Many countries add fluoride to the drinking water in the belief this will improve dental health. As with anything else public health, some people think this is a grave mistake. So on the positive side, it seems to help with the intended target, a good start for any social policy. Wikipedia’s summary is kinda what you would expect:
Reviews have shown that water fluoridation reduces cavities in children. A conclusion for the efficacy in adults is less clear with some reviews finding benefit and others not. Studies in the U.S. in the 1950s and 1960s showed that water fluoridation reduced childhood cavities by fifty to sixty percent, while studies in 1989 and 1990 showed lower reductions (40% and 18% respectively), likely due to increasing use of fluoride from other sources, notably toothpaste, and also the ‘halo effect’ of food and drink that is made in fluoridated areas and consumed in unfluoridated ones.
A 2000 UK systematic review (York) found that water fluoridation was associated with a decreased proportion of children with cavities of 15% and with a decrease in decayed, missing, and filled primary teeth (average decreases was 2.25 teeth). The review found that the evidence was of moderate quality: few studies attempted to reduce observer bias, control for confounding factors, report variance measures, or use appropriate analysis. Although no major differences between natural and artificial fluoridation were apparent, the evidence was inadequate for a conclusion about any differences. A 2007 Australian systematic review used the same inclusion criteria as York’s, plus one additional study. This did not affect the York conclusions. A 2011 European Commission systematic review based its efficacy on York’s review conclusion. A 2015 Cochrane systematic review estimated a reduction in cavities when water fluoridation was used by children who had no access to other sources of fluoride to be 35% in baby teeth and 26% in permanent teeth. The evidence was of poor quality. A 2020 study in the Journal of Political Economy found that water fluoridation significantly improved dental health and labor market outcomes, but had non-significant effects on cognitive ability.
Fluoride may also prevent cavities in adults of all ages. A 2007 meta-analysis by CDC researchers found that water fluoridation prevented an estimated 27% of cavities in adults, about the same fraction as prevented by exposure to any delivery method of fluoride (29% average). A 2011 European Commission review found that the benefits of water fluoridation for adult in terms of reductions in decay are limited. A 2015 Cochrane review found no conclusive research regarding the effectiveness of water fluoridation in adults. A 2016 review found variable quality evidence that, overall, stopping of community water fluoridation programs was typically followed by an increase in cavities.
Most countries in Europe have experienced substantial declines in cavities without the use of water fluoridation due to the introduction of fluoridated toothpaste and the large use of other fluoride-containing products, including mouthrinse, dietary supplements, and professionally applied or prescribed gel, foam, or varnish. For example, in Finland and Germany, tooth decay rates remained stable or continued to decline after water fluoridation stopped in communities with widespread fluoride exposure from other sources. Fluoridation is however still clearly necessary in the U.S. because unlike most European countries, the U.S. does not have school-based dental care, many children do not visit a dentist regularly, and for many U.S. children water fluoridation is the prime source of exposure to fluoride. The effectiveness of water fluoridation can vary according to circumstances such as whether preventive dental care is free to all children.
A nice form of evidence is natural experiments. Since fluoride content of drinking water varies for natural reasons related to sediment, this can be exploited to study effects on dental health. Denmark has extensive natural variation:
Compare this to a map of socioeconomic status of the areas. This very detailed map gives mean income for working adults by parish:
It is somewhat hard to read with this coloration, if we zoom to commune level:
Main point to note is that the southeast islands (Lolland, Falster, and Møn) are pretty poor but high in fluoride, and west Jutland is also poor and low in fluoride. Anyway, so since Nordic countries have great register data, we can link the tooth decay data from the public health system, which e.g. these people did, to figure out the protective effect:
So overall, reasonable enough. What are the other people saying? Surprisingly maybe, we can find Richard Lynn there!
Cheng, H., & Lynn, R. (2013). The adverse effect of fluoride on children’s intelligence: a systematic review. Mankind Quarterly, 53(3/4), 306.
The authors conducted a systematic review on published studies to date to investigate the effect of fluoride exposure on children’s intelligence quotient (IQ) scores. PSYCHINFO, Web of Science, MEDLINE, SCI, and the China National Knowledge Infrastructure (CNKI) search engines were employed for all documents published up to 2012, in English and in Chinese. In total, 38 studies of fluoride exposure and children’s scores on IQ tests were included in this review. The weighted mean effect size (WMES) on children’s IQ scores between higher and lower regions of fluoride exposure was -.46 (CI 95% -.57 to -.35; p<.001) equivalent to 6.9 IQ points. Sensitivity analyses showed that after excluding studies that had other elements contaminations, the adverse effects of fluoride exposure on children’s intelligence remained to be significant. Further, six of the studies reported significant negative correlations between fluoride in the body and intelligence. The evidence suggests that fluoride in drinking water is a serious public health hazard.
