Animal research can only be carried out in the EU when there is no suitable non-animal alternative. All project proposals that require the use of animals also have to be approved by the ethics committee at each institution that conducts animal research. The research will not be allowed if the potential benefits of research do not outweigh the suffering that might be experienced by the animals used. 

Animal research in the European Union (EU) is strictly regulated under Directive 2010/63/EU on the protection of animals used for scientific purposes, which is also closely followed in non-EU countries such as Switzerland, the UK and Norway. The Directive’s aim is to protect animals used in research by regulating both the use and care of these animals. No procedure is allowed to be carried out without justification and formal approval by national authorities. Besides the Directive, the transport of animals is regulated by a separate regulation (Council Regulation (EC) No 1/2005, currently under revision); the testing of chemicals and other substances is regulated via dedicated laws and guidelines; and there is a ban on animal testing for cosmetics in place in the EU. 

Similar standards to ensure the proper regulation and welfare of research animals exist in many other parts of the world, such as the USA, Australia and South Korea. 

Directive and other relevant regulations
Directive 

EU Member States each have their national regulations which protect animals in research. EU Directive 2010/63 harmonises animal research legislation throughout the EU, to ensure high standards of animal welfare and scientific research. It was implemented into national law in each EU Member State in 2013. 

This Directive dictates which animals may be used in research, under which conditions and what measures need to be taken when animals are used. 

Animals can only be used if all the following conditions are fulfilled: 

  • There is a convincing scientific justification. 
  • The expected benefits of the research outweigh the potential risks in terms of animal suffering. 
  • The scientific objectives cannot be achieved using non-animal alternative methods. 

Projects that do not fulfil all these requirements are not authorised. 

The ultimate aim of the Directive is to replace all animal research with non-animal methods of research, such as organoids, organs-on-chips or computer simulations (see also the EARA feature, ‘When will animals be replaced in biomedical research?’). 

Central to this aim and to the Directive as a whole is the requirement to implement the principles of the 3Rs – replacement, reduction and refinement: 

  • Replacement – First, attempt to replace animals with non-animal methods wherever possible. Non-animal methods have priority over animal methods wherever scientifically feasible. 
  • Reduction – If this is not scientifically feasible, reduce the number of animals used to a minimum while still obtaining scientifically valid and robust results. 
  • Refinement– Lastly, refine practices to minimise any potential pain, suffering, distress or lasting harm to the animal by optimising procedures, care and housing. 

Read more about the 3Rs principle here. 

Since its transposition into national law, the Directive has been reviewed to address any scientific, ethical or regulatory developments in laboratory animal science, especially the use of non-human primates and non-animal alternatives. The most recent review took place in 2019. 

Directive 2010/63/EU
Safety assessments for chemicals, medicines and pesticides

Systems are in place in the EU to ensure the safety of different substances and products for human health, animal health and the environment. Chemicals, medicines and pesticides first need to be assessed for safety before they are allowed to enter the market. For each type of product, there are regulations that define what information is required, while agencies decide for each individual product how those information requirements can be fulfilled, including whether animal tests are needed. 

The REACH (Registration, Evaluation, Authorisation and Restriction of Chemicals) Regulation defines the requirements for chemicals, with the European Chemicals Agency (ECHA) overseeing its implementation 

The development and safety assessment of medicines and vaccines in the EU are governed by legislation including Directive 2001/83/EC. Preclinical studies must evaluate safety and effectiveness before medicines or vaccines can proceed to clinical trials in humans. In many cases, animal studies are a part of these preclinical studies. It is the European Medicines Agency (EMA) that oversees these assessments. 

For pesticides and biocides, Regulation 1107/2009 and 528/2012 define what information must be collected to ensure safety for human health, animals and the environment. Animal studies may be required for such studies and the European Food Safety Authority (EFSA) oversees this process. 

EU ban on animal testing for cosmetics

An EU-wide ban on animal testing for cosmetics has been complete since 2013. This ban prohibits both the practice of animal testing for cosmetics and the import of cosmetics that were tested on animals elsewhere. This is possible because science has progressed far enough to make testing for simpler biological risks such as skin and eye irritation possible with animal-free methods. 

There are only limited exceptions related to workers’ safety. Factory workers may be exposed repeatedly and for a long time to chemical ingredients of cosmetics at high concentrations, not only through skin or eye contact, but also through, for example, inhalation. Therefore, animal testing may still be required under the EU’s chemicals regulation REACH to protect their health, when their exposure cannot be assessed yet using animal-free models. 

