Category Archives: Statistics

Animal Research Statistics in Austria, Hungary and Slovenia

Speaking of Research try to keep on top of the latest statistics coming from governments around the world. This post will look at three countries which have recently published their 2016 statistics. All three are regulated by EU Directive 2010/63 which requires countries to produce national statistics on their animal use.


Austria reported 236,459 procedures on animals in 2016, a 4% rise on the previous year.


Procedures on animals in Austria for research in 2016. Click to Enlarge

The rise in animal experiments appears to be mainly due to large increases in the number of birds (mainly chickens) used (up 116%), other mammals (up 89%; mainly pigs) and rats (up 23%).


The vast majority of research was conducted on mice (81.2%), with rabbits (6.2%) and fish (4.4%) the next most common species. It is interesting that Austria, rabbits are the second most common species, a fact not seen anywhere else in Europe, though neighbouring Germany also has a relatively high number (4% of total). No primates were used in Austria in 2016 (or the previous two years) and dogs and cats accounted for less than 0.08% of all animals used despite the rises in the number of procedures for the former.

Trend over time in animal experiments in Austria. Click to Enlarge.

From historical statistics, we can see that while there has been an overall decline of almost 50% since 1990, the numbers have been edging upwards since their nadir in 1999. These numbers tend to reflect changing science funding environments within the country.

This year was the third year where there were retrospective assessment and reporting of severity (i.e. reporting how much an animal actually suffered rather than how much it was predicted to suffer prior to the study). Reassuringly the proportions in each severity banding were similar to previous years, suggesting the system has been well understood. The report showed that 2.5% as non-recovery (4% in 2015) , 63% of procedures were classed as mild (60% in 2015), 27.5% as moderate (24% in 2015), and 7.3% as severe (down from 12% in 2015).

Finally the statistics note that 41.4% of procedures involved genetically altered animals. These were mainly mice and zebrafish, but also included rats and other fish.

See previous reports:


Hungary reported 170,075 procedures on animals in 2016, a 7.9% fall on the previous year.


Procedures on animals in Hungary for research in 2016. Click to Enlarge

The fall in animal experiments appears to be mainly due to falls in the number of procedures on birds (down 5%), rats (down 9%), reptiles and amphibians (down 25%) and fish (down 72%).


Together, mice, rats and birds account for 84% of all procedures in Hungary, this figure rises to over 95% when fish, reptiles, amphibians and guinea pigs are included. The use of primates went from 3 in 2015, down to 0 in 2016 (presumably after a specific study ended). Dogs and cats accounted for less than 0.4% of all animals used.


Trend over time in animal experiments in Hungary. Click to Enlarge.

Using the trend graph we can see how – bar an anomalous year in 2013 – there has been a steady downward trend in animal procedures in Hungary from over 300,000 in 2007, to less than 175,000 in 2016. Perhaps coincidentally the 2013 high point coincides with the implementation of the EU Directive (and its rules around counting procedures), meaning it is possible that this figure is a statistical error caused by incorrect data from the first year under a new counting regime.

This year was the third year where there were retrospective assessment and reporting of severity (i.e. reporting how much an animal actually suffered rather than how much it was predicted to suffer prior to the study). The report showed that 19% as non-recovery (8% in 2015) , 49% of procedures were classed as mild (71% in 2015), 24% as moderate (15% in 2015), and 8.2% as severe (up from 6% in 2015).

Lastly, of note, only 4.7% of animal procedures were on genetically altered animal (up from 2.8% in 2015) – a much lower proportion than, say, the UK, where almost half of procedures were the breeding of a genetically altered animal.

See previous reports:


Slovenia reported 6,819 procedures on animals in 2016, a 25% fall on the previous year.


Procedures on animals in Sloevnia for research in 2016. Click to Enlarge

The main changes was a 24% fall in the use of mice. This is particularly noteworthy given the 22% fall between 2014 and 2015. Neither fish nor horses were used in 2016 (there were 57 and 2 procedures respectively). The only rise was in pigs which went from 8 procedures to 32 in 2016.

As is the case in many smaller countries (but not all), most of the research was animal testing for regulatory purposes (66%), followed by translational and applied research (26%) and basic research (6%).

