42 What is Animal Welfare and How is it Assessed?

Sustainable

Agriculture

Editor: Christine Jakobsson

Sustainable Agriculture

Defined by Broom (1986): The welfare of an animal is its

state as regards to its attempts to cope with its environ-ment.

So, what does this mean? Animals have to cope with a complex environment, such as high temperature, hunger, fear of predation. Animals have a range of means of cop-ing with these stressors, principally through the adrenal stress response. This is the same mechanism of response – the flight or fight response – irrespective of the type of stressor. Increased respiration rate, heart rate, increased blood supply to the muscles, erection of hair, pupil dila-tion. These physiological changes are linked to behav-ioural responses that ameliorate the stressors: move to a cooler shaded area, find a food source, escape from the predator.

Stress is therefore an entirely normal response to challenges from the environment, and animals are able to cope with small amounts of stress, and with stressors from which they can escape. Stress is therefore not nec-essarily a cause of poor welfare, but distress does have a negative effect on welfare.

However, the animal may be unable to cope. There may be no shaded area, no food available, and flight from the predator may not be possible. The animal’s welfare, and wellbeing will then be compromised, and this leads

What is Animal Welfare

and How is it Assessed?

David Arney

to a chronic stress response, related to raised levels of cortisol. In this case the animal’s fitness may be reduced, it may not grow and it may fail to reproduce. We can rank the state of welfare from poor to good, using a range of measures, rather than all or nothing.

Measures that are commonly used include:

• Fertility; but some species, notably dogs, can repro-duce prolifically in very poor conditions. Indeed, almost by definition domesticated animals will do so, as they have been selected to produce large numbers of offspring in captivity.

• Disease; but this, and the following two on the list, do not necessarily indicate poor conditions.

• Morbidity. • Mortality.

• Levels of stress hormones.

• Behavioural indicators; but these may be learnt and not indicate current welfare conditions.

Individual parameters may be caused by incidental fac-tors, unrelated to the welfare conditions prevailing on an animal unit. Measurements should be collected to pro-duce a portfolio of evidence rather than reliance on a sin-gle measure. The first four of these are simply a matter of

record. The capture of stress hormones is made difficult as the nature of collecting them can itself be stress-induc-ing. However, residues can be collected from faeces and saliva as well as directly from blood.

Preferred methods of assessing welfare are behaviour-al measures, as they are:

• Easy to use • Non-invasive • Non-intrusive

• Expressions of emotions • Reflect first attempts to cope • More sensitive than injury or disease Behavioural measures include:

• Abnormal behaviour (Figure 42.1).

• Comparison with normal behaviour (such as time budgets). The researcher will need to know the nor-mal behaviour profile in order to compare differences (Figure 42.2).

• Incidents of aggressive behaviour (Figure 42.3) • Stereotypies: repetitive patterns of behaviour with no

obvious function (Figures 42.4 and 42.5). • Increased vocalisations

• Choice tests

Problems of behavioural measures of stress:

• Not always reliable: some of the abnormal behaviours may be learnt behaviours resulting from previous stress events that are no longer extant. Example: crib-biting by a horse may be a learnt behaviour from a previous stable.

• Human interpretation of the cause or purpose of some of these behaviours may be inaccurate. We can guess, but our perception is limited by our human experi-ence.

• Individual animals will respond differently to the same source of stress. For example, individual animals have different temperature thresholds, as do individual humans.

Figure 42.1. On this farm several cows were observed standing in the drinking trough. This may have been a means of cooling inflamed hooves. These were treated and the behaviour ceased. Photo: M. Ots.

Figure 42.2. Piglets do not normally lie on the sow’s belly. If they do so this indicates that the sow may have a medical problem. Photo: A. Pavlenko.

• Stereotypies may be a method of coping with stress: Wiepkema et al. (1987) showed that veal calves per-forming stereotypies more often had fewer abomasal ulcers than the other calves.

However, stereotypies are often observed in animals in poor housing conditions. Broom and Johnson (1993) pro-posed the following assessment tool (See fact box 1).

