Article misses fundamental points of human study designs.

Posted by Joe Braun and Heather Volk

A recent Los Angeles Times article by Marissa Cavallos described three new studies reporting that pesticide exposure during pregnancy may be associated with lower IQ in children. The three studies – conducted in New York City and Salinas, Calif. – measured pesticides in pregnant women and followed their children for up to nine years. They found that children born to pregnant women with higher pesticide concentrations had lower IQ scores.

Cavallos misses four important features of these studies: 1) effects of small IQ shifts, 2) pesticide exposure measures 3) sources of pesticide exposure and 4) consistency across studies.

First, she underestimates the effects of small IQ shifts. She states that the pesticides are "slightly" harmful and that the 1 to 7 point drop in IQ found in the studies was "small." For an individual, these changes may not be substantial. However, these are not trivial changes in a population where small shifts in the mean IQ can have substantial effects on the number of children who need special education.  

Imagine two populations of one million children each. One population has a mean IQ of 100. At the high and low ends of the distribution – called the tails – about 23,000 people at the high end would be called gifted (IQ greater than 130) and 23,000 people at the low end would be intellectually disabled (IQ less than 70). In the other population the mean IQ is 98 – now 16,000 would be gifted and 31,000 would be intellectually disabled. Even if pesticide exposure is only associated with a two-point drop in IQ, that population would have thousands fewer gifted and highly productive children and thousands more intellectually disabled children who require additional resources for education. These "slight" decreases in IQ can have profound effects on the education and livelihoods of children.

Second, even though Cavallos points out that the IQ differences "only appeared by looking at the data in a certain way" the methods used to measure human exposure to pesticides are valid and some of the best available to scientists. The Berkeley researchers measured pesticide concentrations twice during pregnancy: once in the first half and once in the second half. The researchers found a consistent pattern regardless of whether they used the individual measurements or averaged the two – higher pesticide concentrations were associated with lower IQ. In fact, the average loss of IQ was greater when the concentrations were averaged.

Third, there are many potential ways to be exposed to pesticides, beyond those related to diet, which the article seems to highlight.  One possible route is eating fruits and vegetables.  But Salinas is a large agricultural community where pesticides are commonly used. Even non-farm workers can be exposed outside through the air (drift when spraying) or inside through contaminated dust (carried in on clothes or pets). In contrast, agriculture would not be the main route in New York City. Exposure to these pesticides likely comes from current diet or historical indoor use. Certain pesticides  – including the organophosphates found in the studies – are now banned from residential use, though they were used as recently as 10 years ago in many homes, including inner city lower income housing. Previous studies by these researchers in New York City reported detectable levels in homes, years after the ban took effect. Regardless of the manner of exposure, these pesticides appear to be pervasive and measurable with a demonstrable effect. 

Fourth, when trying to assess the validity of any scientific finding, researchers seek to replicate their findings. These three studies – which were  conducted on different populations of women in different locations – have remarkably consistent results. In addition, the authors of the three papers measured IQ in different ways and each used techniques that assessed different aspects of intelligence and cognitive abilities.  Despite these differences,  a consistent relationship between higher pesticide exposure and lower IQ was observed in each study, giving a higher level of validity to this relationship.

These issues are important when interpreting studies in humans. These four points highlight crucial information that journalists can use to further the public's understanding and put in perspective research methods and findings.  

The above work by Environmental Health News is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.
Based on a work at www.environmentalhealthnews.org.