See previous reports:

That’s all for today but we will endeavor to provide the latest statistics as they are published by national governments. All our statistics can be found on the Statistics Overview page.

Animal experiments in the UK decline by 5% in 2016

The UK Home Office has published the 2016 annual statistics showing the number of animal procedures carried out in Great Britain under the Animals (Scientific Procedures) Act, 1986; this covers all vertebrate species (and Cephalopods). In 2016 there were 3.94 million procedures carried out, down 5% from 2015 (4.14 million).

While we often describe these statistics as being for the UK, they do not include Northern Ireland (who carried out 22,508 procedures in 2015), and so are technically the figures for Great Britain (England, Scotland and Wales).


Procedures on animals in Great Britain for research in 2016. Click to Enlarge

Overall, 96.6% of animals used in scientific studies were mice, rats, fish or birds. Dogs, cats and primates (which are offered special protections under UK law) together accounted for 0.22% of the total (similar to in previous years). The statistics also reveal that almost half of all experiments (48.6%) were the breeding of GM animals which were not used in further experiments – this is almost identical to 2014 and 2015. Overall, 64.9% (almost two thirds) of all experiments involved genetically modified animals, this was down from 67% in 2015.

Different colours represent changes to the counting method in 1987 and 2014.

Trend over time in animal experiments in the UK. Click to Enlarge.

Using the trend graph we can see how the number of animals used in research appears to be flattening out after almost two decades of rising numbers. It should be noted that the Home Office reported in 2015 that it believes there were statistical errors in the 2014 report as a result of new counting methodologies, but that this has been ironed out for 2015 and 2016.

The last five years appears to show animal experiments remaining relatively constant around 4 million. While this is higher than in the 1990s, it remains much lower than the 5.5+ million animals used in the mid 1960s.

Procedures on non-human primates fell slightly from 3,612 procedures in 2015, to 3,569 (down 1%) in 2016. The number of procedures on cats fell by 19 to 190 procedures and on dogs rose to 4,932.

A ban on cosmetic testing on animals (1998) and of using great apes (gorillas, orang-utans and chimpanzees) in research (1986) meant both had zero procedures in 2015. It should be noted that some research may continue on great apes in zoos, however such research can be observation-based only as “procedures” on great apes are illegal under ASPA.

For the third time the UK statistics include retrospective reporting of suffering. Rather than just submitting licence proposals to the Home Office that include estimated levels of suffering, the researchers now have to report on what was actually seen (using a variety of measures). Unfortunately, the statistics put these in two separate tables (Table 3 and 8). So we have combined them to get severity for all procedures in 2016. We can see most experiments are sub threshold (38%; less than the introduction of a hypodermic needle), non-recovery (4%; the animal never awakes from anaesthesia) or mild (38%), with remainder as moderate (16%), or severe (3.9%) Overall the proportion of moderate and severe rose from 18.2% in 2015 to 19.9% in 2015 (though numbers in severe category fell). Also of note in the severity data is the rising proportion of sub-threshold experiments in the creation/breeding of genetically modified animals from 45% in 2014, to 55% in 2015, to 65% in 2016, suggesting improving methods.


Severity of animal research in the UK in 2016

Other things to note in the UK statistics:

  • 48.6% of procedures were for the creation and breeding of genetically altered animals (not used in other experiments), 28.5% were for basic research, 13.5% was for regulatory purposes and 8.6% was translational/applied research [Table 1]
  • Over the experimental procedures, 64% of the “severe” experimental procedures were conducted on mice for regulatory purposes. This is often because death is an endpoint in such procedures [Table 3.1]
  • Over 97% of the animals were born in the UK [Table 2.1]
  • 49.2% of procedures were conducted in universities and medical schools, 25.3% were in commercial organisations (e.g. pharmaceuticals), 13.2% were done at non-profit making organisations (e.g. medical research charities), and 12.2% were done at other public bodies. [Table 11]

Speaking of Research congratulate the UK government on continuing to produce the most comprehensive statistics on animal experiments worldwide. It is also important to note that these statistics are released as a press conference each year where representatives from the scientific community speak about the importance of animals in research.