Choice tests involve giving animals a choice between two or more environments or resources, such as offering dairy cows the choice of entering a cubicle area or a straw yard. The assumption is that the animal will choose the option that is in its best interest. Problems with choice tests include:

• The choice may satisfy some transitory need or pref-erence. An animal may make the choice in its short-term interest at the expense of its long-short-term interest. • Individual animals may have individual preferences

not indicative of the group as a whole.

• Choices may change over time: age of the animal, time of day, stage of reproductive cycle.

• A choice may be preferred, but actually be poor for welfare: e.g. the choice of a sugar-rich food by rats does not indicate that these should form the diet of the housed rat.

• The choices offered are relative. They do not indicate good or poor welfare, only that one of the options may be better than the other. Both options may be good or bad for the animal’s welfare.

Choice tests can be refined to indicate the strength of preference for one option over another. This can help to determine whether a resource is a necessity or a luxury. The less favoured option can be baited with a desirable resource (often food). The favoured option can be en-cumbered with difficulty or unpleasantness (the length of a darkened walkway in the case of the dairy cow, or the length of a cat urine-soaked walkway in the case of the laboratory mouse). Animals can be trained to use an operant device that can be calibrated to altered levels of work required by the animal to choose one option over another.

The UK farm animal welfare council suggested five freedoms that should be satisfied for an animal to be re-garded as being in a state of good welfare. These were first formulated in 1979, principally regarding farm ani-mals, but there is a general consensus of agreement among animal welfare scientists that they are a useful guide to assessing the welfare of an animal. The five freedoms re-main the basis on which the European Union frames its animal welfare policy. The five freedoms are:

Figure 42.3. Tail biting, a common outcome of aggressive behav-iour shown by pigs in confined, barren housing conditions. Photo: A. Pavlenko.

Figure 42.4. Sterotypic behaviour; tongue-rolling shown by a housed dairy cow. Photo: A. Pavlenko.

1. Freedom from hunger and thirst, access to fresh

water and a diet for full health and vigour.

2. Freedom from discomfort, an appropriate

environ-ment with shelter, and comfortable rest areas.

3. Freedom from pain, injury and disease, prevention

or rapid treatment.

4. Freedom from fear and distress, conditions and

treatment which avoid mental suffering.

5. Freedom to express normal behaviour, adequate

space and facilities, company of animal’s own kind. (European Commission, 2007) More recently, Bartussek (2001) has proposed an animal needs index that uses a scoring system for a variety of fac-tors, leading to a sum that is designed to represent a mean-ingful welfare assessment of housing conditions. There are five broad sections to this system of assessment: 1. Possibility of mobility

2. Social contact 3. Condition of flooring

4. Stable climate (including light and noise) 5. Quality of stockman’s care

However, this system does not seem to have been adopted much outside its native Austria.

There is a method for determining the priorities of animals for resources through measuring their motivation for access to the resources. The concept is derived from human behaviour expressed by the economic theory of demand function (Dawkins, 1983; Hursh, 1984). In the context of animal behaviour, ‘cost’ is the amount of time spent on an activity or work the animal is prepared to do to achieve the resource. For example; rats pushing a lever or chickens or mink pushing through an increas-ingly weighted gate to access nesting material. This de-mand may be described as elastic if the animal adjusts the price it is prepared to pay for differing quantities of the resource, such as space. As the amount of work, or cost, increases, the animal becomes less prepared to pay the cost in order to achieve the resource. Plotting the out-comes on a logarithmic scale produces a straight line, the demand function. Resources producing this kind of response are typically described as luxury resources. In contrast, an inelastic demand is for an essential resource, such as food. Changing the amount of work the animal has to do to achieve an essential resource has little ef-fect – the price will continue to be paid. In addition, the expenditure an animal is prepared to make, in amount of time or work, can be used to rank the importance of a range of resources. The steepness of the demand function gradient indicates the degree of motivation. The resource may be the opportunity to perform a behavioural

activ-FACT BOX 1

Relationship between stereotypies and animal welfare.

