Low-dose theory in the area of food safety – EFSA sets the record straight


The European Safety Agency (EFSA) has added more certainty to the discussion about the so-called ‘low dose’ theory: “it is so far not supported for substances in the area of food safety.”

Over the years there have been multiple allegations that very low exposure to certain chemical substances, such as Bisphenol A (BPA), could have more harmful effects on health than higher exposures. Scientists call this a non-monotonic dose-response (NMDR), meaning that the response does not always increase/decrease as the dose increases/decreases.

As we know, people are incidentally exposed to BPA in miniscule amounts. Some researchers have turned to the low-dose theory to attempt to argue for unjustifiably low thresholds on BPA, arguing that BPA is more harmful at low doses because the health response is greater at lower doses. But it turns out that this theory is not supported by the evidence.

In May, a study refuted the low-dose theory. The study was commissioned by the European Safety Agency (EFSA) and scientists from renowned institutions such as the Austrian Agency for Health and Food Safety GmhH (AGES), the French Agency for Food, Environment and Occupational Health & Safety (ANSES), the Dutch National Institute for Public Health and the Environment (RIVM), and the Swedish Institute of Environmental Medicine, Karolinska Institutet (IMM).

The agencies used more than 10,000 identified studies, looking for potential evidence for NMDR, as they performed a complex and complete scientific review of literature since 2002, which was published in May. The review did not find plausible evidence for non-monotonic dose-response for substances in the area of food safety. This of course includes BPA.

If non-monotonic dose-response (NMDR) is not an issue we should worry about then why have scientists claimed that low doses of BPA can be bad for us? Experimental data almost always contain errors. Therefore further statistical evaluation would be required by at least another independent study. It would reduce the possibility that conclusions are based on possible statistical mistakes rather than on actual biological phenomena.

Stats.org has published a detailed post examining the statistical issues in the studies that find an alleged non-monotonic dose-response. It tells us what, whenever you read BPA is bad for you even at very low doses, there are most likely problems with the study: is the ‘low’ dose really that low, or does it far exceed actual human exposure levels? Was the sample size adequate to really measure the response? If the test was conducted using animals, was the sample group selected to avoid familial effects? Did the design of the test favor a finding of low-dose effects?

The amount of BPA we ingest on average every day is tiny, and well below the level EFSA considers as the safety threshold. In most food contact materials made using BPA-based materials, the BPA does not leach into the environment as it is being “transformed” into another product. Studies have indeed consistently confirmed that in our everyday lives, for adults and babies, BPA is safely excreted after, being converted by the liver into a kind of sugar.

Leave a Reply or Comment

Your email address will not be published. Required fields are marked *