Water Fluoridation & IQ Reduction: Statistical vs. Clinical Significance

This really depends on what you mean by significance.

Living in Washington, I have watched the political maneuvering in Portland, OR recently over the decision on fluoridation of the public water supply.  In Washington this topic has, over the 25 years that I’ve lived here, been among the more divisive and controversial topics.  If you want to get educated and otherwise rational adults into a heated and emotional argument, bring up the topic of water fluoridation.  In a discussion at home tonight my son urged me to look into this and consider a post on the topic.  He had heard that IQs were “significantly reduced” in communities with water fluoridation. He suspected that the IQ difference was huge, maybe 20 IQ points.  After all, “27 studies” show lower IQs in children from communities with water fluoridation.  I was pretty certain this could not be the case, so I went to the source of this claim.

A study published online in Environmental Health Perspectives in Oct 2012 seems to be the source of the claim that fluoridation of water supplies “significantly” reduces IQs.  This study was a meta-analysis of published studies both in the English language literature and the Chinese literature.  27 studies were included in the meta-analysis, of which a negative association between fluoridation levels and IQ was in fact noted in 26 of the 27.  The IQ difference in the 26 studies ranges from -0.95 to -0.1 absolute IQ  points difference.  In the pooled data of the meta-analysis the IQ difference associated with water fluoridation was -.045.  This difference was deemed statistically significant.  Statistical significance means that the likelihood that the findings occur by change is low.  Usually a chance of less than 5% that the findings are due to random variability makes the findings statistically significant.  Statistical significance does not mean that there is practical or clinical significance to the findings.

In the case of fluoridation of water the difference in IQ between populations where water fluoride levels were high  compared to populations where fluoride levels were low was “statistically significant” although whether it is clinically significant is much less clear.  Is there a real difference in a population where the mean IQ is 100.2 vs a population where the mean IQ is 99.8?  I have no idea.  The authors of this study did not suggest that fluoridation of water be stopped or that the net value of water fluoridation is negative.  Their conclusion states,

“Conclusions: The results support the possibility of an adverse effect of high fluoride exposure on children’s neurodevelopment. Future research should include detailed individual-level information on prenatal exposure, neurobehavioral performance, and covariates for adjustment.”

This is a long way from the headlines and conclusions of the anti-fluoridation activists.  Is a headline shouting, “Harvard Study Confirms Fluoride Reduces IQ” and the use of Significant IQ Reduction seem prevalent?  I’m all for further research into any and all effects of water fluoridation. There is good evidence that water fluoridation reduces tooth decay by about 25%.  Is this significant?  I’ll let readers decide.  Is an IQ reduction of 0.4 significant?  Again, decide for yourself, but understand the difference between statistical significance and practical or clinical significance.

Leave a reply