The pesticide and plastics industry have a lot invested in the safety of chemicals like bisphenol A and atrazine. Such “endocrine-disrupting” chemicals mimic human hormones, and research has tied them to health problems like cancer and infertility. But these industries have always held up studies that look at exposure to huge doses of endocrine disruptors. In massive quantities, the industries point out, these chemicals don’t cause problems. Therefore, they must be safe.

But those huge doses may actually obscure the chemicals’ effects, a new study argues. Endocrine-disrupting compounds “can have effects at low doses that are not predicted by effects at higher doses,” the authors write. In other words, low levels of exposure to these chemicals — like the levels that you’d get from, say, drinking water out of a BPA-laced plastic bottle — can have worse effects than high levels of exposure.

It’s a little confusing, because we’ve been trained to assume that more chemicals equal more problems. But some chemicals affect our bodies differently in small and large doses. Environmental Health News pulls out this example:

Reader support helps sustain our work. Donate today to keep our climate news free. All donations DOUBLED!

The breast cancer drug tamoxifen “provides an excellent example for how high-dose testing cannot be used to predict the effects of low doses,” according to the report. At low doses, it stimulates breast cancer growth. At higher ones, it inhibits it.

Grist thanks its sponsors. Become one.

Industries that rely on these chemicals also claim to have proven that low doses aren’t dangerous. But the study authors aren’t buying those claims:

vom Saal and other scientists have said that tests that do not find low-dose effects of chemicals such as BPA are often industry-funded, and they often have tested the wrong animals or the wrong doses, or don’t expose the animals during the most vulnerable time of fetal growth …

“There truly are no safe doses for chemicals that act like hormones, because the endocrine system is designed to act at very low levels,” Vandenberg, a postdoctoral fellow at Tufts University’s Levin Lab Center for Regenerative and Developmental Biology, told Environmental Health News.

Believe whoever you want, but given industry’s long history of funding obfuscating research on health issues like these, it’s probably safer to side with the scientists.

Grist thanks its sponsors. Become one.