Stereotypies Welfare

Occasional stereotypies caused Very good by minor frustration

Stereotypies for 5% of active time

Stereotypies for 40% of active time Very poor

adapted from Broom and Johnson (1993)

Figure 42.5. Stereotypic behaviour; Nose-pressing shown by a housed dairy cow. Photo: A. Pavlenko.

ity, such as digging, swimming or even rest. Criticisms of this method include: the artificial nature of the test, the influence of prior deprivation, prior experience and external factors, including the time of day.

Conclusions

The welfare of animals concerns not simply stress, ex-perienced by an animal, but its ability to manage stress, whether it be physical or mental stress. Animals in the wild and in captivity are exposed to many and varied stressors. If the stressor continues despite an animal’s ef-forts to remove it, we can consider the animal to be in a poor state of welfare. Welfare can be ranked; there is not simply good or bad welfare, but many gradations of wellbeing. We can measure and record many factors that can indicate poor or good welfare, and the more meas-ures we take the more likely we are to make an accurate assessment. Behavioural measures are particularly good as they are, among other things, relatively easy to use, are (or ought to be) non-intrusive, show an animal’s first responses to cope and are sensitive. However, caution should be used with such measures as they can be mis-leading to a human observer. Stereotypies can also indi-cate poor welfare, and particularly a poor environment. Letting an animal choose from options, using a choice test, can give us clues as to the relative preference of the options offered, and this can be refined to include posi-tive and negaposi-tive outcomes to assess the strength of pref-erence of one option over another. The five freedoms are an enduring guide to the assessment of welfare, but other systems are also in use. An index system of scoring wel-fare factors and deriving an overall index of welwel-fare has been developed, though there is some doubt as to whether an animal’s welfare can be adequately expressed by a sin-gle number.

decreased ammonia emission. In: Livestock Production Science. 76, pp. 171 – 179.

Reynal, S.M. and Broderick, G.A. 2005. Effect of Dietary Level on Rumen-Degraded Protein on Production and Nitrogen Metabolism in Lactating Dairy Cows. In: J. Dairy Sci. 88, pp. 4045 – 4064. Steinfeld, H., Gerber, P., Wassenaar, T., Castel, V., Rosales, M. and de

Haan, C. 2006. Livestock’s long shadow. Environmental issues and

options. http://www.virtualcentre.org/en/library/key_pub/longshad/

A0701E00.pdf

Swensson, C. 2003. Relationship between content of crude protein in rations for dairy cows, N in urine and ammonia release. In: Livestock

Production Science. 84, pp. 125 – 133.

van Duinkerken, G., Andre, G., Smits, M.C.J., Monteny, G.J. and Šebek, L.B.J. 2005. Effect of Rumen-Degradable Protein Balance and Forage Type on Bulk Milk Urea Concentration and Emission of Ammonia from dairy Cow Houses. In: J. Dairy Sci. 88, pp. 1099 – 1112.

Chapter 42

Bartussek, H. 2001. An historical account of the development of the an-imal needs index ANI-35L as part of the attempt to promote and reg-ulate farm animal welfare in Austria. An example of the interaction between animal welfare science and society. In: Acta Agriculturae

Scandinavica A, vol. 51, suppl. 30, pp.34-41

Broom, D.M. 1986. Indicators of poor welfare. In: British Veterinary

Journal. Vol. 142, pp. 524–526.

Broom, D. and Johnson, K.G. 1993. Approaching questions of stress and welfare. In: Stress and animal welfare. Kluwer Academic Publishers, pp 1-7.

Dawkins, M.S. 1983. Battery hens name their price: consumer demand theory and the measurement of ethological ‘needs’. In: Anim. Behav. 31, pp. 1195–1205

European Commission 2007. Animal welfare factsheet. Director-General for Health and Consumer Protection.

Hursh, S.R. 1984. Behavioural Economics. In: J. Exp. Anal. Behav. 42, pp. 435–452.

