UK Home Office releases 2015 statistics on animals used in scientific procedures

The UK Home Office has today released its annual statistics of scientific procedures on living animal in Great Britain during 2015.

Dogs in home pen

Dogs in home pen


Data collected on animal use changed from 2014 onwards, with European Union Directive 2010/63/EU setting out a common format for all member states. Key changes include reporting of the actual severity of procedures experienced by animals, differences in the categorisation of research, and the source of animals is now reported as place of birth. The Home Office note that data collected in 2015 are more likely to be robust than those collected in 2014.

For all species, 4,142,631 were conducted on 4,069,349 in 2015, which shows a modest increase from 2014 (3,867,439 procedures in 3,800,000 animals).

A total of 3,405 dogs were used in 4,643 procedures in 2015, which shows an increase over the 2,742 dogs reported as being used in 4,107 procedures in 2014. However, the apparent drop in dog use reported for 2014 may have been a result of the new data collection format, with numbers for 2013 being more comparable with 2015 (3,554 dogs). See our graph below to see how Great Britain’s dog use compares to other countries.


Numbers of dogs used in scientific research globally, showing the most recent data reported for each country. Note that this does not include countries thought to use large numbers of dogs, but which do not report animal numbers.

Numbers of dogs used in scientific research globally, showing the most recent data reported for each country. Note that this does not include countries thought to use large numbers of dogs, but which do not report animal numbers.

Numbers of dogs used in scientific research globally, showing the most recent data reported for each country. Note that this does not include countries thought to use large numbers of dogs, but which do not report animal numbers.


Dogs used in scientific research within the UK must be obtained from a designated supplier. Data for 2015 show that most dogs were bred at designated establishments within the UK or other member states. However, 1,012 beagle dogs were obtained from breeders in non-EU countries. This represents a considerable increased from the 640 dogs reported as being bred in non-EU countries in 2014.

One of the most significant changes which has occurred in data collection since the implementation is the reporting of actual severity caused by regulated procedures. Regulated procedures are defined by the Animals (Scientific Procedures) Act (1986, updated 2012) as being those with the potential to cause pain, suffering, distress or lasting harm. The retrospective severity  assessment is conducted following the completion of procedures and assessed the actual severity experienced by each animal. Severity is reported in the following categories: sub-threshold, non-recovery, mild, moderate and severe.

The data for 2015 show that the majority of dogs experienced mild severity procedures. The break down in as follows: sub-threshold (14), non-recovery (49), mild (3,097), moderate (1,480) and severe (3). Only three dogs experienced severe procedures, a decrease from 13 in 2014.

The predominant use of dogs in scientific research in the UK is to meet regulatory requirements for new chemical substances, such as medicines and pesticides. The data for 2015 demonstrate that this is still the case, with 3,085 of the total 3,405 procedures (90.6%) being conducted for regulatory purposes. Of these procedures, 2,454 were for repeated dose toxicity.

The full report can be found on the Home Office’s Animal Research and Testing portal.

More information on the use of dogs in scientific research can be found on our website.

3195 Total Views 2 Views Today


  1. Ed Hayes

    Firstly, I’d like to welcome the intention of the wider work this website is promoting to refine the experience of dogs in animal experiments, refinement is an important component of the 3Rs.

    I wonder if you would like to expand on your comment that the change in statistical reporting in 2014 could account for a 20% reduction (712 dogs) from 2013 to 2014, and then a 24% increase (663 dogs) from 2014 to 2015, of the number of dogs being used in scientific procedures. Such a sizeable difference on a ‘specially protected species’ would appear to be more than an artefact of statistical reporting.

    We’re talking about whether a dog was used in a scientific procedure or not, not determining whether a given procedure counts as one or two (or even more) in the new reporting system. As I understand the overwhelming majority of dogs used in regulatory procedures are killed at the end of procedures to allow for examination of organs etc, given that 72% (480 dogs) of the increase in procedures from 2014 to 2015 involved this form of experiment I would be concerned if the scientists involved were unable to decide whether this was a recordable procedure or not.

    Furthermore, the overall increase in total number of animals between 2014 to 2015 was approx 6-7%, it seems unlikely that there would be such a big change in dogs, unless there is something which makes it inherently more difficult to determine whether they’ve been used in a scientific procedure. The 2014 figure was also very similar to the 2011 figures (2,865 dogs), so surely it’s more of a case of natural variation in the number of experiments year to year, rather than an artefact of statistical reporting?

    Your graph above comparing countries while being slightly interesting is far from illuminating. What would be much more instructive would be a year by year comparison of dog use in the UK in absolute numbers, and also by nature of use, such as toxicology testing, or basic research, and in future years severity. Though I’d readily admit that even this would have limited value in reflecting whether there has been advances in replacement and reduction of dogs in research.

    Furthermore with respect to your graph above, yes the US is using by far the largest numbers of dogs in animal testing across the world, but absolute numbers are only part of the equation. What are the dogs being used for? What is the severity level? Are they conducting much more science in an area which dogs are currently perceived necessary by that section of the scientific community i.e. proportionally more regulatory testing than other nations?

    I’d also speculate that 60,000 dogs being used in the US in mild procedures for development of new medicines, would for some be more palatable than 18,000 dogs being tested in the EU in severe procedures, for the development of new household toilet cleaning products. I’m not suggesting this is the case, just illustrating that the graph presented might demonstrate UK use of dogs is lower than other countries but it’s off little additional value.

    • Dear Ed, thank you for your interest in the website and your comment on the post. I have copied the following from the document released by the Home Office today:

      ” [T]hroughout this release, 2015 data are compared with 2013 data, as neither year of data are subject to the same data quality issues as the 2014 data. However, comparisons between 2015 and 2013 should still be exercised with a degree of caution due to the methodological change in 2014.”

      A number of changes to the way data were collected from 2014 onwards influenced how data were reported for that year, in particular changing from reporting procedures started in a given year to procedures completed, which resulted in discrepancies with the preceding year as some procedures begin and end in different years. This does not mean that data were not collected for any animals, only that the year to which they were attributed has changed. Since the transposition of the EU Directive, it has been a requirement for dogs to have an individual record which records their personal histories, so it is not the case that data were not recorded for any dog, which is a legislative requirement. If you would like more details on how the change in data collection has affected the reported animal use for 2014, I would recommend that you contact the Home Office directly using the information on this page:

      The graph is not intended to compare the quality of research conducted by individual countries, only to reflect the data as they are reported. Some of your questions may be answered by reading our page on statistics and legislation: This expands further on the statistics collected by other countries, along with links to the relevant legislation. These show that for example, 75% of dogs in the USA are used in regulatory research, compared to 90% in the 2015 data for Great Britain.

      However, some of your other questions cannot be answered easily, as the frequency and detail of data collection vary considerably between countries, making direct comparisons difficult. Dogs are given special protection under EU legislation, and their use is therefore subject to additional scrutiny in EU member states. The differences in legislation between countries means that it is important to understand in which countries dogs are used. You will note that only three dogs out of 3,405 experienced severe procedures and this is likely to be similar in other EU member states as they are operating under the same legislation. If you are interested in differences in legislation, I recommend that you explore the links in our page on statistics and legislation:

  2. Ben

    Using dogs for these studies sometimes feels when seen according to human perspective since they are going to give out the exact results of how a drug or chemical is going to affect a human being. Being a dog lover however i feel bad for the high number of canines being subjected to these trials and tests especially in the US and UK. Still we need out research to go on maybe at the cost of these pooches. Your data was really an eye opener for me.

Leave Comment

Your email address will not be published. Required fields are marked *