Animal welfare science is a relatively young scientific discipline. It is a formidable challenge for this scientific field to meet expectations from policy-makers, funding agencies, and society. They often ask scientists to provide clear-cut, unambiguous, and indisputable conclusions about specific and relevant welfare issues, for example, whether or not animal welfare is better in one type of housing/management system than in another one. People that expect scientists to provide such evidence often do not realise the complexity of this type of research question. Frequently, answers to these questions are urgently needed and solutions ought to be delivered within relatively short time-frames. Research grant applications in which it is promised that these expectations will be fulfilled quickly and cheaply, are often more likely to attract funding. Given the limited budget and time allocated to these research projects, the animal welfare problems are often investigated using a limited set of ‘standard’ welfare indicators known to be reliable for the species involved. There is rarely opportunity for developing and validating new measures that perhaps are better suited for addressing the research question concerned, nor for developing complex methodologies for integrating these different measurements into an overall assessment of animal welfare - let alone for checking that these measures and integration methods truly reflect the public's understanding. The latter, though, has been convincingly advocated to be essential for socially-constructed concepts such as ‘animal welfare’ (Fraser 2003).

The assessment and improvement of animal welfare are the tasks and joint responsibility of many stakeholders involved in the agro-food chain. This paper first looks at the supply side of the chain, and presents different stakeholder views on farm animal welfare, discussing the potential for market differentiation, communication and labelling related to farm animal welfare standards. From the demand side, the paper then examines the duality that exists between citizens’ attitudes and consumer behaviour in relation to animal welfare and livestock products, and identifies distinct segments of citizens and consumers. Although the importance that citizens claim to attach to animal welfare seems relatively strong, consumers’ interest in information about animal welfare is only moderate compared to other product attributes, and the market shares of products with a distinct animal welfare identity remain small. The paper concludes that while there seems to be substantial consensus between supply chain stakeholders, citizens and consumers about what is relevant for achieving an acceptable level of farm animal welfare, the differentiation and satisfaction potential of increased animal welfare per se as a stand-alone product attribute seems limited to particular niche market segments. It argues that improved farm animal welfare is more likely be realised and valued by consumers when it is integrated within a broader concept of quality, such as quality assurance or sustainability schemes.

There are increasing local and international pressures for farm animal welfare monitoring schemes. Housing of farm animals is a contentious issue for many, although the impact of the housing system may be overestimated by some. In contrast, the topic of stockmanship has received relatively little attention, even though research has shown that animal carers or stockpeople have a major impact on the welfare of their livestock. While welfare monitoring schemes are likely to improve animal welfare, the impact of such schemes will only be realised by recognising the limitations of stockpeople, monitoring ‘stockmanship’ and providing specific stockperson training to target key aspects of stockmanship. Appropriate strategies to recruit and train stock-people in the livestock industries are integral to safeguarding the welfare of livestock. Monitoring the key job-related characteristics of the stockperson, attitudes to animals and to working with these animals, empathy, work motivation and technical knowledge and skills, provides the opportunity to detect deficits in stockmanship and the necessity for further targeted training. Due to the strong relationships between stockperson attitudes and behaviours and animal fear responses, as well as the relationships between attitudes and other job-related characteristics, we believe attitudes, together with empathy, work motivation and technical knowledge and skills, should be the principal focus of measuring stockmanship in on-farm welfare monitoring schemes.

Animal welfare concerns stem from recognition of the fact that animals can experience emotions such as pain or joy. Nevertheless, discussion of animal emotions is often considered anthropomorphic, and there is a clear need to use explanatory frameworks to understand animals' emotions. We borrowed appraisal theories developed in cognitive psychology to study sheep emotions. Emotions are viewed as the result of how an individual evaluates a triggering situation, following a sequence of checks, including the relevance of the situation (its suddenness, familiarity, predictability, and intrinsic pleasantness), its implications for the individual (including consistency with the individual's expectations), the potential for control, and both internal and external standards. We assumed that if the outcome of checks has an impact on the animal's emotional responses, then animals do not only show emotional responses but also feel emotions. We showed that sheep use similar checks to those used by humans to evaluate their environment, ie suddenness, familiarity, predictability, consistency with expectations, and control. Furthermore, this evaluation affects their emotional responses (behavioural responses, such as startle, ear postures, and cardiac activity). It is concluded that sheep are able to experience emotions such as fear, anger, rage, despair, boredom, disgust and happiness because they use the same checks involved in such emotions as humans. For instance, despair is triggered by situations which are evaluated as sudden, unfamiliar, unpredictable, discrepant from expectations, and uncontrollable, whereas boredom results from an overly predictable environment, and all these checks have been found to affect emotional responses in sheep. These results have implications for animal welfare: although a completely invariable and totally predictable environment should be avoided to prevent boredom, sudden events should probably be minimised, the animals should be offered the possibility to control their environment, and care should be taken to ensure a degree of predictability concerning the various events.

