counts, action counts and AW scores. Only three moderate to strong correlations (蟻 鈮?#xA0;0.4, P 鈮?#xA0;0.001) between single behaviours/actions were found (鈥渟lapping rear鈥?and 鈥渟lapping front鈥? 鈥減rodding鈥?and 鈥渟houting鈥? and 鈥減rodding鈥?and 鈥渂eating rear鈥?in the driving race). The correlation between AW scores based on behaviours and actions was statistically significant but rather weak both in the driving race (蟻 = 0.37, P < 0.0001) and stun box (蟻 = 0.22, P = 0.0002). The effects of slaughter plant and animal category on behaviour counts and AW scores were estimated using standard or zero-inflated negative-binomial regression. The risks of most behaviours related to negative AW differed considerably between plants. In the stun box, adult bulls had a 2.5 times higher risk of 鈥渟truggling-kicking鈥?(P = 0.016) and a 2.0 times higher risk of displaying 鈥渂acking-turning鈥?(P = 0.016) than had dairy cows, indicating poorer welfare for the bulls.
