Wood-chip pads represent a low-cost alternative to housing for cattle during the winter. Considering the negative welfare implications associated with housing indoors on concrete, they may also offer welfare benefits to replacement dairy heifers. However, these animals may not be able to withstand winter weather conditions on a grass silage diet. The aim of this experiment was to evaluate behaviour, limb injuries, dirtiness scores, performance and climatic energy demand (CED) of yearling dairy heifers on two levels of nutrition kept outdoors on a wood-chip pad or indoors in cubicles during the winter. Ninety-six 10-month-old heifers were blocked and assigned in groups of eight, to one of the following four treatments in a 2 3 2 factorial design: (a) indoors, silage only; (b) indoors, silage plus concentrate; (c) outdoors, silage only; and (d) outdoors, silage plus concentrate. There were three replicate groups per treatment. All animals were inspected for skin lesions and were weighed and body condition scored (BCS) at the beginning and end of the trial. Instantaneous scan sampling and continuous all-occurrence behaviour sampling were used to collect behaviour data during two 24-h periods. Animals were also dirtiness scored and group feed intakes were recorded during the trial. Significantly more comfort, social and play behaviours were recorded outdoors (P , 0.05) while trips, slips and falls were only recorded indoors (P , 0.001). Groups outdoors had significantly lower limb lesion scores at the end of the experiment (P , 0.05) and fewer groups outdoors were affected by all categories of limb lesions. However, they were consistently dirtier than animals indoors (P , 0.001). Low-nutrition animals had lower feed intakes, smaller BCS changes and lower average daily weight gains than high-nutrition animals (P , 0.01). Heifers outdoors had significantly lower average daily weight gains and BCS changes (P , 0.05) explained by lower feed intakes (P , 0.01). However, outdoor heifers on both the high-and low-nutrition diets and indoor animals on the low-nutrition diet had lower UFL (feed unit for maintenance and lactation (Irish Republic)) intakes (20.36, 20.35 and 20.22, respectively) than that required to meet the daily live-weight gains they achieved. Heifers indoors on the high-nutrition diet gained 0.98 kg per day but consumed 0.17 UFL more than what would be recommended to achieve a daily weight gain of 1.0 kg. The CED for outdoor heifers was higher than that of indoor heifers (6.18 v. 5.47 MJ/day per m 2 body surface area; P , 0.001, s.e.d. 0.044). However, CED did not exceed heat production in any treatment. Although animal performance was reduced outdoors, the wood-chip pad was associated with welfare benefits compared with cubicle housing.

Incorporating indicators for pig health and welfare at meat inspection could reduce the need for on-farm assessments. Skin and tail lesions are important welfare indicators in pigs with good potential to record during meat inspection and could possibly function as iceberg indicators of on farm welfare. The aim of this study was to validate the use of these carcass lesions at meat inspection for the assessment of pig health and welfare on farm. Thirty-one farrow-to-finish pig farms (∼12% of Irish herds) were assessed using an adapted version of the Welfare Quality protocol by inspecting 6 randomly selected pens of pigs in the first weaner (4 to 8 wk), second weaner (8 to 13 wk) and finisher stage (13 to 23 wk). The average prevalence of welfare outcomes for each stage was calculated. One batch of pigs was observed at slaughter and skin and tail lesions were scored according to severity for each carcass. The average prevalence of carcass lesion outcomes was calculated for each farm. Linear regression models were developed to predict the prevalence of each welfare outcome in each stage based on the prevalence of the different carcass lesions. The welfare outcomes of different welfare aspects that were best predicted by abattoir information (highest ) were poor body condition (first weaner stage), bursitis (second weaner stage), huddling (first weaner stage), severe tail lesions (finisher stage) and coughing (second weaner stage). Regression trees and receiver-operating curves (ROC) were used to evaluate the usefulness of carcass lesions as monitoring tools. Receiver-operating curves were created using the 75th percentile to classify farms as a problem farm for these welfare outcomes. Cut-off values of predictive carcass lesion prevalence were similar using both techniques. Models for predicting problem farms with poor body condition, bursitis and severe tail lesions were moderately accurate. Sensitivity and specificity ranged from 75 to 100% and 70 to 87%, respectively at the optimal cut-off value of the predictive carcass lesion prevalence. Results show potential for using carcass skin and tail lesions as iceberg indicators of pig health and welfare on farm. Future work is needed to evaluate the cost of including carcass lesion recording at meat inspection, the cost of failing to identify problem farms and the cost of incorrectly visiting or penalizing problem farms before carcass lesions can be used as welfare indicators in a commercial setting.

Tail biting is common on farms that are more prone to disease. Tail lesions can be associated with poor health either directly by providing a route of entry for pathogens or indirectly as conditions which trigger tail biting may also affect pig health. The aim of this study was to investigate the relationship between carcass tail lesion and lung lesion severity scores in Irish slaughter pigs. Factory visits occurred over 5 days (Jan-March 2015) and tail lesion score (0-4) according to severity, sex, and kill number was recorded for every pig after scalding and dehairing. The lungs from each carcass were scored for lesions using an adapted version of the BPEX pig health scheme. Presence of pleuropneumonia (APP), abscesses and pyaemia was recorded. The severity of enzootic pneumonia (EP-like lesions) was recorded on a scale of 0-50 and grouped into none/ mild (score 0-10), moderate (score 11-20) and severe (score 21-50). Severity of pleurisy was scored on a 0-2 scale with a separate variable for lungs that were attached to the chest wall (score 2). These lungs could not be assessed for other lung lesions and therefore a separate database was created for assessing the pleurisy lesions (n = 5,628) while the final database contained animals (n = 4,491) with records for tail lesions, EP-like lesion scores and pleurisy (level 0 + 1). Associations between tail lesions and sex and the different lung lesion outcomes were analysed using generalized linear mixed models (PROC GLIMMIX) with random effect for batch. EP-like lesions and pleurisy were the most common lesions, though large variation existed between batches. Pigs with severe tail lesions tended to have more severe pleurisy where the lungs remained attached to the chest wall than pigs with moderate tail lesions. No other associations between tail lesion scores and lung lesions were found. Although tail lesions on the carcass may not be an accurate predictor of lung health, tail lesions are important welfare indicators and respiratory disease is among the most significant infectious conditions affecting pigs. Thus, recording of tail and lung lesions at meat inspection provides valuable information regarding the on-farm health welfare of pigs.