Watch an explanatory video by Monique Sundin, EARA Policy Officer.

Regulation on the transport of animals 

When research animals are transported in the EU, Regulation (EC) No1/2005 on the protection of animals during transport applies, just like for other animals. However, because these animals are for scientific purposes, Directive 2010/63/EU also sets additional rules. This means that the transport of research animals is more strictly regulated than the transport of other animals. 

The regulation on the transport of animals is currently under revision. 

Compliance with the Directive is ensured by a layered system and animal research needs to be approved on different organisational levels: 

  • Institutional license – Organisations that breed, supply or use animals for scientific purposes need to be authorised to do so. The local competent authorities grant licenses to institutions if they have appropriate facilities and staff that comply with the welfare and housing standards of the Directive. The wide range of general and species-specific requirements are set out in Annex III of the Directive. 
  • Certified staff – Every individual working with research animals must have the right certification depending on their role: carrying out experimental procedures, designing procedures and projects, taking care of the animals or euthanising animals. 
  • Project licenses – Individual research projects have to be authorised by the local competent authority before they can start. Project applications follow a pre-defined template and include extensive details on procedures and schedules, scientific explanations, ethical considerations and a non-technical summary of the project (see the EARA feature, ‘Non-technical summaries explained’). An important component of this assessment is the ‘harm-benefit analysis’ by which the competent authorities assess if the potential outcomes of a project outweigh the harm done to the animals. Once approved, any changes to protocols that deviate from the granted project license require filing an amendment with the competent authority. 
  • Retrospective assessment – In certain cases, a retrospective assessment is required at the end of a project, specifically for projects involving non-human primates and with procedures classified as ‘severe’. This assessment includes whether the objectives of the project were achieved and the actual severity of the procedures as carried out. 
  • Inspections – Member States are obliged to carry out regular inspections, including inspections of at least one third of all users each year. A portion of these inspections must be unannounced. The inspections are carried out by the competent authorities. 
Who does what?

Official national bodies appointed in each EU Member State that oversee and enforce the regulations under Directive 2010/63/EU. They grant authorisation for animal research projects and carry out inspections at establishments (breeders, suppliers and research organisations).

Each authorised institution is obliged to have an Animal Welfare Body (AWB) installed. This AWB advises and supports staff on all animal welfare matters related to acquisition, care, housing and use of animals and on how to apply the 3Rs principle, they keep track of how research projects affect the animals and they give advice on project applications. They consist of at least one person with practical knowledge of animal care, housing and husbandry, one scientific member and a designate veterinarian who oversees animal health and veterinary care.

Every Member State has a national committee which advises the competent authorities and animal welfare bodies on matters relating to research animals, including their acquisition, breeding, accommodation, care and use. All national committees then share information at the EU level about best practice and the functioning of animal-welfare bodies.

The Directive defines an animal procedure as an action that leads to a level of pain, suffering, distress or lasting harm that is equivalent to or higher than an injection with a needle (as performed by a skilled professional). When such procedures are performed for scientific purposes, they count as a procedure under the Directive, while the identical procedure will not count as an animal experiment if performed in a veterinary clinic.  

For example: 

Procedures on animals for scientific purposes are subject to the Directive and therefore can only be performed with permission and under the conditions stated in the Directive. 

  • Observing animals in the wild from a distance without any interaction is not considered an animal experiment, while doing a small surgery to implant a tracking device in wild animals to study their behaviour is.  
  • Killing an animal (using specified, approved methods) without any preceding procedure solely to make use of its tissues in a study is not considered an animal experiment according to the EU-definition. Some EU Member States have a stricter definition of “animal experiment” that includes killing the animal without any preceding procedures. 
  • Blood sampling of your pet dog at a veterinary clinic as part of diagnosis or treatment is not considered an animal experiment, while blood sampling of a dog for testing that a new dog medication is safe and effective is considered an animal experiment. 
How are procedures classified? 

The level of pain, suffering, distress, or lasting harm that is experienced by an animal during or as the result of a procedure is taken into account by the Directive. Which severity level applies for an animal depends on the predicted cumulative severity across the animal’s lifetime. 

In practice, there are four levels of severity for procedures: non-recovery, mild, moderate and severe. 

How is the distribution of severity of procedures in the EU? Click to see the data.

Procedures during which the animal is under deep anaesthesia during the entire procedure and never regains consciousness.  

  • For example: where researchers need to measure or collect organs while they are still functioning. 
  • Measures taken: the animal has to be under deep anaesthesia, meaning it is fully asleep the whole time, so it does not experience pain or distress. 