In-transit losses (ITL) in fattening pigs refers to mortality occurring after having left the farm but prior to stunning at the abattoir. The purpose of this observational study was to identify the associations between environmental and truck temperatures, distances travelled, feed withdrawal, farm, transport company and abattoir and in-transit losses of fattening pigs marketed in Ontario, Canada from 2001 to 2004. A prospective study of 104 trips was conducted to determine temperatures inside the truck and identify the factors associated with this. In 2001, ITL was 0.017%, with 75% of producers losing < 5 pigs annually. In-transit losses increased between distances travelled of 590 to 720 km and decreased at distances greater than 980 km. The Pig Comfort Index, a combination of temperature and humidity, was used to identify thresholds of environmental conditions above which in-transit losses increased. The farm at which the pig was raised explained more variation of ITL (25%) than transport company (8%) or abattoir (16%). The within-farm ITL in 2003 had a positive association with those in 2001 and 2002. Withdrawing food prior to transport may decrease ITL on some farms. The temperature in truck compartments holding pigs increased by 0.99°C as the environmental temperature increased by 1°C and by 0.1°C as the relative humidity increased by 1%. Truck temperature decreased 0.06°C for each increase in driving speed of 10 km h−1 and increased by 7°C with an increase in pig density from one to 2.6 pigs per m2.

Animal welfare is multidimensional; its assessment relies on complementary measures covering all dimensions. Welfare Quality® constructed a multicriteria evaluation model for its assessment at unit level (farms, slaughterhouses). Four welfare principles are distinguished (‘Good feeding’, ‘Good housing’, ‘Good health’, and ‘Appropriate behaviour’). An animal unit receives four principle scores (expressed on a 0-100 value scale). These scores are aggregated together to form the overall assessment by sorting animal units into predefined welfare categories boundaried by reference profiles. A unit is assigned to the welfare category above the profile it is considered at least as good as. Several assignment procedures were tested on a set of 69 dairy farms and compared with observers’ general impressions. The welfare categories, reference profiles and assignment procedure were defined in consultation with social scientists, animal scientists and stakeholders. Four welfare categories were defined: ‘Excellent’, ‘Enhanced’, ‘Acceptable’, and ‘Not classified’. The reference profiles were set at 80, 55 and 20, corresponding to aspiration values for Excellent, Enhanced and Acceptable. The assignment procedure resulted from a compromise between theoretical opinion on what should be considered excellent, enhanced or acceptable, and what can realistically be achieved in practice: to be assigned to a given category, a unit must reach its aspiration value on 2 or 3 of the 4 principles, and not score below the aspiration value for the next lowest category on the other principle(s). The model can be used for several purposes, including identifying welfare problems on a farm to advise farmers, or checking compliance with certification schemes.

In the vast majority of European countries, piglets are surgically castrated in order to eliminate the risk of boar taint, an odour or flavour that can be present when pork from entire males is cooked. However, surgical castration is the subject of much debate and criticism as a result of its negative implications for piglets’ welfare, integrity and health. At present, there is much ongoing research into potential alternatives, among them immunocastration. This practice involves the injection of a vaccine that inhibits the production of the hormones responsible for boar taint. Although satisfactory results are associated with immunocastration in terms of meat quality and production parameters, uncertainty concerning consumer acceptance is often put forward as a key element in the quest for a successful market introduction. This study focuses on consumer awareness of piglet castration and attitudes towards immunocastration by means of a web-based questionnaire among 225 Flemish consumers. We noted approximately 40% awareness of the routine practice of castrating piglets and this limited awareness is accompanied by a moderate level of concern regarding castration, especially in comparison to food safety and other pork production system-related animal welfare issues. Sixty percent of the sample had a general appreciation for the concept of immunocastration, as opposed to surgical castration. Informing consumers about the potential benefits and/or risks from immunocastration did not tend to have much effect in terms of altering their attitudes. Immunocastration did not emerge as a problem in terms of consumer acceptance: special attention should be paid to consumers’ perception of pricing, food safety and the taste of the meat from immunocastrated pigs.