Wiepkema, P.R., Van Hellemond, K.K., Roessingh, P. and Romberg, H. 1987. Behaviour and abomasal damage in individual veal calves. In:

Applied Animal Behaviour Science. Vol. 18, p.257.

Further reading

Appleby and Hughes 1997. Animal welfare.

Benson and Rollin 2004. The well-being of farm animals

Broom and Fraser 2003. Domestic animal welfare and behaviour. 4th

edition

Martin and Bateson 1993. Measuring behaviour.

The following journals:

Animal Welfare, Applied Animal Behaviour Science, Animal.

Alcaro, A., Huber, R. and Panksepp, J. 2007. Behavioral functions of the mesolimbic dopaminergic system: An affective neuroethologi-cal perspective. In: Brain Research Reviews, 56: 283-321. Burghardt, G.M. 2005. The genesis of animal play. Cambridge, MA:

MIT Press.

Nelson, E.E. and Panksepp, J. 1998 Brain substrates of infant-mother attachment: Contributions of opioids, oxytocin, and norepinephrine. In: Neuroscience & Biobehavioral Reviews, 22: 437-452.

Numan, M. and Insel, T.R. 2003. The neurobiology of parental behav-ior. New York: Springer-Verlag.

Panksepp, J. 1990. The psychoneurology of fear: Evolutionary perspec-tives and the role of animal models in understanding human anxi-ety. In: Roth, M., Burrows, G.D. and Noyes, R. (eds) Handbook of

anxiety. pp. 3-58 Amsterdam Elsevier/North-Holland: Biomedical

Press.

Panksepp, J. 1998. Affective neuroscience: The foundations of human

and animal emotions. New York: Oxford University Press.

Panksepp, J. 2007. Can PLAY diminish ADHD and facilitate the con-struction of the social brain. In: Journal of the Canadian Academy

of Child and Adolescent Psychiatry, 10: 57-66.

Panksepp, J., Fuchs, T., and Iacabucci, P. 2010. The basic neuroscience of emotional experiences in mammals: The Case of subcortical FEAR circuitry and implications for clinical anxiety. In: Applied

Animal Behaviour Science, in press.

Panksepp, J. and Moskal, J. 2008. Dopamine and SEEKING: Subcortical ‘reward’ systems and appetitive urges. In: Elliot, A. (ed.) Handbook

of approach and avoidance motivation, pp. 67-87 Mahwah, NJ.:

Lawrence Erlbaum Associates,

Panksepp, J. and Zellner, M. 2004. Towards a neurobiologically based unified theory of aggression. In: Revue Internationale de

Psychologie Sociale/International Review of Social Psychology. 17,

pp. 37-61.

Pellis and Pellis 2009. The playful brain: Venturing to the limits of

neu-roscience. Oxford, UK: Oneworld Pubs.

Pfaff, D.W. 1999. Drive: Neurobiological and molecular mechanisms

of sexual behavior. Cambridge, MA: MIT Press.

Siegel, A. 2005. The neurobiology of aggression and rage. Boca Raton, FL: CRC Press.

Watt, D.F. and Panksepp, J. 2009. Depression: an evolutionarily con-served mechanism to terminate separation-distress? A review of aminergic, peptidergic, and neural network perspectives. In:

Neuropsychoanalysis, 11, pp. 5-104.

Further Reading

McMillan, F. (ed.). 2003. Mental Health and Well-being in Animals. Ames, Iowa: Iowa State University Press.

Panksepp, J. 2003. At the interface of affective, behavioral and cogni-tive neurosciences: Decoding the emotional feelings of the brain. In:

Brain and Cognition, 52, 4-14.

Panksepp, J. 2005. Affective consciousness: Core emotional feelings in animals and humans. In: Consciousness & Cognition, 14, 30-80. Panksepp, J. 2006. Emotional endophenotypes in evolutionary

psy-chiatry. In: Progress in Neuro-Psychopharmacology & Biological