For animal research statistics of countries around the world please see our statistics page.

Source of UK Statistics:

See previous years’ reports:

Speaking of Research

Switzerland’s animal research in numbers for 2016

The statistics for animal research conducted in Switzerland in 2016 were released today. We have translated these tables to English and these data are summarized below. Overall, there were 629,773 animals used in research and animal testing in Switzerland in 2016 — a 7.7% decrease compared to the previous year.


Animal research in Switzerland for 2016 by species [Click to Enlarge]


Animal research in Switzerland by species and use.
* We have added a column titled Total 2015 to aid with the ease of comparison to the previous year.

According to the Federal Veterinary Office, the BLV, this decrease is “mainly due to the completion of various projects with a large number of fish and amphibians”. Most of these animals were involved in basic research (64.4%), with “discovery, development and quality control” being the next most common (19.4%). The remainder were used for other reasons including disease diagnosis, education and training and protecting the environment. Mice were again the most prevalently used species (65.19%).


97.1% of the animals used in research and testing was conducted on mice, rats, fish, and birds, similar to other European countries. Monkeys (198), cats (411) and dogs (616) together accounted for 0.2% of all research animals, with an overall decrease of 2553 animals from 2015 for these species.

Severity in Switzerland works in the follow way (Translated definitions of the severity grading procedure for animals used in experiments in Switzerland):

The following four categories are used for constraints on animals resulting from procedures or measures in the context of animal experiments:

  • Severity grade 0: no constraint: Procedures and actions performed on animals for experimental purposes that do not inflict pain, sufffering or harm on the animals, engendr fear or impair their general well-being;
  • Severity grade 1: mild constraint: Procedures and actions performed on animals for experimental purposes that cause short-term mild pain or harm or mild impairment of general well-being;
  • Severity grade 2: moderate constraint: Procedures and actions performed on animals for experimental purposes that cause short-term moderate or medium-long term mild pain, suffering or harm, short term moderate fear or short to medium-term severe impairment of general well-being;
  • Severity grade 3: severeconstraint: Procedures and actions performed on animals for experimental purposes that cause medium to long-term moderate pain or severe pain, medium to long term moderate harm or severe harm, long-term severe fear or a severe impairement of general well-being

Pain, suffering and harm, were also measured and classified under four grades of severity; 0, 1, 2 and 3. In 2016, 38% of experiments were Grade 0, 31% were Grade 1, 21% were Grade 2 and 2% were Grade 3.


Severity of animal experiments by species, 2016


Severity grade data are only available from 1997 onwards

These numbers are relatively consistent across time, with on average 78% of all animals being exposed to no or minor short-lasting pain and distress.


Overall there has been a steady downward trend in the number of animals used in research in Switzerland over the last 30 years.

See details of Switzerland’s 2015 statistics.


Jeremy D. Bailoo

USDA publishes 2016 animal research statistics – 7% rise in animal use

The USDA/APHIS has published the 2016 animal research statistics. Overall, the number of animals (covered by the Animal Welfare Act) used in research in the US rose 6.9% from 767,622 (2015) to 820,812 (2016). This includes both public and private institutions.

These statistics do not include all animals as most mice, rats, and fish are not covered by the Animal Welfare Act – though they are still covered by other regulations that protect animal welfare. We also have not included the 137,444 animals which were kept in research facilities in 2016 but were not involved in any research studies.


Click to Enlarge

The statistics show that 52% of research is on guinea pigs, hamsters and rabbits, 10% is on farm animal species, while 11% is on dogs or cats and 9% on non-human primates. In the UK, where mice, rats, fish and birds are counted in the annual statistics, over 97% of research is on rodents, birds and fish. Across the EU, which measures animal use slightly differently, 93% of research is on species not counted under the Animal Welfare Act (AWA). If similar proportions were applied the US, the total number of vertebrates used in research in the US would be between 12 and 27 million, however, there are no published statistics to confirm this.


Comparing the 2015 and 2016 statistics there has been a small rise in the use of most species, apart from dogs (down 0.2%) and cats (down 5.2%). The largest rises were found in non-human primates (up 15%) and sheep (up 14%). Furthermore, it should be noted that this 6.9% rise comes a year after an 8% fall, putting the total number of animals used in 2016 slightly below the levels in 2014.