Procedures where the animal is likely to experience short-term mild pain, suffering or distress. These are procedures with no significant impairment to the well-being or general condition of the animal. This category also includes animals that are  not observed to have experienced pain, suffering, distress or lasting harm above the minimum threshold. 

  • For example: administering anaesthesia, performing non-invasive imaging (CT- or PET-scan), short-term social isolation (several hours), and taking a blood sample. 
  • Measures taken: procedures are brief, animals are monitored during and shortly after the procedure and returned to their housing and routine promptly, according to what is appropriate for the species and procedure.

Procedures where the animal is likely to experience short-term moderate pain, suffering or distress, or long-lasting mild pain, suffering or distress. These are procedures that are likely to cause moderate impairment to the well-being or general condition of the animals. 

  • For example: implantation of telemetry devices (trackers), catheter implantation, social isolation that lasts several days or weeks, repeated injections or blood sampling over several days or weeks. 
  • Measures taken: the use of effective analgesia and/or anaesthesia (unless anaesthesia would disturb the experiment or measurement and it’s ethically justified) and more extensive monitoring, which often includes defined clinical scoring and daily or more frequent checks, depending on what is appropriate for the species and procedure. Predefined endpoints determine how to act depending on the observations made. 

Procedures where the animal is likely to experience severe pain, suffering or distress, or long-lasting moderate pain, suffering or distress. These are procedures that are likely to cause severe impairment to the well -being or the general condition of the animal. 

  • For example: any test where death is the end-point, or fatalities are expected; testing a device that could cause pain or death if the device were to fail; breeding animals with genetic disorders that are expected to experience severe and persistent impairment of their general condition (for example, Huntington’s disease or muscular dystrophy). 
  • Measures taken: the animal must be anaesthetised, and monitoring has to occur at a high frequency, usually multiple times per day. Mandatory pre-defined endpoints determine when animals have to be euthanised or removed from the procedure before prolonged severe suffering occurs. In some cases, a retrospective assessment is required to confirm the severity level after the study is performed. 

Severity plays a central role in project authorisation, as the authorities assess whether the expected severity is justified by the expected benefits. Severe procedures are highly restricted and require exceptional justification. The predicted severity of procedures also dictates what measures should be taken by the researchers and animal caretakers to minimise pain, suffering and distress. 

Severity is assessed in advance (predicted severity) in the project licence application and during the study (actual severity). Animals must never experience a level of suffering greater than what has been authorised.  

With a higher predicted severity, stronger safeguards are needed to relieve pain and monitor animal welfare. In the project licence application, clear humane endpoints are agreed upon. These endpoints indicate when animals should be removed from a procedure, receive certain treatment or pain relief or be euthanised before unnecessary suffering occurs. 

You can find the predicted and retrospective (actual) severity levels of procedures of approved projects as part of the non-technical summary in the ALURES database. 

There is a vast array of diseases and disorders which stem from the (dys-)function of genes. By altering the genetics of an animal, the various functions of different genes can be identified more easily. Researchers may add in a new gene, silence or remove an existing gene, or edit an existing gene. Genetically altered animals (GAA) can contribute to our understanding of how animals and humans develop and function, as these identified genes, or similar genes, are often also present in humans. They can also help in the development of diagnostics, treatments and preventive strategies. 

How are genetic alterations made? Click to learn about the common methods.
How many animals are genetically-altered in the EU? Click to see the data.
How is the use of GAA in research regulated? 

The use of genetically altered animals also raises important ethical and welfare considerations. Genetic modification may unintentionally cause pain, suffering, or long-term health issues. Animals may only be genetically altered for research purposes under strict conditions set out in Directive 2010/63/EU. 

Breeding genetically altered animals is classified as  a procedure, with its severity level depending on whether the alteration is expected to have an effect on the animal’s wellbeing. Any potential harm to the animal is carefully weighed against the anticipated scientific or medical benefit, and animals are closely monitored throughout their lives. Breeding lines that develop unexpected harmful effects may be refined, restricted, or discontinued.  

Which animals can be genetically altered? 

In the EU, great apes are the only animal species that is banned outright from genetic alteration; however, genetic alternation of other species does require strong scientific justification, particularly for larger animal species. This means that genetic editing is performed almost exclusively on the smaller animals more commonly used in procedures: mice, zebrafish and fruit flies (the latter not being classified as an experimental animal under the Directive). Like any procedure, all projects involving genetically altered animals must comply with the principles of the 3Rs.