The objective of this study was to investigate inter-observer and test-retest reliability of different behavioural observations to be used in an on-farm, animal welfare monitoring system for veal calves. Twenty-three veal calf farms, varying in size, housing system, feeding regime and age of the calves were visited twice with two observers, simultaneously. Behavioural tests were conducted in eight pens per farm, measuring the response of calves to: a human entering the barn; a novel object; a passive, unfamiliar person; disturbance in the pen and an active approach by an unfamiliar and a familiar person. Furthermore, behaviour was recorded 20 min before and 20 min after feeding in eight other pens per farm. For all behavioural tests, inter-observer reliability was very high. Farm effects and test-retest reliabilities were high and significant for all behavioural tests, except for the test measuring response to disturbance in the pen. Although the active approach test with the familiar person was reliable, it was not feasible in practice due to the availability of the farmer. Since the active approach test with the unfamiliar person gave similar results, this test was recommended for an on-farm animal welfare monitoring system. For most behavioural elements recorded around feeding, farms differed significantly and interobserver and test-retest reliabilities were high as well as being significant. The behavioural tests with entering the barn, novel object and unfamiliar person, and the behavioural observations before and after feeding were feasible and distinctive and reliable enough to be performed on-farm. These methods are promising tools to use as a monitor of animal welfare in veal calves.

This study analysed the free use of phrases related to animal-based and resource-based measures of animal welfare in Swedish state animal welfare inspection reports on conventional (C) and organic (O) farms. From 244 reports by 35 inspectors, 88 were analysed as matched pairs of C and O farms (same inspector, species and size of farm). They were analysed ‘blind’ for negative comments referring to the animals or to the resources (buildings and facilities). The most commonly reported deficiencies were poor body and hoof condition and dirty animals, accounting for a total of 79% of all animal-based remarks. Deficiencies in measures and equipment or excessively high stocking density together accounted for 78% of all resource-based remarks. The total number of general (non-compulsory) comments was similar for O and C farms. But the number of (compulsory) requirements for change to comply with legislation was almost twice as high for O than C. There were significantly more comments about body condition and hooves in C than O but a tendency for the opposite to be the case for animal health. Despite this, the number of requirements for change was greater for O-farms regarding their animals. There was no difference in number of comments on resources, but once more a tendency for more requirements for change was seen on O-farms. The study demonstrates that the analysis of inspection reports can be useful in terms of identifying where, in practice, animal welfare problems lie as well as further developing the methodology of animal welfare control.

Many animal welfare traits vary on a continuous scale but are commonly scored using an ordinal scale with few categories. The rationale behind this practice is rarely stated but appears largely based on the debatable conviction that it increases data reliability. Using 54 observers of varying levels of expertise, inter-observer reliability (IOR) and user-satisfaction were compared between a 3-point ordinal scale (OS) and a continuous modified visual analogue scale with multiple anchors (VAS) for scoring lameness in dairy cattle from video. IOR was significantly better for the VAS than for the OS. IOR increased with self-reported level of expertise for the VAS, whereas for the OS it was highest for observers with a moderate level of expertise. The mean continuous scores and the mean categorical scores were highly correlated. Three times as many observers stated a preference for the VAS (n = 27) compared to the OS (n = 9) in investigating differences in lameness between herds. Contrary to common perception, these results illustrate that it is possible for a continuous cattle lameness score to be more reliable and to have greater user acceptability than a simple categorical scale. As continuous scales are also potentially more sensitive, and produce data more amenable to algebraic processing and more powerful parametric analyses, the scepticism against their application for assessing animal welfare traits should be reconsidered.

