More than a quarter of a million animals were used for research, testing and teaching in 2016, new figures reveal.
The 2016 Ministry for Primary Industries (MPI) figures show the number of animals used in studies that were completed during the year and reported to MPI.
MPI manager of animal welfare Dr Kate Littin said the statistics showed that a total of 254,453 animals were used in research, testing and teaching in 2016.
This figure was up 13 per cent from the record low achieved in the previous year.
"The increase for 2016 is mainly due to more livestock used for veterinary research, testing and teaching," Littin said.
"It was good to see there has been a decrease in the number of animals that died or were euthanised, and that the number of animals that returned to owners or were released to the wild were up compared to previous years."
The number of animals that were rehomed was similar to last year, she said.
"While rehoming may not always be appropriate for all animals, there are groups that will work with research organisations to rehome animals, and we encourage organisations to do this wherever possible."
MPI took the role of protecting the welfare of animals used in research, teaching and testing very seriously, Littin said.
"The Animal Welfare Act 1999 requires a code of ethical conduct to be approved by MPI before any animals are used for research, testing and teaching.
"Each project must be scrutinised and approved by an animal ethics committee that has been established under the code."
The New Zealand Anti-Vivisection Society (NZAVS) is a non-profit organisation that campaigns to abolish animal testing.
NZAVS executive director Tara Jackson said it was an unsurprising report, that failed to address the problem of using these animals to predict a human response.
"If anything, it is just totally a reminder to the public that thousands of animals are still being used in New Zealand for this," Jackson said.
"It is an outdated process and it's time that we see a drastic change happen."
Of the animals categorised in the "high and "very high" impact categories according to MPI statistics, the large majority were rodents and rabbits.
Jackson said there were multiple reasons rodents and rabbits were used: they were commonly smaller animals, easier to handle, relatively non-aggressive and were not expensive.
"It's what has been entrenched in the industry for a long-time, which is unfortunate because when you do analyse the science correctly and independently you see that using rodents to model the human response is so flawed."
Jackson said these figures showed there was still a long way to go and the goal remained abolishing animal testing.
"That's the ultimate goal and I think one day we will get there."
Key statistics in the report
• In 2016, 26 institutions carried out research under their own approved codes of ethical conduct, and 109 under another organisation's approved code.
• Livestock accounted for almost 65 per cent of the animals used.
• Cattle were the most commonly reported species making up 41 per cent of the total number, followed by sheep at 19 per cent, mice at 13 per cent and fish at 8 per cent.
• The majority of animals underwent procedures which had no impact or little impact on them, at nearly 84 per cent of the total number.
• The number of animals that experienced "high" or "very high" impact in 2016 changed with the number of animals in the "high" impact category increasing by 11.9 per cent and the number within the "very high" impact decreasing by 35.3 per cent.
• Of the animals categorised in the "high and "very high" impact categories, the large majority were rodents and rabbits (6897 out of 8596). These animals were classified in these impact grades largely due to use in vaccine testing, veterinary research, and production and evaluation of biological reagents and other medical research detailed in the report.
• There was a decrease in the number of animals that died or were euthanised, with the number of animals that were returned to owners or released to the wild up from the previous year.