Junk Science

Why Research Mice And Rats Are “Vanishing” Mid-Study

When mice and rats vanish in the middle of high-stakes scientific experiments, it's a tad suspicious

Junk Science
Jan 04, 2016 at 4:06 PM ET

Science has a credibility problem—and it doesn’t help that some researchers are apparently unable to keep track of their own lab animals. Mice and rats routinely “vanish” throughout the course of scientific studies, according to a new paper published in PLoS Biology. Regardless of the reason for the missing rodents (negligence, tampering, Pinky and the Brain trying to take over the world) such errors have skewed the results of hundreds of animal studies.

This paper is among the first in a series of studies due to be published by PLoS Biology as part of a new section of the journal dedicated to “meta-research” or, more precisely, the study of studies. Experts will submit papers to this section in an effort to critique inaccuracies and poor methodologies, with studies that (hopefully) steer the next couple of years of science in the right direction.

More Junk Science: Mouse Trials Are Meaningless

Meta-research of this kind cannot come soon enough. One 2005 study found that, given current research practices, scientific claims are more likely to be false than true. A recent effort to replicate 100 psychology studies found that only 39 percent could be reproduced. Studies suggest 85 percent of research investments (roughly $200 billion) is wasted on bad science every year, and that, specifically, $28 billion of that cash is being spent on preclinical research that cannot be replicated. Inefficient research is a heavy burden on the scientific community.

Nonetheless, whipping science into shape is a tall order. For instance, the first study in PLoS Biology’s new section involves an analysis of 441 biomedical journal articles on PubMed since 2000. The authors report that only one of the articles provided a complete summary of the research protocol, and none of them made the raw data publicly available. Most studies didn’t specify where the funding came from, and they often failed to disclose conflicts of interest. In other words, there’s no way to holistically assess the validity of 440 out of 441 studies.

Which brings us to the curious case of the disappearing mice and rats.

The second meta-research study published by PLoS revolves around missing mice—and its findings are even more worrisome. Researchers examined hundreds of stroke and cancer experiments involving mice and rats. They were surprised to find that the majority of studies failed to mention how many mice or rats were actually used in the experiments and that, when they did list the amount of rodents involved, the animals appeared to “vanish” throughout the course of the study without any explanation as to where they went.

More The Fraud Behind Some Of The Most Common Prescription Drugs

Now, even robust mouse trials have their issues (the research is important, but the results very rarely translate directly to humans) but when scientists fail to report the number of mice involved, it’s hard to know whether we can take the findings seriously. Without a sample size, we can’t figure out whether the results are significant—whether we’re looking at a cancer drug that worked on three mice or 3,000. Worse, if mice disappear mid-study, one has to wonder whether the scientists intentionally excluded certain mice after the fact, tampering with the results to show statistically significant effects even when there are none.

That’s an expensive problem. Pharmaceutical companies (and the U.S. government) spend billions of dollars testing potential drugs on humans and primates, all based on the preliminary findings reported in mice and rats. If the accuracy of these mouse studies cannot be verified, that means we’re throwing money at new drugs without any real reason to believe they might work. When mice and rats disappear without any note from the researchers accounting for the missing rodents, that omission harms the entire drug development process.

Worse, it gives the public yet another reason to mistrust scientists. And, in a world full of science deniers who seize on isolated instances of scientific inaccuracy to undermine vaccines, climate change and GMOs—science credibility is more important than ever.