And it isn’t just him, as we can easily find out:
Tang, Q. Q., Du, J., Ma, H. H., Jiang, S. J., & Zhou, X. J. (2008). Fluoride and children’s intelligence: a meta-analysis. Biological trace element research, 126(1), 115-120.
This paper presents a systematic review of the literature concerning fluoride that was carried out to investigate whether fluoride exposure increases the risk of low intelligence quotient (IQ) in China over the past 20 years. MEDLINE, SCI, and CNKI search were organized for all documents published, in English and Chinese, between 1988 and 2008 using the following keywords: fluorosis, fluoride, intelligence, and IQ. Further search was undertaken in the website www.fluorideresearch.org because this is a professional website concerning research on fluoride. Sixteen case–control studies that assessed the development of low IQ in children who had been exposed to fluoride earlier in their life were included in this review. A qualitative review of the studies found a consistent and strong association between the exposure to fluoride and low IQ. The meta-analyses of the case–control studies estimated that the odds ratio of IQ in endemic fluoride areas compared with nonfluoride areas or slight fluoride areas. The summarized weighted mean difference is −4.97 (95%confidence interval [CI] = −5.58 to −4.36; p < 0.01) using a fixed-effect model and −5.03 (95%CI = −6.51 to 3.55; p < 0.01) using a random-effect model, which means that children who live in a fluorosis area have five times higher odds of developing low IQ than those who live in a nonfluorosis area or a slight fluorosis area.
Choi, A. L., Sun, G., Zhang, Y., & Grandjean, P. (2012). Developmental fluoride neurotoxicity: a systematic review and meta-analysis. Environmental health perspectives, 120(10), 1362-1368.
Background: Although fluoride may cause neurotoxicity in animal models and acute fluoride poisoning causes neurotoxicity in adults, very little is known of its effects on children’s neurodevelopment.
Objective: We performed a systematic review and meta-analysis of published studies to investigate the effects of increased fluoride exposure and delayed neurobehavioral development.
Methods: We searched the MEDLINE, EMBASE, Water Resources Abstracts, and TOXNET databases through 2011 for eligible studies. We also searched the China National Knowledge Infrastructure (CNKI) database, because many studies on fluoride neurotoxicity have been published in Chinese journals only. In total, we identified 27 eligible epidemiological studies with high and reference exposures, end points of IQ scores, or related cognitive function measures with means and variances for the two exposure groups. Using random-effects models, we estimated the standardized mean difference between exposed and reference groups across all studies. We conducted sensitivity analyses restricted to studies using the same outcome assessment and having drinking-water fluoride as the only exposure. We performed the Cochran test for heterogeneity between studies, Begg’s funnel plot, and Egger test to assess publication bias, and conducted meta-regressions to explore sources of variation in mean differences among the studies.
Results: The standardized weighted mean difference in IQ score between exposed and reference populations was –0.45 (95% confidence interval: –0.56, –0.35) using a random-effects model. Thus, children in high-fluoride areas had significantly lower IQ scores than those who lived in low-fluoride areas. Subgroup and sensitivity analyses also indicated inverse associations, although the substantial heterogeneity did not appear to decrease.
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.
If this effect is for real, it could be a major contributing factor to anti-Flynn effects. So… is it? Meta-analysis is a biased method when publication bias is present. And regional studies can have uncontrolled confounds. Better would be to track people over their entire life and know what water they drink and then test them later on a good intelligence test… which… in fact… we can do with the Nordic data. Not only can we do this, it has been done, aaaand:
- Aggeborn, L., & Öhman, M. (2021). The effects of fluoride in drinking water. Journal of Political Economy, 129(2), 465-491.
Water fluoridation is a common but debated public policy. In this paper, we use Swedish registry data to study the causal effects of fluoride in drinking water. We exploit exogenous variation in natural fluoride stemming from variation in geological characteristics at water sources to identify its effects. First, we reconfirm the long-established positive effect of fluoride on dental health. Second, we estimate a zero effect on cognitive ability in contrast to several recent debated epidemiological studies. Third, fluoride is furthermore found to increase labor income. This effect is foremost driven by individuals from a lower socioeconomic background.
