Skip to content

Why Journalists Should Stop Publishing Studies Conducted With Mice

When it comes to determining what will make humans healthy and disease-free, research involving mice has had remarkably few successes.

A researcher holds a mouse.
A researcher holds a mouse in an in vitro fertilization lab in 2002. In addition to the fact that mice have a number of biological and genetic differences from humans, a lot of mouse research is of extremely low quality. (Joel Page/The Associated Press)

On a near-daily basis, the public is bombarded with news headlines announcing the latest breakthrough” findings related to cancer, obesity, psychological disorders, and so on. Here are just a few examples from the past couple of years: 

The problem is that the research studies highlighted in the above articles — and in thousands of articles like them — were based on animals, usually mice. That is, not humans. So as exciting as these studies might sound, journalists should actually stop covering them, period.

To be sure, studies involving mice can be immensely useful for some purposes. Mice are relatively cheap, they are plentiful, they can be genetically bred for specific aims, and they can be killed and dissected to see how they were affected over the course of an experiment. If you want to know whether a substance is so toxic that it will make an organism drop dead or cause its brain to stop functioning normally, give it to a mouse.

But when it comes to determining what will make humans healthy and disease-free, research involving mice has had remarkably few successes.

Take stroke research, for instance. One team of researchers found that out of about 500 treatments reported to improve stroke outcomes in animals, only two treatments worked in humans (and one of those was aspirin, with its blood-thinning effect).

ALS — or Lou Gehrig’s disease — provides another example. The head of the ALS Therapy Development Institute wrote: We have tested more than 100 potential drugs in an established mouse model of this disease … Many of these drugs had been reported to slow down disease in that same mouse model; none was found to be beneficial in our experiments.” He further noted that eight of these compounds ultimately failed in clinical trials, which together involved thousands of people.”

Or, consider cancer research. Many of the breakthrough” cancer experiments touted in mainstream news reports turn out to have been performed on mice. Yet, as Terry Van Dyke of the National Cancer Institute said, the success rate of such research has traditionally been only about 5%,” because the mouse cancer doesn’t truly reflect the original cancer in the human.”

Finally, there are studies related to sepsis, burns, and trauma. Unsurprisingly, it is far easier to perform such research on mice than on humans. However, the research with mice may be largely useless. As The New York Times reported — based on an article in the Proceedings of the National Academy of Sciences — mice simply don’t have the same immune system or genetic makeup as humans. Therefore, the reason why every one of nearly 150 drugs tested at a huge expense in patients with sepsis has failed” may be that the drug tests all were based on studies in mice.” One researcher, upon reading the paper, was quoted as saying: I was stunned by just how bad the mouse data are. It’s really amazing — no correlation at all [with humans.]”

In addition to the simple fact that mice have a number of biological and genetic differences from humans, a significant amount of mouse research is of extremely low quality. A recent review of 2,672 animal experiments showed that between 70 and 99 percent of them failed to include one or more of the basic scientific techniques that make human clinical trials reliable, such as using randomization, blinding the researcher to the treatment, and using a sufficient sample size. Due to these quality problems, most mouse research doesn’t even provide reliable information about mice, let alone humans.

Of course, there are many efforts underway to improve the quality and reproducibility of research on mice and other animals — including a replication project that my foundation is funding. And newer genetic engineering methods may make some mouse research considerably more replicable in humans.

But for now, what should journalists do when faced with yet another breathless press release on the latest mouse study claiming to have found a possible cure for cancer?

My answer: As soon as journalists see the word mouse” in a study (or rat,” hamster,” etc.), they should put down the press release, delete any related emails, and refuse to write a word about that study. As a society, we waste billions of dollars and innumerable hours pursuing remedies and diets and exercise plans that may not even work in humans. The media bear a great deal of responsibility for this. To be fair, there have been many instances in which a reporter has pointed out that the study featured in his or her article was conducted on mice and may not be applicable to humans. But this is not nearly enough. When readers see a breakthrough” treatment or diet plan advertised in the headline, they have already been misinformed — even if there is a modest disclaimer several paragraphs later.

Like oncologist and author Vinay Prasad says, a newspaper article on a mouse study is roughly the equivalent of interviewing someone who bought a lottery ticket and asking them what they plan to do with the money. Better to wait until the lottery ticket actually wins and then write a story about it.