Previous studies have shown that the analysis of high frequency stress calls in pigs can serve as a reliable tool in welfare research. Our study focuses on the classification of three different classes of piglet vocalisation: grunting, squealing and screaming. In a castration experiment (Experiment 1), 3,285 vocalisations from 42 piglets were analysed for 21 different vocal characteristics. A first discriminant function for the three vocal types was derived from recordings made under laboratory-like conditions. A second discriminant function was derived from non-calibrated measurements of the relative sound energy content. These two classifications revealed 86.7% identical assignments of vocalisations to the three vocal types. The second classification allowed for vocalisation analyses of animals under on-farm recording conditions. This technique was validated during an open-field test (Experiment 2) with piglets housed in two different farrowing systems (11,089 vocalisations, 22 piglets). The proportion of screaming sounds was lower for piglets from a group-farrowing (GF) system than for those from a single-farrowing (SF) system. Sound properties showed differences between as well as within the two experiments for all three vocal types. Vocalisations from SF and GF piglets differed significantly in the duration, energy, and relative maximum levels. We conclude that vocal-type analysis can not only help to identify vocalisation indicative of pain during castration, but also vocal behaviour changes indicating separation stress during the open-field test. Therefore, classification of vocal types can add valuable information to studies that use pig vocalisation for the assessment of welfare.

The validity of an ANI-based welfare index for cattle, the A-Index, was evaluated using Test Theory methods described for quality of life assessments in human medicine. Content validity was considered during the modification of the Index for Finnish beef production. In total, 43 items evaluating locomotion abilities, lying area, social environment, management, feeding, and health of animals over six months up to slaughter were formed. Index scorings were performed on 237 farms. A most consistent partial A-Index was constructed based on qualitative item analyses. A positive relationship was established between the full A-Index and daily carcase gain. In addition, a preliminary negative relationship was established between the partial A-Index, body fat and mortality. Based on our observations, methods used for quality of life indices in human beings would appear suitable for animal welfare assessment. Practicability and well-defined methodology are the main advantages of Test Theory approach.

The aim of the present study was to investigate the effects of selection on low mortality in combination with brooding by a mother hen on open-field response at 5-6 weeks of age and on plumage and body condition at 42 weeks of age. Birds in the experiment were either selected for low mortality in group housing (low mortality line) or randomly selected (control line) for two generations. These lines originated from the same population. Twenty groups of 10 female birds from each line were used. Within each line, ten groups were brooded by a foster mother and ten groups were non-brooded. At 5-6 weeks of age, the chicks were tested in an open-field test for five minutes. At 42 weeks of age, plumage condition and incidence of comb lesions and toe wounds of all birds was recorded. It was found that both brooded chicks and chicks from the low mortality line were more active in the open-field test at 5-6 weeks of age, indicating that they were less fearful or had a stronger exploratory motivation. No interactions were found between selection on low mortality and brooding. Birds from the low mortality line also had a lower incidence of comb and toe wounds compared with the control line at 42 weeks of age. No effect of brooding on plumage condition or incidence of wounds was found. This study indicates that selection on low mortality is a promising way forward to reduce maladaptive behaviour in laying hens, especially if such an approach is combined with improved rearing conditions.

Increasing lameness problems associated with intensified dairy cattle production has lead to the development of several techniques to automatically detect these problems. Comparisons of these new measuring techniques of cow locomotion with the conventional subjective observer scoring are scarce. In order to better understand human observers’ gait scoring, cows walking on a pressure-sensitive mat were evaluated for kinematic gait variables and a visual assessment of gait was also made via video recording. Forty of these videos were used for subjective gait scoring on a 3-point scale, and the observers were also asked to report any observed abnormalities (lameness indicators) that had influenced their scoring. Relationships between reported lameness indicators and subjective gait scores, between subjective gait scores and measured kinematic variables of cow locomotion and between reported lameness indicators and measured kinematic variables of cow locomotion were investigated. In general, observers based their gait score on reported indicators such as ‘tenderness’, ‘arched back’, ‘irregular gait’ and ‘increased abduction’. All of these four reported lameness indicators were correlated with measured kinematic ‘variables of asymmetry’, ‘stance time’ or both, suggesting that human observers are capable of detecting changes within these lameness indicators as measured by the pressure-sensitive mat. ‘Increased abduction’ appeared harder to detect and was reported more frequently by observers already experienced with gait scoring. Also, the measured kinematic variables of ‘stance time’ and ‘measures of asymmetry between left and right limbs’ as measured by the pressure-sensitive mat, show potential in predicting the gait score given. These reported lameness indicators and measured kinematic variables —mutually correlated and both related to the gait scores — were considered promising for subjective gait scoring in general.