Animals used in researchand testing in the US 1973 - 2016

Trend in number of animals used in research in the US, 1973 – 2016 – Click to Enlarge

There has been a general downward trend in the number of animals used in the US over the past three decades; the number of animals used has more than halved (from 1.8million in 1986), with the use of dogs and cats down by over 65%. It is likely that a move towards using more genetically altered mice and fish has reduced the numbers of many other AWA-covered animals used. That said, non-human primates are one of the few species to have risen in use, from an average of 54,000 animals per year from 1977-2006, to 67,000 in 2007-2016.

In the UK, where mice, rats, fish and birds are counted in the annual statistics, over 97% of research is on rodents, birds and fish. Across the EU, which measures animal use slightly differently, 93% of research is on species not counted under the Animal Welfare Act. If similar proportions were applied the US, the total number of vertebrates used in research in the US would be between 12 and 27 million.

Rises and falls in the number of animals used reflects many factors including the level of biomedical activity in a country, trending areas of research, changes to legislations at home and abroad, outsourcing research to and from other countries, and new technologies (which may either replace animal studies or create reasons for new animal experiments).

The annual statistics are one example of openness and transparency in animal research, but the last few years have seen a greater number of institutions from all over the world publically acknowledging their animal research in statements on their website. This week, two separate openness initiatives were announced, with Americans for Medical Progress launching their “Come See Our World” website of free-to-use animal research images, and Understanding Animal Research promoting a 3D tour of four animal facilities in the UK.

Using the virtual tour you walk around real research facilities like this one at the University of Oxford.

On the subject of openness, it was disappointing that neither the USDA, nor APHIS decided to press release the figures when they were released on June 7th 2017, or even mention them in the website’s News and Announcements. The US could follow the past example of the UK, where the Home Office, in conjunction with the Science Media Centre, held a press conference each year to announce the annual statistics and to offer experts to explain and discuss the numbers.

Source of US Statistics:

Speaking of Research Coverage:

We will continue to bring you the latest national statistics as and when they are released.

Speaking of Research

Animal experiments in Israel rise by 51% in 2016

Israel used 507,018 animals for research and testing on animals in 2016 according to statistics released by The Ministry of Health’s Council for Animal Experimentation. This represents a 51% rise on 2015 – with the increase mainly due to a fourteen-fold increase in the number of cold-blooded mammals used (99% fish).

Testing on Animals in Israel for research in 2016. Click to Enlarge

There were moderate decreases in the number of rabbits, but the huge increase came from cold-blood animals, up from 12,784 in 2015 to 180,253 in 2016. According to the chairman of the National Council for Animal Experimentation, Prof Jacob Gopas, who spoke to Haaretz:

“If it’s possible to use fish, you don’t use mice, for example, and if it’s possible to use mice then you don’t use pigs,” Gopas says. Both the move toward using fish rather than mice in experiments and the efforts being made to raise fish with as few diseases as possible have contributed to the spike in the number of fish being used. Gopas notes that the vast majority of the fish used in research, 154,000 of the 178,000 that were used last year, were returned to their previous habitats.

No cats or dogs have been used in experiments in Israel since 2012. Primate numbers have edged up, increasing from 42 to 46 in 2016, though this is still less than 0.01% of total animal numbers. Primate experiments were under threat in 2014, resulting in seven Nobel Laureates, and seven major universities writing to President Netanyahu urging him not to further restrict animal studies.

Animals used in research in Israel in 2016. Click to Enlarge

Mice are still the most commonly used species in Israel, accounting for 51% of total animal numbers. Fish are the next most common at 35% (36% when other cold-blooded mammals are included). Rats and birds take the next two slots, with 8% and 4% respectively.

Trends in Israeli animal experiments 2004-16. Click to Enlarge.

Historical statistics show that until the spike in 2016, the number of animals has been fairly constant, fluctuating between around 275,000 and 340,000. The slight variations may account for individual projects which used a lot of animals, or from slight changes in science funding over the years. It appears that the sharp rise in 2016 is due to one or more research projects specifically working with fish – that account for most of the 170 thousand rise in animal numbers.

