Focus on Arts and Ecology

Nearly every scientist who has used mice or rats to study depression is familiar with the forced-swim test. The animal is dropped into a tank of water while researchers watch to see how long it tries to stay afloat. In theory, a depressed rodent will give up more quickly than a happy one — an assumption that has guided decades of research on antidepressants and genetic modifications intended to induce depression in lab mice.

But mental-health researchers have become increasingly sceptical in recent years about whether the forced-swim test is a good model for depression in people. It is not clear whether mice stop swimming because they are despondent or because they have learnt that a lab technician will scoop them out of the tank when they stop moving. Factors such as water temperature also seem to affect the results.

“We don’t know what depression looks like in a mouse,” says Eric Nestler, a neuroscientist at the Icahn School of Medicine at Mount Sinai in New York City.

Now, the animal-rights group People for the Ethical Treatment of Animals (PETA) is jumping into the fray. The group wants the US National Institute of Mental Health (NIMH) in Bethesda, Maryland, to stop supporting the use of the forced-swim test and similar behavioural assessments by its employees and grant recipients. The tests “create intense fear, anxiety, terror, and depression in small animals” without providing useful data, PETA said in a letter to the agency on 12 July.

The animal-rights group also singled out NIMH director Joshua Gordon for using the forced-swim test in the early 2000s, when he was a researcher at Columbia University in New York City.

“The National Institute of Mental Health has for some time been discouraging the use of certain behavioral assays, including the forced swim and tail suspension test, as models of depression,” Gordon said in a statement to Nature. “While no single animal test can capture the full complexity of a human disorder, these tests in particular are recognized by many scientists as lacking sufficient mechanistic specificity to be of general use in clarifying the neurobiological mechanisms underlying human depression.”

But Gordon said that the tests are still “crucial” for some specific scientific questions, and that the NIMH will continue to fund such studies.

Although scientists insist that behavioural tests that cause stress in animals are necessary for developing human treatments, the PETA campaign dovetails with scientists’ growing concern about the quality of data produced by forced-swim tests, says Hanno Würbel, a behavioural biologist at the University of Bern. “The point is that scientists shouldn’t use these tests anymore,” he says. “In my opinion it’s just bad science.”

Scientists developed the forced-swim test in the 1970s. One of its earliest applications was studying the efficacy of drugs known as selective serotonin reuptake inhibitors (SSRIs) — a class of antidepressants that includes Prozac (fluoxetine). Mice and rats that received SSRIs swam for longer periods than animals that did not.

The test’s popularity grew in the early 2000s, when scientists began modifying mouse genomes to mimic mutations linked to depression in people. Many of these researchers adopted the forced-swim test as a “quick and dirty” way to assess their ability to induce depression, even though it was not designed for that purpose, says Trevor Robbins, a neuroscientist at the University of Cambridge, UK.

By 2015, mental-health researchers were publishing an average of one paper a day that used the procedure, according to an analysis by researchers at Leiden University in the Netherlands1. Yet the swim test’s track record is mixed. It has accurately predicted whether different SSRIs are effective treatments for depression, but yields inconsistent results when used with other types of antidepressant.

And some aspects of the SSRI results are puzzling. Mice given the drugs show measurable changes in behaviour during swim tests beginning one day after treatment, whereas in people SSRIs often take weeks or months to reduce symptoms of depression.

Due in part to concerns about the forced-swim test’s accuracy, major drug companies such as Roche, Janssen and AbbVie have abandoned the procedure in recent years.

Many researchers feel obligated to use the test, says Ron de Kloet, a neuroendocrinologist at Leiden University Medical Center and a co-author of the 2015 study. “People get their grants based on this test, they write papers based on the test, they make careers,” he says. “It’s a culture which keeps itself alive, even though most of them will admit that the tests are not showing what they are supposed to do.”

Todd Gould, a neurobiologist at the University of Maryland School of Medicine in Baltimore, acknowledges the poor track record of the forced-swim test. But he says the procedure has proved useful for his research into whether the party drug ketamine and related substances are effective antidepressants2.

Gould finds it ironic that an animal-rights group is attacking the NIMH, because Gordon and several of his predecessors have been outspoken advocates of developing objective biological measures of depression and other mental-health disorders. In practical terms, that has meant looking for alternatives to many animal tests. Perhaps unsurprisingly, Gould says that NIMH grant reviewers have tended to push back against proposals of his that have included forced-swim tests.

The agency told Nature that it requires grant applicants to supply written justification for using animals in research, and that its review system “evaluates these descriptions very rigorously to determine whether the use of the animal proposed is appropriate and justified”.

Emily Trunnell, a research associate at PETA’s Laboratory Investigations Department in Norfolk, Virginia, says that the group decided to target the NIMH because of the agency’s prominence in mental-health research. “We believe that if NIMH took a stand, it would set a strong precedent,” she says.

She argues that emerging technologies, such as ‘mini-brains’ grown from human stem cells, could eliminate the need to use rodents in depression studies. Researchers are already using these clumps of human tissue to study the genetics and brain wiring that underlie various mental-health disorders3.

But some scientists say that the best replacement for the forced-swim test may be more sophisticated tests that involve rodents or other animals. Robbins says that one approach could include developing animal tests that accurately measure specific symptoms of depression, such as lack of interest in a favourite food.

And Nestler says that modelling individual signs of depression may produce better data than do attempts to mimic the full complexity of the human disorder in animals. The symptoms and underlying genetics of depression seem to vary widely between people, and the same treatments don’t work for everyone.

“We know human depression is not one disease,” he says.