One effect that has been claimed for fluoridation of drinking water that I didn’t really examine in my previous examination of fluoride toxicity is reduced IQ. The major regulatory reviews have previously concluded there is no evidence for any such link at relevant concentrations, but a new study published ahead of print in Environmental Health Perspectives concluded that “higher fluoride exposure may affect intelligence among children.” Does the study really support an effect on intelligence, and if so what are its implications for water fluoridation?
The authors studied two villages in China. One of these villages, Wamiao, was considered to be in a severe endemic fluorosis area, with a mean level of fluoride in drinking water of 2.47 mg/L (range 0.57 to 4.50 mg/L). The other, Xinhuai, had a mean fluoride level of 0.36 mg/L (range 0.18 to 0.76 mg/L). Children from 8 to 13 years of age attending the village primary schools (222 students from Wamiao and 290 from Xinhuai) were included in the study.
The authors found that serum fluoride levels were correlated with fluoride concentrations in drinking water, which is not particularly surprising. They also found a statistically significant difference in IQ scores between the two villages, with children in Wamiao having on average a lower IQ. IQ was found to be associated with serum fluoride in Wamiao but not in Xinhuai.
There are some significant limitations to this study, however. The relationship between serum fluoride and IQ in Wamiao isn’t particularly strong – barely statistically significant, and with an r2 value of only 0.0266 (indicating that serum fluoride accounts for only a very small amount of the overall variability in IQ. Furthermore, the only analysis undertaken was a linear regression, but the data do not appear to indicate a linear trend. When I look at the plot of serum fluoride vs. IQ in the paper (Figure 3 – only available as a PDF right now so I haven’t imported it here), if there is a trend it appears to be driven by a few low IQ scores associated with very high serum fluoride levels (which in turn appear to be associated with concentrations in water around 4 mg/L or higher); I suspect if you removed the very high exposures the data would not show a noticeable trend.
The difference in IQ between the two villages could have many other causes. The authors ruled out differences in urinary iodine or blood lead levels, and both villages had similar average family income and average education level. However, the villages are characterized by “lack of communication with the outside world,” which in turn implies that they are likely genetically distinct populations – that alone could account for the differences in IQ. Narrowing the cause of an IQ difference down to a single factor would require the study of more than 2 villages. To relate the results to municipal fluoridation policies, further studies would preferably also include villages with average fluoride concentrations in water approximating 1 mg/L.
Overall the data do not, in my opinion, suggest an effect of fluoride on IQ at anything close to the fluoridation levels used in North American water supplies. There is some indication of a possible effect at very high concentrations (around 4 mg/L or higher), however.
Xiang, Q., Liang, Y., & Chen, B. (2010). Serum Fluoride Level and Children’s Intelligence Quotient in Two Villages in China Environmental Health Perspectives DOI: 10.1289/ehp.1003171