It should also be noted that Israel works hard to rehabilitate animals used in research. According to Israel Hayom,

The council noted that in 2016, its post-testing animal rehabilitation rate for monkeys, carnivores, farm animals and wildlife, excluding rodents and poultry, stood at 88%.

“The rates of animal rehabilitation in Israel are outstanding by any standard. Israel has been a leader in this field for years compared with the data published worldwide,” the council said, adding that it has so far funded nine projects aimed at developing methods that would minimize animal testing by finding alternatives that would not compromise research studies.

More information (in Hewbrew) can be found about the severity of animal experiments in Israel in 2016. The statistics show that 14% of projects were categorised as severe, 31% as moderate, 28% as mild, 18% as below mild, and 9% killed humanely for the purpose of collecting organs (not all countries collect this data). It is likely that projects are estimated at or above the actual severity level, and researchers would be in breach of protocol if they exceed their estimated severity.

Check out all the latest international statistics on our Animal Research Statistics page.

Speaking of Research

monkey animal experiment playing

Monkey on an Israeli primate breeding facility


Czech Republic sees 2% fall in animal research numbers for 2016

The Czech Republic has reported a 2.1% fall in the number of animal research procedures in 2016, with 229,465 procedures on animals. This is down from 234,366 procedures in 2015. The falls were mainly in fish (down 11%) and rats (down 17%), while the biggest rise was in birds (up 17%).

Procedures on animals in the Czech Republic for research and testing in 2016. Click to Enlarge

Fish were the most common animal used (35%), followed by mice (33%), birds (13%) and rats (11%). Collectively these four species accounted for over 92% of animal research in the Czech Republic (in line with other European countries). Dogs, cats and primates together continued to account for less than 0.5% of research procedures (919)

The most common areas of research were “basic research” (35.4%),  “Conservation of the natural environment in the interests of the health or welfare of people or animals” (21.0%) and “Translational and applied research” (11.4%).

The trend in animal experiments in the Czech Republic. Click to Enlarge.

The number of animals used since 2013 has remained quite flat, at around 230,000 procedures, though it is not immediately clear why. The drop since 2012 may be a result of the new reporting criteria brought about by EU Directive 2010/63, which came into force in 2014 (though some countries implemented new counting procedures before then).

Source of Czech Statistics:

We will continue to bring you the latest national statistics as and when they are released.

Speaking of Research

Animal Research in South Korea in 2016

In February 2017 the Animal and Plant Quarantine Agency (APQA) of South Korea released its animal research numbers for 2016. We spoke to the Animal Protection & Welfare Division and have been able to get a translated copy of these figures. The tables below were produced by the APQA, and we thank Dr Lee for providing these figures.

In 2016, South Korea used 2,878,907 animals in research, up 14.8% from the previous year.

Animal research in South Korea for 2016 by species

Rodents, fish and birds accounted for over 97% of animals used in research – similar to figures found in Europe. Most of the rise in animal experiments came from an increase in rodents (+19.5%), though numbers for fish (+15.2%) and birds (+60.7%) also contributed. There were falls in several categories, including primate experiments, which fell 18.8%.

Severity of animal experiments in South Korea

South Korea also produced severity statistics, similar to those in Europe. 2.6% of research showed no harm to the animal, 28.4% was mild, 35.5% was moderate and was 33.4% severe.We are unclear if these categorizations are based on pre-experiment licenses (what the researcher believed the severity would be) or post-experiment evaluation (what the researcher saw the severity to be).

Trends in South Korean animal experiments 2008-2016

The number of animals used in research has risen sharply over the last nine years, up 279% over the period, rising at a fairly steady rate of over 250,000 animals per year. To see why, take a look at a graph, produced by Nature, on the growth of R&D in South Korea over the same period.

The huge rise in spending on basic and applied research means that animal experiments were likely to rise (and did) over the same period. In 2013, South Korea had more researchers per thousand people in employment (12.84) than Japan (10.19), the USA  (8.81) or Germany (8.54). Medical and health sciences were the largest discipline (by publications) in South Korea (see Nature article).

If you know of any animal research statistics not on our list, please contact us.