I have discussed the paper by Grandjean & Landrigan (2014) before in the article Repeating bad science on fluoride. So have other bloggers and commenters. On the other hand anti-fluoride propagandists are still promoting it heavily in social media and “natural” health web sites.
But the cogs of science publishing have ground slowly on and the scientific critiques are starting to appear. The last issue of The Lancet (which published the original paper) has several articles in the correspondence section critical of the paper (see Neurodevelopmental toxicity: still more questions than answers). Two of these related to claims made about fluoride.
Unsubstantiated claims, misquoted studies
Virginia Feldman wrote that the authors:
“make unsubstantiated claims and misquote previous studies to pull together heterogeneous elements and drugs into a group of substances termed neurotoxicants.
The investigators’ claim of new data is undermined by 24 of the cited references being their own previous, mostly review articles . . . . . . Their “strong evidence” for adding fluoride was a finding from Grandjean’s own review of older Chinese studies. Contrary to their statement—“Confounding from other substances seemed unlikely in most of these studies”—findings from many previous meta-analyses have shown the faults of using intelligence quotient (IQ) data from countries with highly polluted air and water; non-validated IQ tests; poor controls for parent IQ, socioeconomics, and other variables; and studying mega-doses in animals and in human beings. By contrast with this review of Chinese studies, all of problematic methodological robustness, more than 3000 studies of the safety of water fluoridation stretch over 65 years. During this time, as fluoridation increased from 0% to 72% of US households, average US IQs have not decreased, but have instead increased by 15 points.”
She also directs the authors’ attention to a paper by one of their frequent co-authors, David Bellinger, about determining IQ points lost (Grandjean & Landrigan 2014 claimed fluoride can cause a 7 point drop in IQ). Bellinger specifically says the meta-analyses of the sort used by Grandjean & Landrigan (2014) are just not suitable for calculating IQ points lost.
Julianna Gelinas and Myron Allukian were also concerned about the authors’ reliance on very flimsy evidence. The “claim that fluoride might cause neurodevelopmental harm” is “based on only one paper, of which Grandjean is a coauthor.” Further describing the limitations of that study they say “it contains several flaws that undermine its credibility and calls into question its applicability to the community water fluoridation programme in the USA.”
“The study is a meta-analysis of 27 cross-sectional studies done in poor, rural communities in China, Mongolia, and Iran, countries where the drinking water contains high levels of naturally occurring fluoride. The 27 original studies did not adequately control for a variety of intervening and confounding variables that could have affected intelligence quotient (IQ) scores, such as parents’ education and socioeconomic status and air and water pollution. It is unfortunate that Grandjean and Landrigan did not mention these limitations.
Additionally, they did not clearly state that the reference groups in their article use water fluoridated at about the recommended level. Thus, another interpretation of their analysis could be that communities fluoridated at the recommended level have a higher IQ.”
Nor did Grandjean and Landrigan acknowledge research showing fluoride was not harmful:
“No credible scientific studies show a relation between fluoride consumption and IQ levels; however, several have shown that fluoride ingested at recommended levels is not harmful. Grandjean and Landrigan did not acknowledge the animal study that showed no evidence of a neurotoxic effect of fluoride, even at levels up to 230 times the recommended concentration; an earlier study showing that fluoride causes no harm to children; two formal reviews that delineate weaknesses in the Chinese fluoride and IQ studies; and the conclusion by one of these sets of investigators that biological plausibility for a link between fluoridated water and IQ has not been established.”
Request for statement on fluoridation
They are also concerned at the way the original paper is being used by anti-fluoride propagandists:
Unfortunately, Grandjean and Landrigan’s Review has been aggressively and improperly used by antifluoridationists to frighten the public about the effects of fluoridation, a well-established public health measure that has been shown to be cost-effective and safe. As a result, the public’s oral health, especially that of the most vulnerable people, is put in jeopardy. . . .
A statement from Grandjean and Landrigan clearly stating that their addition of fluoride to their list of neurotoxins does not apply to fluoridation at the recommended levels of 0·7—1·2 ppm would clarify our concerns on the misuse and misinterpretation of their paper.”
The authors’ response to these criticisms was brief and rather flippant. They claim other reviewers of their meta-analysis were “without access to important background information” – without providing information on this. They dismissed Feldman’s reference to the increase in population mean IQ as a “serious error” – again without justification or explanation.
The remained silent about the way they had relied on very few sources, often their own, in their review.
So some important criticisms of the original paper and ones that the authors did not respond to properly. But will this stop anti-fluoride propagandists relying on this paper – one that very few of them have bothered reading?
I doubt it.