Why is animal research necessary?

Research using animals is essential for understanding the biology that underpins health and disease in both humans and other animals. Without such research, we would have few of the modern medicines, antibiotics, vaccines and surgical techniques that we take for granted in both human and veterinary medicine.

Animals are used to investigate fundamental biology, to model disease and to test potential new treatments before they are tested in humans. Animal research is only undertaken where there is no alternative.

What types of animal do you use at Cambridge?

The majority of the animals we use are mice and zebrafish – they make up 97% of all procedures at Cambridge. Where these species are not suitable, we use a small number of other animals, such as xenopus frogs, rats and sheep, as well as non-human primates, namely marmosets. 

What types of animal research do you carry out?

Some of the work carried out is fundamental research, aimed at understanding how humans and animals develop and how our immune systems and brains work, for example. This knowledge is essential for underpinning our understanding of health and disease for both medical and veterinary purposes. 

Other work is aimed at tackling specific diseases, for example in helping us understand how Parkinson’s disease affects the brain and motor system and how it might be tackled, or in developing new treatments for autoimmune diseases such as type 1 diabetes and multiple sclerosis.

How many procedures are undertaken on animals at the University of Cambridge each year?

The Home Office collects data on the number of procedures carried out, rather than the number of animals used. The following numbers of procedures based on Home Office Returns were carried out on animals for research at the University of Cambridge:


2021 2020      


2018 2017
Mouse  152,471 102,050 99,351 107,140 134,118
Rat 2,095 2,511 2,137 3,202 3,511
Zebra fish  42,354 71,789 12,456 30,589 18,747
Other fish  1,496 220 0 0 0
Primate  33 41 43 53 39 
Sheep  122 84 110 130 62
Guinea pig  48 0 0 144 76
Other rodents 78 33 14 21 21
Rabbits 2 0 0 0 0
Chicken 0 37 20 30 48
Pigs 59 16 0 0 0
Other birds 0 0 0 0 0
Xenopus 445 438 509 659 1,353 
Total 199,203 177,219 114,640 141,968 157,975 

A ‘procedure’ is regulated and counts towards our Home Office Returns if it is carried out on a protected animal (such as those species listed above) for a scientific or educational purpose and may cause that animal a level of pain, suffering, distress or lasting harm equivalent to, or higher than, that caused by inserting a hypodermic needle according to good veterinary practice.

Procedures that are regulated include modifying the genes of a protected animal if this is done for and experimental or other scientific purpose or applied for an educational purpose; for example, breeding mice with genetic alteration is a regulated procedure if the intention is to keep the animals produced beyond two-thirds of the way through their gestation period. This helps explain why the numbers of mice and zebrafish are so high: each offspring born is counted as a procedure.

How severe were these procedures?

Researchers must assign a severity classification to the procedures they plan to undertake before authority to do their work is authorised by the Home Office (prospective severity). Once their licence has been granted they must also record the actual level of suffering (actual severity), i.e. severity experienced by each animal during the course of a procedure.

The prospective severity classification of a procedure is an assessment of the degree of pain, suffering, distress or lasting harm that is expected to be experienced by an individual animal during the course of the procedure. The prospective severity classifications are defined by the EU Directive as Non-recovery, Mild, Moderate and Severe, examples of which are provided below. 

During each procedure the researcher must monitor their animals and keep a record of the level of suffering (severity) each animal experiences. After the completion of each procedure, the researcher must record the highest level of suffering and any cumulative suffering experience by an animal during the procedure. This is the actual severity for each animal. Actual severity classifications are Non-Recovery, Sub-threshold, Mild, Moderate and Severe. The table below is a record by species of the numbers of animals assigned to each actual severity classification. 


Procedures, which are performed entirely under general anaesthesia from which the animal shall not recover consciousness.


Procedures on animals as a result of which the animals are likely to experience short term mild pain, suffering or distress, as well as procedures with no significant impairment of the wellbeing or general condition of the animals.

Mild procedures include:

  • anaesthesia
  • non-invasive imaging, like and MRI scan
  • short-term social isolation
  • taking a blood sample
  • superficial non-surgical procedures e.g. ear biopsies in mice and non-surgical implantation of recording devices and minipumps


Procedures on animals as a result of which the animals are likely to experience short term moderate pain, suffering or distress, or long-lasting mild pain, suffering or distress as well as procedures that are likely to cause moderate impairment of the wellbeing or general condition of the animals.

Moderate procedures include:

  • invasive surgery under general anaesthetic e.g. surgical implantation of a catheter into a blood vessel for long term drug delivery
  • causing cancer in an animal where the tumour growth impairs normal behaviour
  • feeding a modified diet which is deficient in an essential nutrient such that it affects the health of the animal
  • exposing the animal to something that they would normally run away from, without enabling them to run away
  • the breeding of genetically altered animals where the animals health is affected, e.g. genetic models of diabetes.


Procedures on animals as a result of which the animals are likely to experience severe pain, suffering or distress, or long-lasting moderate pain, suffering or distress as well as procedures, that are likely to cause severe impairment of the wellbeing or general condition of the animals.

