Last week, an article appeared in The Lancet Neurology (doi:10.1016/S1474-4422(13)70278-3) which reviewed studies and asserted that there is a "pandemic" of developmental toxicity. This led to a press release and a variety of stories in the media linking things like pesticides to brain disorders in children.

The new article is an update of the authors’ 2006 paper on the same topic. Unfortunately, the authors continue to forget or ignore the fundamental principle of toxicology that underpins the effects that chemicals can have on living organisms, dose-response. As a result, this 2014 review suffers from the same use of faulty logic applied in 2006 in Grandjean and Landrigan’s apparent desire to promote their opinions without considering the basis in reason or scientific method.

As before, Grandjean and Landrigan compile a list of chemicals that they claim is expanding in terms of having been identified as causing developmental neurotoxicity in humans, again relying on epidemiological studies where there was no clear measure of exposure. They mention ethanol as an example of a human developmental neurotoxin, which exemplifies their faulty logic. Abuse during pregnancy causes fetal alcohol syndrome, which is a highly variable group of birth defects that includes mental retardation, low birth weight, and malformations of the skull and face that tend to occur in babies born to mothers that repeatedly ingested high doses of alcohol during pregnancy.

Pesticides were implicated in ADD, autism and cognitive delays in a new paper. But does the science show that? Image link: Harvard School of Public Health

However, there are no data to show that maternal exposure to low, environmentally relevant levels of ethanol (i.e., concentrations that are known to be present in foods we eat every day) places the fetus at any risk of developmental toxicity. It is the dose of the chemical, and the pattern of exposure, that determines whether a chemical produces an adverse effect on an organism, not simply the presence of a chemical, even for developmental neurotoxicity.

Just as a critical concentration at the site of action is needed before a drug can produce its beneficial effects in humans, the same principle applies to toxicity produced by a chemical. Effects that might be reported at high doses will not occur at lower doses if the concentration at the site of action falls below the threshold for toxicity.

As I pointed out in 2006, if Grandjean and Landrigan’s logic is applied to drugs, an “outbreak of cures” would be predicted to be triggered by any dose of any drug. Evidence-based medical practice has proven that this is not the case.

As a result, evidence-based toxicology and epidemiology dictates that the dose of chemical is the critical factor when examining the risk posed by a chemical and argue against a “pandemic” of developmental neurotoxicity associated with chemical exposure.

Although the issues of autism, preterm birth rates, pediatric bipolar disorder, and other neurological conditions mentioned by the authors clearly are issues of concern in medicine today, improvements in diagnostic methods and criteria for such diseases may account for much of the purported increase in developmental neurotoxicity that is discussed by Grandjjean and Landrigan.

As a result, Grandjean and Landrigan’s conclusions are flawed and lack a sound basis in toxicology and the science of risk assessment.


Philippe Grandjean MD, Philip J Landrigan MD, 'Neurobehavioural effects of developmental toxicity', The Lancet Neurology, Volume 13, Issue 3, Pages 330 - 338, March 2014  doi:10.1016/S1474-4422(13)70278-3