The Welfare Quality® project aims to develop a European on-farm welfare assessment standard for pigs, amongst other species. A prototype monitoring system was developed for sows and piglets using predominantly animal-based measures of behaviour, health and physiology to assess welfare. The prototype monitoring system was evaluated on a total of 82 farms in the UK and The Netherlands, encompassing a wide variety of farming systems. Preliminary analysis of the data indicates that the incidence of clinical welfare problems were low, with 1.2, 0.8, 1.1 and 0.1%, respectively of pregnant and lactating sows recorded as having substantial skin lesions, bursitis and vulval lesions, and extremely poor body condition. Some clinical problems were more prevalent in certain types of feeding system than others, particularly the severity of vulval lesions with electronic sow feeding (ESF) systems. Fear of humans, assessed by the extent of withdrawal behaviour from an unknown human, was low, and median scores were similar for sows housed in groups indoors, outdoors and in stalls. Some form of stereotyped behaviour was observed on almost 75% of farms visited, with sham chewing the most commonly recorded stereotyped behaviour. System design affected the practicality of some measures, particularly on outdoor farms. Full investigation of the complete data set will enable a refined, final, on-farm monitoring system to be developed and benchmark standards established.

This paper discusses the current state of development of on-farm cattle welfare assessment systems with special regard to the approach of Welfare Quality® that focuses on animal-related measures. The central criteria validity, reliability and feasibility are considered with regard to selected welfare measures. All welfare measures incorporated into the Welfare Quality® protocol possess face validity, but for most of them construct or criterion validity as, eg shown for lameness, have not been demonstrated. Exemplarily the cases of qualitative behaviour assessment and measurement of avoidance distance towards humans or social licking are discussed. Reliability issues have often been neglected in the past and require more thorough investigation and discussion in the future, especially with respect to appropriate test statistics and limits of acceptability. Means of improving reliability are the refinement of definitions or recording methods and training. Consistency of results over time requires further attention, especially if farms are to be certified, based on infrequent recordings. Considering feasibility, time constraints are the main concern for assessment systems that focus seriously on animal-based measures; currently they require several hours of on-farm recordings, eg about 6 h for a herd of 60 dairy cows. The Welfare Quality® project has promoted knowledge and discussion about validity, reliability and feasibility issues. Many welfare measures applied in the Welfare Quality® on-farm assessment approach can be regarded sufficiently valid, reliable and feasible. However, there are still a considerable number of challenges. They should be tackled while using the present assessment system in order to constantly improve it.

Merino ewes have been selected, over 18 generations, for calm (C) or nervous (N) temperament using using an arena test and an isolation box test. We investigated the relative contributions of genotype versus the post-partum behaviour of the dam on the temperament of the lambs using a cross-fostering procedure. Forty-eight multiparous calm and 52 nervous ewes were artificially inseminated with the semen of a sire of the same temperament. At birth, 32 lambs of a given temperament line were cross fostered to ewes from the other line (16 N × C, 16 C × N), 34 lambs were cross fostered to ewes from the same line (15 C × C, 19 N × N) and 30 lambs were left with their birth mother (15 C, 15 N), to control for the effect of cross fostering. The temperament of the progeny was assessed at two occasions, one week after birth by measuring locomotor activity during an open-field test and at weaning (16 weeks) by measuring locomotor activity during an arena test and agitation score measured during an isolation box test. There was a genotype effect but no maternal or fostering effect on the lamb temperament at one week. This may be because the maternal behaviour of the foster ewes did not differ considerably between the calm and nervous mothers during adoption or within the first week, post partum. Similarly, at weaning, only a genotype effect was found on the locomotor and agitation score. Therefore, it appears that temperament in Merino sheep is mainly determined by the genetic transmission of the trait across generations rather than behaviours learned from the mother.