  • any test where death is the end-point or where deaths are expected and it is not easy to determine when an animal is likely to die, e.g. models of aortic aneurysm
  • testing a device that could cause pain/death if it were to fail, e.g. testing devices designed to support patients at risk of heart disease
  • inescapable electric shock treatments, e.g. to induce a model of learned helplessness
  • breeding animals with genetic disorders that are expected to experience severe and persistent impairment of general condition, for example Huntington’s disease, and muscular dystrophy

Actual severity:
The above definitions and examples also provide a good insight into what animals could have experienced when undergoing procedures and so reflect how actual severity is determined.  The only difference is that in the UK the Home Office introduced a further actual severity classification known as sub-threshold.  This classification therefore appears when UK annual returns of procedures are published.


This is for procedures which were originally assigned an above-threshold pain or suffering classification, but when the work was undertaken the actual level of suffering was below that which would exceed the threshold at which procedures are licenced under the Act.

Sub-threshold procedures include:

  • breeding of genetically altered animals under project licence authority but without a harmful phenotype
  • dosing with a compound in feed where the animals ate normally and suffered no consequences of being dosed

Finally, any animals that are undergoing experimental procedures that are found dead and death could have been procedure related will be automatically classified as Severe unless the researcher can proved that the death was not procedure related.

Example procedures courtesy of Understanding Animal Research

Species and severity classifications No.of procedures (2021) No. of proc. (2020)

No. of proc. (2018)

No. of proc. (2018) No. of proc. (2017)
Mice 152,471 102,050 99,351 107,140 134,118
Non-recovery 939 1,345 1,801 1,393 2,004
Mild 56,066 36,503 32,517 38,614 42,857
Moderate 24,678 11,642 13,613 15,677 18,179
Severe 2,115 1,362 1,735 1,675 2,451
Sub-threshold 68,673 51,198 49,685 49,781 68,627
Rats  2,095 2,511 2,137 3,202 3,511
Non-recovery 50 68 137 164 331
Mild 440 618 552 859 623
Moderate 1,271 1,357 1,302 1,868 2,172
Severe 26 9 34 64 43
Sub-threshold 308 459 112 247 342
Rabbits 2 0 0 0 0
Mild 2 0 0 0 0
Zebra fish 42,354 71,789 12,456 30,589 18,747
Mild 30,163 56,744 8,550 17,462 10,697
Moderate 10 243 93 2 144
Severe 209 23 26 2 12
Sub-threshold 11,972 14,779 3,787 13,123 7,894​
Other fish 1,496 220 0 0 0
Mild 676 174 0 0 0
Moderate 210 46 0 0 0
Severe 8 0 0 0 0
Sub-threshold 600 0 0 0 0
Non-recovery 2 0 0 0 0
Primates 33 41 43 53 39
Mild 9 11 19 16 12
Moderate 21 30 23 37 27
Severe 3 0 1 0 0
Sheep 122 84 110 130 62
Non-recovery 0 8 4 5 0
Mild 38 17 11 4 4
Moderate 65 55 94 119 58
Severe 3 4 1 0 0
Sub-threshold 16 0 0 2 0
Pigs 59 16 0 0 0
Non-recovery 50 13 0 0 0
Mild 6 0 0 0 0
Moderate 3 3 0 0 0
Guinea-Pigs 48 0 0 144 76
Non-recovery 0 0 0 0 6
Mild 48 0 0 0 0
Moderate 0 0 0 144 69
Severe 0 0 0 0 1
Other Rodents 78 33 14 21 21
Mild 78 33 14 21 21
Domestic Fowl 0 37 20 30 48
Mild 0 37 20 30 48
Xenopus 445 438 509 659 1,353
Mild 445 438 509 658 1,350
Moderate 0 0 0 1 1
Severe 0 0 0 0 2
Total 199,203 177,219 114,640 141,968 157,975

Additional Data

In 2017 every UK establishment provided data under the EU Directive (2010/63/EU), which required Member States to collect additional data that included the number of animals bred or produced for scientific procedures that were killed without being used in procedures. The reasons for this include:

  • The animals were ‘wild type’ breeding animals;
  • They were used to produce tissue only and were not subjected to regulated procedures;
  • The animals were the wrong gender required for the studies;
  • They were a surplus resulting from breeding animals to ensure adequate supply for scientific purposes.

Only animals not reported in the annual return of procedures were counted.

Since 2017, the University of Cambridge has continued to collect these additional data annually and is publishing them in the interests of openness and transparency.


How do you ensure high standards of animal welfare?

We believe that good science and good animal welfare go hand in hand.  The UK has the most rigorous animal welfare regulations in the world, and we consider adherence to these regulations as a minimum and will continue to aim for the highest possible standards of animal care.

We strongly agree with, and rigidly follow, the guiding principles emphasised by the Home Office on the need to refine protocols, keep the numbers of animals used to a minimum and replace the use of animals with other methods where possible. To this end, we encourage all staff involved in animal research and husbandry to continuously develop and improve on existing welfare standards, offering incentives to those that contribute, and rewarding those that are recognised by the laboratory animal welfare organisations.

Are you looking for alternatives to animal use?

We are committed to refining, reducing and replacing the use of animals in research - known as the 3Rs. Animals are only used where no alternatives are viable. Cambridge scientists are also leading research looking at finding viable alternatives. For example, in 2014, Dr Meritxell Huch from the Gurdon Institute won the UK’s international prize for the scientific and technological advance with the most potential to achieve the 3Rs for work to grow “mini-livers” from adult mouse stem cells.

Do you test cosmetics and household products on animals?

No. It is not permitted anywhere within the UK or the European Union to test cosmetics or household products on animals.