Harm Minimisation

The 3Rs: Replacement, Reduction & Refinement

Animal-based scientists are required by law to make sure that they keep any pain, suffering or other harm they cause to the animals they use for research, teaching and testing as low as possible.

The 3Rs Principle is applied at the planning stages before any direct work with animals begins. Its purpose is to help scientists to minimise the invasiveness, unpleasantness or noxiousness of anything they do to animals.

The 3Rs Principle is a practical guide to scientists, and is designed to ensure that:

  • animals which might suffer are only used when necessary (Replacement),
  • that no more and no fewer animals are used than are required to achieve the objectives of the work (Reduction), and
  • that if any noxiousness is caused during the work, it is kept as low as possible (Refinement).


Replacement means that animals should not be used at all if the same research, teaching or testing aim can be achieved in other ways. The word "animal" refers to those higher order animals that are capable of suffering or feeling pain.

The first question scientists must ask themselves at the planning stage of a study is "Do I need to use higher order animals at all?" If the answer is "Yes" then Reduction and Refinement must be applied.

The three important questions of replacement

  • What alternatives can be used instead of higher order animals?
    Non-animal alternatives
    • Computer models
    • Chemical models
    • Charts, diagrams, manikins and physical models
    • Mathematical and statistical models
    • Use of plants.
    Alternatives derived from animals

    Lower order animals

    • Micro-organisms
    • Cells derived from invertebrates and lower order vertebrates
    • Intact invertebrates and lower order vertebrates.

    Higher order animals

    • Tissue culture using cells derived from higher order animals
    • Videos of procedures conducted on animals to avoid repetition.

    Human beings

    • Voluntarily donated human tissues (e.g. the placenta and other tissues)
    • Human volunteers.
  • When is it appropriate scientifically to use alternatives to higher order animals?

    When the alternative to animal use can genuinely reveal new knowledge or demonstrate particular features of the body organ or tissue or the whole body processes of interest.

    To date, replacement alternatives have been used extensively in teaching and during some stages of drug and cosmetic testing, but less extensively during research designed to understand how the body as a whole works.

    Computer, mathematical and other models are helpful for analysing, presenting or accessing knowledge we have already obtained about body processes, but they are less helpful in generating new knowledge.

  • When is it not appropriate scientifcally to use alternatives to higher order animals?
    • When chemical or computer or physical or mathematical models cannot reveal relevant new knowledge or demonstrate the known fact or principle.
    • When microbial or tissue cultures cannot be applied to achieve the desired goals.
    • When the processes to be studied or demonstrated cannot be effectively modelled using non-vertebrate or lower vertebrate animals.
    • When the processes to be studied or demonstrated can only be modelled effectively using the chosen species of higher order animal - e.g. when functions in the particular chosen species (e.g. sheep) closely parallel functions in another animal species (e.g. goats, cattle) or in human beings.
    • When the processes to be studied relate explicitly to the chosen species of higher order animal (e.g. studies of pregnant sheep to reduce death or sickness in newborn lambs).


Reduction means keeping the number of animals used to the minimum necessary to achieve the research, teaching or testing purposes of the work. This avoids using unnecessarily large numbers of animals.

It is equally important to avoid using too few animals, as if not enough animals are used it will not be possible to interpret the results, and the animals used would have been wasted. The second question scientists must therefore ask is "What is the lowest number of animals needed for this work?"

  • How can the number of animals used be reduced to the minimum needed to achieve the aim(s) of the work?

    • If related but different work has been done before, it can be used to assess the number of animals that will need to be studied to produce a definite result.
    • Science-based statisticians can advise on the minimum numbers required.
    • In some situations, the animals can be studied during spontaneous occurrences of the illness or injury or other phenomena of interest, so that no extra animals are required for the purposes of the study.
    • Some studies can be done, painlessly and with no distress, in animals which have been studied for other scientific purposes. That can occur during a short period of unconsciousness while they are under anaesthetic just before they are euthanased using an overdose of that anaesthetic, or immediately after euthanasia.
    • Tissues from animals killed in abattoirs can also be retrieved immediately after death and used for research, teaching or testing, thereby avoiding the need to use extra animals.


Refinement refers to keeping any pain, suffering or other harm which may be caused as low as possible for each and every animal used in the work.

This means that every aspect of the work must be reviewed carefully, and great care taken to minimise any noxious effects on the animals. Thus, the third question scientists must ask is "How can I minimise the noxiousness of every aspect of this work?"

It is worth noting that many studies cause very low or no pain, suffering or other harm to the animals involved, while others do indeed have noxious effects.

  • How can the noxiousness of each study be minimised?

    There are many different ways of refining procedures. For example, studies can involve:

    • Non-invasive behavioural observations of conscious animals; non-invasive methods such as ultrasound scanning or X-ray examinations or other sophisticated imaging techniques in conscious animals;
    • Non-invasive methods such as external collection of urine and faeces;
    • Limiting the invasiveness by taking blood samples using a needle the minimum number of times required to achieve the desired goals;
    • A hormone implant is placed under the skin using local anaesthesia instead of, for example, surgically modifying the animal to change the way it produces the hormone;
    • The animals are kept unconscious with a general anaesthetic throughout the study at the end of which they are killed with an overdose of that anaesthetic – unconscious and dead animals cannot suffer or experience pain;
    • Anaesthetics, pain-killers (analgesics), sedatives and/or tranquillisers are used to relieve anxiety, fear, pain or distress in conscious animals;
    • Any surgery on the animals is done by expert surgeons and great care is taken to keep damage to body tissues to an absolute minimum, thereby reducing any pain experienced when the anaesthetic used during the surgery wears off;
    • When the animals are likely to experience pain, suffering or other harm (e.g., in studies of diseases), their condition is assessed very regularly, the earliest signs that the study objectives have been met are identified, and the animals are withdrawn from the study or euthanased at that earliest time.

    Also animals are handled gently throughout and, where possible, are given extra attention, including food treats and stroking, to enhance their well-being.