Soon the U.S. Environmental Protection Agency (EPA) will launch its regulatory review of the active ingredient in the weed killer Roundup. It’s called glyphosate, and the debate over its safety has already begun. The EPA has a long tradition of conducting thorough, objective reviews based on rigorous science. Unfortunately, glyphosate’s opponents are trying to influence the process by injecting irrational fear.

In my years as a pediatrician, an ICU physician, and a toxicologist, I have seen firsthand the damage that can result when fear undermines science. In 1998, for example, a research paper was published claiming a link between the measles-mumps-rubella (MMR) vaccine and autism. The paper was revealed to be fraudulent, and later studies clearly showed that MMR plays no role in autism. But the damage was done. Thousands of parents gave in to irrational fear and decided not to vaccinate their children. The result? Today measles is again a public health risk, and, because of guilt by association, pertussis is increasingly common. A polio outbreak in the United States could also become a reality and tragedy.

Now opponents of crops that are genetically modified to make them resistant to herbicides and pesticides are seizing on routine reviews of glyphosate by EPA and the International Agency for Research on Cancer as an opportunity. In an effort to restrict glyphosate use – and therefore indirectly plantings of genetically modified crops that have built-in resistance to it – they are urging the agencies to consider studies that have been produced linking the herbicide to everything from diabetes to cancer to, yes, autism. All of these studies can be easily debunked. They either rely on such a small sample size that they are statistically invalid, or they confuse correlation with causation. As any high school science student knows, just because two trends are occurring at the same time doesn’t mean they’re related. GMO opponents have even tried aggregating these studies to try to give them more weight. But combining bad studies does not magically create a good study.

In fact, glyphosate has been studied extensively using real science. In 2005, researchers published an Agricultural Health Study, evaluating almost 60,000 farm workers who had worked with glyphosate controlling for important factors like tobacco use. The study concluded that “glyphosate exposure was not associated with cancer incidence.”

The vast majority of corn, soybeans, and other crops grown in the U.S. are now glyphosate-resistant. Farmers like them because they make weed control much easier and farming more efficient. These and other GMO crops have increased yields here and are doing the same in poorer countries around the world that have often struggled to feed their people.

It’s unfortunate that misinformation is once again putting science on the defensive. A rational assessment of risk clearly shows that the risk of lower crop yields needed to feed hungry people far outweighs any risk from glyphosate-resistant or other GMO crops. We have seen the damage that irrational fear can do. I can only hope we have learned our lesson and won’t let history repeat itself.

Banner is the president-elect of the American Association of Poison Control Centers, a past president of the American Academy of Clinical Toxicology, and a past chair of the American College of Medical Toxicology.