The greatest challenge to the welfare of dairy cows occurs in the peripartum period. Given the perception that cow welfare is better in more natural environments, it was hypothesised that cows in a PASTURE-based production system (cubicle housing with grass silage pre-partum and rotational grazing with concentrate supplementation post-partum) would have improved peripartal welfare compared to cows in a HOUSED production system (cubicle housing with a total mixed ration [TMR], pre-partum and post-partum). Blood samples were analysed for acute phase proteins (APP), cortisol, white blood cell (WBC) differential and counts and other biochemical metabolites as non-specific indicators of sub-clinical ill-health and nutritional stress. Daily monitoring of rectal temperature (RT) and rumen fill (RF) scores were used to monitor ill-health and nutritional status. Reproductive health and welfare (calving difficulty, retained placenta, puerperal metritis, endometritis and oestrous cyclicity) was also recorded. No differences were found between treatments for APP, cortisol or WBC. Blood metabolite differences indicated that PASTURE cows were under greater nutritional stress than HOUSED cows. HOUSED cows showed an increase in RF score from day 0 to 10 post-partum and had a higher RF score than PASTURE cows. PASTURE cows had an overall lower RT and lower incidence of reproductive disorders. Results primarily reflect nutritional differences between treatments with PASTURE cows showing greater potential nutritional/metabolic stress in early lactation which has attendant implications for welfare. Nevertheless, this did not result in inferior health and, in accordance with our hypothesis, PASTURE cows’ reproductive health and welfare tended to be better than that of HOUSED cows.

One million pet rabbits are kept in The Netherlands, but there are no data available on their behaviour and welfare. This study seeks to assess the welfare of pet rabbits in Dutch households and is a first step in the development of a welfare assessment system. In an internet survey, housing systems, general up-keep and behaviour of pet rabbits were reported by their owners. The answers of 912 respondents were analysed with behavioural observations carried out on 66 rabbits in as many households. The rabbits were observed in their home cage and during three fear-related tests: a contact test, a handling test and an open-field test. The survey revealed that the average lifespan of the rabbits is approximately 4.2 years (the maximum potential lifespan is 13 years) and solitary housing appears to reduce lifespan. Close to half of respondents subjected their rabbit(s) to solitary housing and the majority housed them in relatively small cages (< 5,000 cm2). Health risks may arise from a failure to inoculate rabbits and via inappropriate diet. During the contact test, solitary-housed rabbits made more contacts with a human than group-housed rabbits and rabbits in a small housing system made more contacts than those in a large system. Observations in the home cage differed greatly compared with the natural time budget of rabbits, ie displaying increased stereotypic behaviour and decreased foraging and, in solitary-housed rabbits, a complete lack of social behaviour. Nearly 25% of rabbits displayed strong resistance to being picked up, indicating socialisation problems. During the open-field test, solitary-housed rabbits sat up more than social-housed rabbits suggesting increased fearfulness. These findings indicate that the conditions in which pet rabbits are kept often have a negative impact on their welfare, further underlining the need to study this in greater detail.

The aim of this article is to discuss the farming industry's development and use of welfare assessment schemes. A welfare assessment scheme developed by the Danish Cattle Federation (DCF) is used as a case study. The declared aim of the DCF scheme is to improve animal welfare, farm profitability and dialogue with the public. It is the purpose of this article to attempt to understand the dilemmas arising from this broad aim. We ask how DCF measures of welfare compare with alternative measures in which economic factors receive less emphasis. We bring in farmers’ views on whether the DCF's parameters of welfare track welfare effectively and are economically feasible. We also discuss how the views of the Scandinavian public on animal welfare influences the likelihood that the DCF scheme will improve dialogue with the public. The DCF definition of welfare is broad, but the measures it deploys are more limited and indeed very narrow compared with those in other welfare assessment schemes. This may not be a problem if the goal is to improve farm profitability. However, if the goal is to improve welfare, limited measures are problematic, and this may undermine attempts to improve dialogue with the public.