Higher levels of fluoride in drinking water appear to be associated with an increased risk for hypothyroidism in a new study from England, raising concerns about the validity of community fluoridation of water as a safe public-health measure.

In particular, when a comparison was drawn between the West Midlands, a completely fluoridated area (0.7 mg/L or more) and Greater Manchester, a nonfluoridated area (0.3 mg/L or less), nearly twice the risk for hypothyroidism was detected in the West Midlands.

The report, published online February 24 in the Journal of Epidemiology and Community Health, describes the first population-level study investigating the association between fluoride levels in drinking water and hypothyroidism.

Lead investigator Prof Stephen Peckham, from the University of Kent, United Kingdom, told Medscape Medical News: “I think the results clearly demonstrate an increased risk of hypothyroidism associated with areas of [high] water fluoridation.” The study also “raises questions about the safety of community fluoridation,” he and his colleagues add.

Regarding the difference in risk between the West Midlands and Greater Manchester, Prof Peckham said, “Comparing these two similar urban areas does seem to demonstrate a significant increased risk of having a higher level of hypothyroidism.”

Commenting, Philippe Grandjean, MD, from the Harvard School of Public Health, Cambridge, Massachusetts, said:”The findings are meaningful and plausible and deserve to be taken seriously.”

Dr Grandjean and others interviewed by Medscape Medical News suggest that these results should prompt a rethinking of the whole practice of adding fluoride to drinking water in all countries.

But in contrast, the British Dental Association (BDA) expressed doubts about the new study.

“No definitive conclusions can be drawn about cause and effect from the data,” said Damien Walmsley, PhD, from the University of Birmingham, United Kingdom, an advisor to the BDA, adding, “We mustn’t forget that fluoridation has a proven benefit in reducing high rates of tooth decay, particularly among those living in the most deprived communities.”

First Populationwide Data

Prior to this study, a report in 2006 by the US National Research Council highlighted the potential causal relationship between fluoride and hypothyroidism as an area for future investigation.

And other than a 2014 Public Health England (PHE) report — which concluded that water fluoridation was a safe public-health measure but did not analyze data on hypothyroidism — there has been little study of the association and nothing on a populationwide scale. Other literature suggested further research was needed on the effects of children ingesting fluoride.

Against this background, the British researchers have now examined whether there is any difference in hypothyroidism prevalence between practices serving areas where water is fluoridated compared with areas not fluoridated.

The new data were drawn from 2012 Drinking Water Inspectorate figures on fluoride levels in England and diagnosed hypothyroidism prevalence data from the 2012–2013 Quality and Outcomes Framework (a general practice-based pay-per-performance system), as well as 2013 GP-registered patient numbers and 2012 practice-level Index of Multiple Deprivation scores.

A total of 7935 general practices covering around 98% of the English population provided data, and fluoride levels across England were mapped against these catchment areas. Associations and trends were noted, and an analysis carried out using binary logistic regression models.

“Data collection in the UK has improved over recent years and shows that the prevalence of hypothyroidism has become static, with a mean of 3.2% in the general population,” said Prof Peckham, who emphasized the strength of this study.

There was also minimal variation in recording of hypothyroidism between practices, he emphasized to Medscape Medical News. “We believe nearly 100% of hypothyroidism is diagnosed, and we used this raw, unadjusted prevalence data in our analysis. Given the size of this study, any variation was evened out.”

In England, approximately 10% of the population receives fluoridated water at a target level of 1 mg/L (1 ppm), with some areas having high natural fluoride levels and some having fluoride added artificially. Prof Peckham added that while there are other dietary sources of fluoride such as that found in tea, drinking water is the most significant source of ingested fluorides in the United Kingdom.

The analysis aimed to determine whether prevalence of hypothyroidism was affected by practice populations being situated in fluoridated areas, defined as 0.7 mg/L or more, or in those areas with lower levels of fluoride, 0.3 mg/L or less.

Commenting on the methodology from an expert risk assessor’s perspective, Kathleen Thiessen, PhD, from the Oak Ridge Center for Risk Analysis, Tennessee, said that the large size of the study population minimized bias and the researchers accounted for the major known predictors of thyroid function, including age, sex, iodine intake, and perchlorate exposure.

High Fluoridation Found to Increase Risk for Hypothyroidism

The mean prevalence of hypothyroidism for England was 3.18%.

The odds of a practice recording high levels of hypothyroidism was 1.4 times higher (odds ratio [OR], 1.37) in areas with maximum fluoride of greater than 0.3 mg/L but 0.7 mg/L or less and 1.6 times higher in areas with maximum fluoride in excess of 0.7 mg/L (OR, 1.62) than it was for practices in areas with maximum fluoride 0.3 mg/L or less.

