How many mice and rats are used in U.S. labs? Controversial study says more than 100 million

Cages of lab mice at an unnamed institution


The most numerous mammals in U.S. research are also the most invisible. Mice and rats comprise the vast majority of lab mammals, yet no one knows exactly—or even approximately—how many are used in scientific experiments every year. Now, for the first time, someone has attempted to calculate this number using data from U.S. labs, and it’s big: More than 111 million mice and rats are used annually in U.S. biomedical research, according to a new study. That represents more than 99% of all lab animals.

“It’s a very thoughtful and reasonable analysis,” says Sue Leary, president of the Alternatives Research & Development Foundation (ARDF), which seeks to reduce the number of animals in labs and find replacements. The figures are troubling, she says, because mice and rats are not covered by the federal Animal Welfare Act (AWA), which enforces the humane treatment of research animals. “If the numbers are anywhere near correct, the amount of pain and suffering that’s occurring in these animals is completely unacceptable.”

But many in the biomedical community say the figures, published today in Scientific Reports, are a gross overestimation, and that the study itself is deeply flawed. “It’s a really disappointing analysis,” says Allyson Bennett, senior editor at Speaking of Research, which advocates for the use of lab animals. The organization’s own estimates—extrapolated from European labs, which do count mice and rats—put the number or rodents used annually in U.S. facilities at 10 million to 25 million. That would represent 93% to 97% of all U.S. research mammals. Bennett also disagrees with the idea that the animals are at risk of being mistreated because they are not covered by the AWA.

Signed into law in 1966, Congress amended the AWA in 1970 to cover “all warm-blooded animals.” Yet the U.S. Department of Agriculture (USDA), which enforces the act, declined to apply it to most birds, rats, and mice, citing a lack of resources. Several animal advocacy groups, including ARDF, fought the exclusion. But in 2002 Congress amended the AWA to formally exempt these creatures, with very limited exceptions. That means USDA has never counted most rodents or tracked the type of experiments in which they’re used.

Both animal and biomedical advocates have made their best guesses. The National Association for Biomedical Research (NABR) has stated—based on its knowledge of research facilities—that 95% of all lab mammals in the United States are rodents. That would equal about 14.8 million animals, similar to Speaking of Research’s average. ARDF’s number, also extrapolated from European countries, is about 23 million.

Larry Carbone made his own guesstimate when writing a book about animal welfare policy in the early 2000s. Carbone, a veterinarian who worked for 4 decades in laboratory animal care at Cornell University and the University of California, San Francisco, used his knowledge of academic labs to come to a figure of 80 million to 100 million mice and rats. “It was just a half sentence, but that’s the line people picked up on from the book,” he says with a laugh. “Since then, there has been no attempt to do a more rigorous evaluation.”

In the new study, Carbone tried to take a more scientific approach and focus on U.S. data. He also did something he had long been the target of when he ran animal facilities: File Freedom of Information Act requests. Animal rights groups often use such requests to dig up animal welfare infractions, which they then publicize to support their cause. But Carbone’s efforts were much more focused.

He made records requests of large public scientific institutions, asking only for the number of mice and rats they reported to the Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC) International in 2017 and 2018. The organization inspects many U.S. animal facilities, where it counts rodents and other animals, but it keeps the figures confidential. Carbone’s expertise in the field helped. “I knew exactly which appendix of their annual report to point them to,” he says.

Carbone also emailed people he knew at private institutions, which are not subject to public records requests, for their rodent records from the same period.

In all, he was able to collect data from 11 public and five private facilities, representing 16 of the top 30 funded by the U.S. National Institutes of Health. The institutions reported using a total of about 39,000 animals covered by the AWA, including monkeys, dogs, and rabbits. They also reported using more than 5.5 million mice and rats.

That means rats and mice made up 99.3% of all mammals used at those places. If all U.S. institutions use the same proportion of these rodents versus other mammals, then USDA numbers include just 0.7% of all mammals used. Given that USDA reported about 780,000 animals at all biomedical facilities in 2018, Carbone concluded that approximately 111 million mice and rats are being used in U.S. research every year. His data also indicate that mice make up the vast majority of these rodents, more than 97%.

Carbone says the figures argue for including mice and rats in the AWA. “We keep telling the public, ‘Don’t worry about research animals because there is rigorous government oversight,’” he says. “But that oversight covers less than 1% of the animals in labs.”

Leary says the numbers are also important if scientists are serious about one of the cornerstones of biomedical research: the three Rs. The biomedical community has been saying for decades that it is dedicated to reducing and replacing animals in labs, and to finding ways to minimize their suffering, she notes. “How can you measure your progress on the three Rs if you’re not even counting the animals?”

Peter Smith, associate director of Yale University’s Animal Resources Center, thinks Carbone’s estimates are close to the mark. “It’s a well–thought-out approach,” he says, “and about as good as anyone can do.”

Others in the biomedical community disagree. The new analysis is “fundamentally flawed,” and the numbers are “inaccurate and vastly overstated,” says Nadia Rosenthal, scientific director of the Jackson Laboratory, which supplies a broad spectrum of mouse strains to universities and other institutions. Spending and shipping numbers from Charles River Laboratories, the world’s largest commercial distributor of rodents, suggest commercial breeders supply, at most, 15 million mice and rats to U.S. labs every year, Rosenthal says. Combined with what labs themselves likely breed, she says the total number of those animals is likely closer to 20 million.

Bennett faults the study for extrapolating from a handful of top-funded academic institutions to more than 900 other biomedical facilities, including less well-funded academic centers, pharmaceutical companies, and contract research organizations. “If the same ratios of mice and rats don’t hold at these places,” she says, “the entire exercise is invalid.”

Bennett also takes issue with the idea that omitting mice and rats from the AWA means they are unprotected. She notes that all labs that receive federal funding must answer to institutional animal care and use committees, which follow the three Rs and oversee the welfare of all lab animals. Those that don’t receive federal funds typically submit to AAALAC inspections, she says, which also enforce humane standards. “It is unlikely that there are large number of facilities that are uncovered by any regulation.”

Still, Cathy Liss, president of the Animal Welfare Institute, which has also fought to include mice and rats in the AWA, says only this federal law gives someone the ability to file a complaint that must be investigated. “These animals merit the same level of protection that other animals get.”

Matthew Bailey says that’s not realistic. The president of NABR, which has repelled efforts to include rats and mice in the AWA, he says the government doesn’t have the time or money to track so many animals, and that doing so would drain vital resources. “Now is not the time to be seeking additional restrictions on biomedical research or endeavoring to make it more difficult and more expensive.”

Smith agrees that adding mice and rats to the AWA would be a “huge administrative and taxpayer burden” and probably wouldn’t improve the lives of the animals, which he argues are already well cared for.

Carbone says he hopes his study will start an important conversation about whether to count mice and rats. “If we’re serious about reducing the number of lab animals we use and curbing the number of painful experiments, somebody has to be keeping track of these animals,” he says. “You can’t track progress if you can’t measure progress.”