1. We argue that:
• in their application to non-human animals, 'welfare' and 'well-being' are interchangeable words; and that
• good welfare/well-being is the state of being manifest in an animal when its nutritional, environmental, health, behavioural and mental needs are met.
2. These latter are essentially the 'five freedoms' formulated by the Farm Animal Welfare Council of the United Kingdom.
3. Using the five freedoms as a basis, we have developed a system for assessing the impact of a proposed animal experiment or usage. The freedoms are now transformed into 'domains of potential compromise' and are redefined better to emphasise the extent of welfare compromise rather than the ideal of absence of compromise. Domain 1 is Thirst/hunger/malnutrition, 2 is Environmental challenge, 3 is Disease/injury/functional impairment, 4 is Behavioural/interactive restriction, and domain 5 is Anxiety/fear/pain/distress. A proposal would be examined systematically in all domains, and the degree of compromise in each rated on a 5-step non-numerical scale - O, A, B, C, X. Anxiety/fear/pain/distress arising from compromise in domains 1-4 would be cumulated into domain 5. The overall rating would commonly be that given to domain 5, but if this were low or unknown, it would be given to the highest rating in the other domains.
4. The proposer would be required to present to the institutional Animal Ethics Committee his/her assessment of the impact of a proposed experiment on the animals involved, together with an appropriate justification for the work and a cost-benefit analysis.
5. The extent of the justification required for a proposal would be directly related to the severity of compromise expected, being low for grade O and very high for grade X.
6. The cost-benefit decision would be based on the balance between the expected severity of welfare compromise and the expected benefits set out in the justification.
7. The major advantage of this system for assessing the impact on welfare is that it encourages systematic consideration of all sources of possible compromise. Such wider consideration would allow more accurate assessment of the severity of impact and thereby would improve the validity and efficiency of cost-benefit analyses.
8. The philosophical background to our approach is outlined, graded examples of welfare compromise are given and ethical and practical implications of using the system are discussed.
9. We also set out what we consider to be the ethical and practical responsibilities of the researcher to the animals, and to his/her assistants. The conscientiousness and comprehensiveness of the assessments of welfare compromise and the actions taken to minimise it are measures of the researcher's acceptance of ethical responsibility for all features of each experiment which affect the animals adversely.
Mellor, D. J., & Reid, C. S. W. (1994). Concepts of animal well-being and predicting the impact of procedures on experimental animals. Improving the well-being of animals in the research environment, 3-18.