I also don’t know what this has to do with political economy, but that aside, this study trumps the prior ones in rigor. Swedish maps of fluoride are pretty similar to the Danish ones:
They first confirm the dental health effects:
And then for intelligence, measured at age 18 on the army draft test:
Column 1 displays the unconditional treatment effect. In columns 2 and 3, we add fixed effects for cohort and municipality of birth. We then include parental covariates, which results in a reduced sample since we have data on fathers’ cognitive ability only from 1969 and onward. To make the samples comparable with and without these covariates, we run column 4 for the same sample as in column 5. We also run two subsample analyses: in column 6, we run the analysis for those who have lived in the same SAMS in a municipality for the entire period from age 0 to 18, and in column 7 we restrict the sample to those who have moved only within a municipality.8
Looking at the estimates, they are very small and often not statistically significantly different from zero. Sometimes the estimates are negative and sometimes positive, but they are always close to zero. If we take the largest negative point estimates (−0.0047, col. 1) and the largest standard error for that specification (0.0045), the 95% confidence interval would be −0.014 to 0.004. We may thus rule out negative effects larger than 0.14 standard deviations in cognitive ability if fluoride is increased by 1 milligram/liter (the level often considered when artificially fluoridating the water).
However, the effect of fluoride may not be linear. We have therefore run several specifications addressing nonlinearities, and the results are presented in the appendix. Figure A1 displays the effect for each 0.1 milligram/liter of fluoride, table A5 present results for quartile regressions, table A6 is a dose response analysis, and table A7 is an analysis where we have restricted the sample to 1 milligram/liter or higher. Figure A2 is a spline regression where we have predicted cognitive ability on a set of background characteristics. We then use the ranked predicted values to run regressions with fluoride as the independent variable in a flexible interaction model, where fluoride is interacted with a vector of cubic splines. The spline specification picks up nonlinear treatment heterogeneity over the predicted cognitive ability distribution. All in all, we conclude that fluoride does not have an effect on cognitive ability in these nonlinear specifications.
We have furthermore run analyses for noncognitive ability, math test scores, and health, which are presented in section B5 in the appendix. This analysis further strengthens our conclusion that fluoride does not have a negative impact on human capital development.9
And then they continue with actual life outcome data:
Which they explain too:
We now continue with the long-term outcome of annual labor income in 2014 for individuals born between 1985 and 1992. Given our results for cognitive ability, we do not expect negative effects of fluoride. However, positive effects are possible given the results found for dental health.
The results are presented in Table 5. The point estimates are often statistically significant, and the coefficients are always positive. Taking column 6 as an example, where all covariates and fixed effects are included, we find that the point estimate equals 0.0044, meaning that income increases by 4.4% if fluoride is increased by 1 milligram/liter.10 These reduced form estimates may be compared with Glied and Neidell (2010), who, by using American data, found that women who drink fluoridated water have on average 4% higher earnings.11 Our estimated effect on income may also be compared with estimated education premiums. The return of one additional year of education yields an increase in income by 6%–10%, according to the instrumental variable estimates in the review in Card (1999). An increase in fluoride by 1 milligram/liter would thus yield a similar increase as roughly half a year of additional education. Nonlinear specifications are presented in figure A1 and tables A8–A10, which overall supports the findings presented here. In section B5 in the appendix, we present the result for employment status (another margin for labor market status), and we find that fluoride has a positive effect.
We have run several robustness checks for our main outcomes, which are presented and discussed in section B6 in the appendix. These include (1) analyses with older cohorts for income, (2) sensitivity tests to the mapping of the water data, (3) alternative income measures, (4) included interacted fixed effects, (5) an intention-to-treat model, (6) analyses using time variation in fluoride, (7) in utero treatment effects, (8) secondary dentition treatment analyses, (9) analyses including SAMS covariates, (10) specifications for various forms of family robustness, and (11) analyses including covariates for other water characteristics. All in all, after considering these robustness results, we remain with our conclusions presented here that fluoride improves dental health, that fluoride does not affect cognitive ability, and that fluoride has a positive effect on annual labor income. These robustness checks are numerous, and most of them are in line with the results presented here, but some specifications do not go in the expected direction. For a more detailed discussion, see the appendix.
All in all, this is pretty close to being the definitive study to wish for.