Last year, the Roundtable on Science and Welfare in Laboratory Animal Use organized a meeting in which scientists from around the world gathered to discuss the issue of reproducibility in experiments using animal models. They identified several factors that contribute to irreproducibility, including problems with the design and execution of animal experiments. Last month, a summary of this workshop was published. Considering the amount of press that such reproducibility issues have been getting lately, I thought it was important to share these findings with you.
It was clear from reading the workshop summary that the reproducibility issue with animal experiments is real, widespread and the result of many factors. One striking example of this was provided by a speaker who revealed that researchers at pharmaceutical giant Amgen were not able to reproduce data from 47 out of 53 groundbreaking publications on drug discoveries in cancer research. Unfortunately, many clinical trials and hundreds of follow-up publications were based on irreproducible data, putting people at risk and wasting animal lives.
Some attributed reproducibility issues to the pressure that researchers face to publish new and exciting results, as doing so increases their ability to receive grants and job promotions. Because of this “reward system,” scientists may feel compelled to report data selectively, or to place more emphasis on findings that support their hypothesis, which may lead to publication of irreproducible data.
Problems with study design, as previously reported in Science First, is another cause of irreproducible results. Lack of bias-reducing measures, poor protocols, inappropriate statistical tests and lack of important controls in experiments are all contributing factors.
While it is, of course, important to address the issue of reproducibility in science, I would be remiss not to bring up a larger issue—the issue of using animals in science at all. It is clear that poorly-designed studies using animal models are wasteful; but even more concerning is that “better-designed” animal experiments would still use models that lack human relevance. Rather than ignore the limitations with animal models and work to improve the design, execution and reporting of animal experiments, it is critical that researchers consider switching gears and placing their emphasis instead on the development and utilization of more human-relevant approaches that do not rely on animal models.
What are your thoughts on the emerging trend that many of the animal experiments conducted cannot be reproduced, providing strong evidence that investment in this type of research wastes time, resources and animal lives? Send your questions and comments to email@example.com. I look forward to hearing from you.
-Dr. Pam Osenkowski, Director of Science Programs
November 16, 2015
Reproducibility Issues in Research with Animals and Animal Models: Workshop in Brief
Scientific progress is achieved by robust experiments that generate reliable and reproducible results to be used with confidence by the research community. Recent publications have drawn attention to an apparent and concerning prevalence in the number of peer-reviewed studies that cannot be reproduced, particularly those containing data from experiments using animals and animal models.
For more information see: http://www.nap.edu/download.php?record_id=21835#