When a comparison was drawn between the West Midlands, where fluoridation is high (0.7 mg/L or more), and Greater Manchester, where fluoridation is low (0.3 mg/L or less) the adjusted OR was 1.935 for the West Midlands.

“On the whole, we think there’s a 9% elevated or excess hypothyroidism prevalence across England due to fluoridation of drinking water,” Prof Peckham said.

Similar Levels of Fluoridation in the United Kingdom and United States

Levels of fluoridation in the United Kingdom are comparable to those in the United States. In 2011, there was a proposed recommendation to decrease fluoride levels in drinking water to 0.7 mg/L in the United States, but according to Dr Thiessen, the recommendation of 0.7 mg/L is still only a proposal and very few US states or municipalities have made any changes.

“The prior recommendation of a fluoridation range from 0.7 to 1.2 mg/L is still the operative one, with most US states targeting a fluoride level of about 1 mg/L, the same as in the UK. If a similar study were to be conducted in the US, I would expect to see similar results,” she said.

Prof Peckham said, “Based on our data, the 1-mg/L cutoff point for fluoride in drinking water does seem too high. Internationally, most countries are dropping their level to 0.7 mg/L, for example in Ireland and Canada.”

Dr Thiessen also made the point that the results were based on diagnosed hypothyroidism, and undiagnosed or subclinical hypothyroidism is also a concern. “Subclinical hypothyroidism carries an increased risk of progression to overt hypothyroidism as well as other health risks,” she observed.

However, speaking in his capacity as scientific adviser for the BDA, Dr Walmsley said, “The current level [of fluoridation] fits in with the recommended range that is considered by many organizations (for example, WHO, UK Medical Research Council) to be safe.”

He also pointed out that he was surprised that the authors were allowed to draw such sweeping conclusions in the paper, given that cause and effect were not established from the data.

“There may be other factors at play, particularly iodine levels, which would have to be taken into account before any certain scientific conclusions could be drawn.”

Confounding Factors, Including Iodine

Hypothyroidism is mainly found in women over 40 years of age. The researchers adjusted for gender and age and decided that looking only at people over the age of 40 provided the best fit. They also adjusted for deprivation, which is usually a result of nutritional deprivation, although this applies less in England than in other countries, they note.

Other confounding factors that affect the development of hypothyroidism are iodine deficiency and perchlorate contamination of drinking water. Perchlorate levels in the United Kingdom were found to be so low as to be negligible in this study, and according to Prof Peckham, studies on iodine deficiency are few, with only one recent study showing low levels in teenage women. However, background research showed that most iodine is sourced in the diet, and one recent study showed no regional differences in iodine levels in milk.

Reflecting on iodine as a confounding factor, Prof Peckham said: “Unfortunately, we could not adjust for this because the data are not available. But our study was in people aged 40-plus, so the deficiency in young people does not apply.”

Dr Grandjean said this is the largest epidemiological study so far focusing on adverse effects on the thyroid associated with elevated exposure to fluoride.

“Although the exposures and effects were assessed for groups of subjects rather than individuals, the data appear to be unbiased, and confounding seems to be limited,” he observed.

Rethinking of Water Fluoridation Needed?

Reflecting on the significance of the findings in relation to nationwide water-fluoridation schemes, Dr Thiessen stressed that there is certainly a cause for concern.

“This report, as well as other reports on the adverse effects of fluoride exposures and the lack of significant dental benefit from ingested fluoride should prompt a rethink of the whole practice of adding fluoride to drinking water.”

Consideration should be given to reducing all sources of fluoride in the environment, “with cessation of water fluoridation being an obvious place to start,” she commented.

Dr Grandjean remarked that much evidence already supports a link between fluoride exposure and thyroid deficiency and noted that in the United States it has been suggested that fluoridation should aim for 0.7 mg/L because the dental benefits appear to be no greater at levels above 0.7 mg/L.

“This decrease in fluoride levels would be an appropriate change to institute right away, as it would also help limiting any risks. Still, the new study suggests that risks may occur even below this level, and a revised risk/benefit assessment would therefore seem indicated.”

In addition to the risk for hypothyroidism, Dr Grandjean notes that thyroid deficiency during pregnancy can have serious implications for the developing fetal brain. “While thyroid deficiency can be treated, IQ loss or more serious damage to brain development is permanent.”

Prof Peckham was involved in a campaign in Southampton, United Kingdom, to prevent the fluoridation of drinking water supplies; the coauthors report no relevant financial relationships. Dr Thiessen and Dr Grandjean have declared no relevant financial relationships. Dr Walmsley has conducted research for a number of companies over the years related to tooth brushing and has received funding from major manufacturers.

J Epidemiol Community Health. Published online February 24, 2015. Abstract