When Good Science Goes Bad
Thursday, April 12, 2012
When I was working on my doctorate in entomology, I was convinced that all scientists were pure in their pursuit of knowledge and the truth. I still believe most are, but 29 years in this industry have taught me that scientific bias is alive and well, particularly when it comes to a high-profile subject, such as honey bees and pesticides.
Recently we have seen several sensational reports in the media that claim scientists have at last determined that neonicotinoid insecticides are the cause of honey bee declines around the world. This makes a good story, except for the troubling fact that it isn’t true.
The latest articles in Science Magazine reflect a disturbing trend in scientific research and publications. In this case, researchers conducted interesting studies and, with the help of a willing media and a well-coordinated PR campaign, completely overstate their importance. You might call it the triumph of the trivial.
And now we find a press release accompanying a study to be published in the Bulletin of Insectology, which claims imidacloprid is the “likely culprit” behind the worldwide decline in honey bee populations. Unfortunately, the study is heavily biased, poorly designed, factually inaccurate, and seriously flawed. This seems to be a classic case of designing a study to produce the results you want to achieve. I call it bad science.
Why is it these scientists come to conclusions which are contrary to our own testing experiences? In some cases, well-intentioned scientists make claims based on a faulty premise or experimental design, which tends to reinforce the narrative they have come to believe – that is that pesticides must be responsible for the collapse of bees, because that’s what everybody thinks. And unfortunately a few cling to this narrative even when their own research shows otherwise.
The problem of bee health is real and extremely complex. Most experts believe the decline in honey bee populations is due to multiple factors, particularly parasitic mites and associated pathogens. Poor bee health correlates extremely well with the presence of Varroa mites and diseases, but does not correlate at all with pesticides. Unfortunately, the attention given to neonicotinoids overshadows research in other areas and diverts our focus on finding real solutions to the real problem of bee health.
People are entitled to their opinions, but scientists have a higher responsibility and are called upon to test their theories using the scientific method. In the rush to publish, some scientists fail to pressure-test their assumptions, or to explain why their results do not concur with others. These are not bad people, but the science is bad and this can drive others to make bad decisions. With the demand to help feed a hungry planet at an all-time high, the stakes could not be higher.
Fortunately, we have a focused effort in Bayer CropScience to tackle the real problem of bee health and I encourage each of you to learn more about what Bayer is doing to help. Attempts by some to ban useful crop protection products not only would be detrimental to modern agriculture, but also would not improve honey bee health. When good science goes bad, everyone loses.
Jack Boyne, Ph.D
Director of Communications
Bayer CropScience, LP
Check out a post from Randy Oliver of Scientific Beekeeping, a beekeeper in California, who was one of many who posted a rebuttal to the Harvard paper on imidacloprid and bees.
Statements from the Bayer CropScience Newsroom:
Bayer CropScience Says Bee Study is “Seriously Flawed” - Press Statement