As scientists who have researched the health effects of bisphenol A (BPA) for over a decade, we are concerned that a recent Food and Drug Administration decision raises questions about the agency’s ability to protect the health of Americans. The FDA’s decision to reject a petition to ban BPA in food packaging exposes a serious Catch 22 in the federal mindset.

Used in plastic food containers and canned food linings, BPA leaches into food and beverages, exposing people to this estrogen-like substance. We have contributed to the hundreds of peer-reviewed studies of laboratory animals that link BPA exposure to cardiovascular disease, diabetes, obesity, neurobehavioral abnormalities, allergies and asthma, prostate and mammary gland cancer and a wide range of reproductive defects. Further, similar effects have been reported in a number of human studies.

Why doesn’t the FDA consider these results evidence that BPA poses a risk to human health? The answer is simple: They have chosen to ignore most of this data, relying instead on a small number of studies conducted by FDA scientists and scientists funded by industries that manufacture BPA. In a statement released last March explaining its decision to reject the BPA petition, the FDA appeared to justify the omission of hundreds of independent studies from consideration because they used too few animals “to provide confidence in the conclusions." This statement is a jaw dropper because the FDA was previously criticized by its own Science Board for using exactly the same exclusion criterion in its 2008 assessment of BPA, which ultimately was rejected by the Science Board.

Decades ago, the National Institutes of Health published guidelines that restrict university scientists from using excessive numbers of animals in experiments. The NIH Guide mandates that only the smallest number of animals required to identify a significant effect and draw valid conclusions may be used in research conducted in academic settings. As university professors, we are required to follow NIH guidelines. The FDA should know that the strength of conclusions that can be drawn from data is not directly dependent on the number of animals used. The sensitivity of the methods employed, the precision of the measurements and the magnitude of the observed differences are the major factors.

The FDA’s persistent rejection of virtually all the research from independent, university laboratories is disturbing. On one hand, NIH has established a data-based method (power analysis) to ensure that the fewest possible animals are used in research. At the same time, the FDA is rejecting NIH-funded research findings on the grounds that the studies did not use “enough” animals. This is a federal Catch 22 that does not serve the public interest.

In contrast, researchers working in industry-funded laboratories are not bound by NIH guidelines. In its decisions regarding BPA, the FDA has relied heavily on a single industry-funded study that involved killing more than 8,000 rats, even though the study is seriously flawed by use of insensitive methods and lack of appropriate controls.

Science evolves rapidly, and we should be using the very latest and most sensitive techniques to determine if BPA and similar chemicals affect the development of our babies and our long-term health. Much of the research discarded by the FDA has been conducted by leading scientists using state-of-the-art methods of analysis and published in peer-reviewed journals. The FDA’s insistence that hundreds of these studies are not relevant is quite simply indefensible and puts the public health at risk.