Animal Experimentation: Trying to Measure Progress

How do we know if we’re succeeding? It’s a question we are always trying to answer. If we want to improve research and testing by reducing animal experiments, we need to know how many animals are used in laboratories. But we don’t.

Rough estimates of how many animals are used in United States laboratories are as high as 100 million annually, but we have little idea whether that number is accurate. More importantly, we have no idea whether that number is going up or down. This lack of clarity can largely be blamed on a problem unique among industrialized nations – namely, our only federal law designed to regulate the use of animals in labs excludes from the definition of “animal” more than 90 percent of animals used in labs. Most rats and mice and all birds and cold-blooded vertebrates are not covered by the Animal Welfare Act.

For the small percentage of animals covered by the Animal Welfare Act, each research facility must send the U.S. Department of Agriculture (USDA) an annual report listing the number of animals used, categorized by species. Unfortunately, the use of these reports as a monitoring system hit a major snag earlier this month, when the USDA abruptly shut down its animal welfare database.

While the online database was very useful, even when research facilities’ annual reports were available, it’s not clear how much the numbers could be trusted. Under the Animal Welfare Act, every U.S. research facility must have an Institutional Animal Care and Use Committee, which is in charge of evaluating proposed requests for animal use and ensuring legal standards of animal care and husbandry. These committees are the primary entity in charge of compliance with federal law, but according to a USDA audit,over a three-year period ending in 2011, almost half of all them were found to be in violation of the law. One of the primary reasons for the committees’ 1,379 Animal Welfare Act violations over that time was that the groups “did not recognize the importance of submitting an accurate annual report.”

Compare the U.S. system to that of the European Union, where all vertebrates (including rats, mice, birds, and fish) and even octopuses and squid are covered by the region’s law on animal use for scientific purposes. That law also results in the collection of information on not only how many animals and what species are used in experimentation each year but also where the animals originate and for what purpose they are used (broadly speaking). This information, which has been published every three years by the EU in a public report, shows that between 2008 and 2011 the number of animals decreased by more than a half-million. Under a new law, the next report, including even more detail, will be published in 2019. If we had similar information in the U.S., we could know if efforts to reduce animal experimentation are succeeding or failing and which efforts are most successful.

On this side of the Atlantic, how do we get closer to a more transparent system? The best step might be to reverse the 2002 congressional action that changed the Animal Welfare Act’s definition of “animal” to exclude the species mentioned above. That’s unlikely given the current divisiveness on Capitol Hill. However, Republican Rep. Ken Calvert of California and others are trying at least to require that more information about the federal government’s use of animals in testing be made available to the public.

In addition, three scientists at Hannover Medical School in Germany think they might have one answer to this problem of transparency—registries of animal studies, which could include details of each experiment and animal numbers. As the journal Science reports, these registries would address the problem that half of all animal experiments are never published in scientific journals. In contrast to the dearth of publicly available numbers on animal studies, the U.S. Food and Drug Administration already mandates that researchers preregister human clinical trials online.

Ultimately, whatever the system of transparency, it’s clear more is needed. Unless the shutdown of USDA’s animal welfare database is either reversed by the agency or overturned by a court, organizations like ours will be forced to wait months or years to receive records via federal Freedom of Information Act requests, and  may have to bring separate lawsuits against the USDA to obtain the information. Meanwhile, numbers for most animals used in U.S. laboratories aren’t available from any central repository. The federal National Institutes of Health funds approximately $13 billion worth of animal experiments each year. The public deserves to know how that money is spent, and whether we are moving in the right direction—away from animal research.