To improve the public’s health, it is essential to collect and analyze data. Scientists often call this “surveillance data.” This information is important, but I get annoyed when surveillance data are misused. This is the case in recent articles concerning fluoride intake and IQ, and this misuse is causing harm.
Studies that allege an association between water fluoridation and decline in IQ and that have been cited by Robert F. Kennedy, the head of the U.S. Department of Health and Human Services, have measured fluoride incorrectly and have unnecessarily led to water fluoridation cessation in many U.S. cities and in Florida and Utah. Let me summarize my background and views on this topic.
As Professor Emeritus of Newcastle University, United Kingdom, and past president of U.K. societies of pediatric dentistry and dental public health, I have a strong interest in data surveillance of oral health. I ask questions such as, “Are the teeth of 5-year-old children getting better or worse?” and “What is the relation between community water fluoridation and the dental health of children on a national basis?”
My main research interests have been fluoride, nutrition and diet. I was a member of the U.K.’s Scientific Advisory Committee on Nutrition (SACN) for five years. As the first dentist on that committee, I had a particular interest in improving oral health by improving diet.
In 2023, I was humbled to be awarded the International Association for Dental Research Gold Medal for “significant positive impact on global oral health through your scholarly activity on the links between oral health, nutrition, and fluoride” Regarding fluoride, my research over decades has spanned laboratory science, population surveys and reviews.
This work and wider evidence leaves me with no doubt that fluoride has been instrumental in improving oral health throughout the world over the past half-century. Yet, in the past year, dozens of municipalities in the U.S. and some in Canada, including Montreal, are considering removing fluoride or have already begun the process. Fortunately, Calgary, which removed fluoride from its drinking water more than a decade ago, is bucking the trend and will return to community water fluoridation June 30.
Appropriate provision of fluoride is recommended by scientific and health authorities worldwide. “Appropriate” is a key word, because ingestion of too much fluoride is bad for health, just as too much salt, vitamin A or even water is bad for health.
Monitoring fluoride intake is, therefore, sensible. It is possible to measure fluoride intake but doing so is difficult and expensive because food and drink intake must be carefully recorded and the fluoride content determined. Consequently, an alternative method was investigated: measuring excretion of fluoride in urine, given that most of ingested fluoride leaves the body in urine.
The World Health Organization (WHO) undertook the task of determining whether we can measure fluoride intake by measuring fluoride in urine. The WHO publishes advisory and guidance documents including in the oral health field – one of the topics is urinary fluoride monitoring. I was a member of the panel of nine nutritionists and dental scientists who prepared the latest 83-page manual on “Basic methods for assessment of renal fluoride excretion in community prevention programmes for oral health”.
This WHO work shows that many people attack water fluoridation using flawed arguments. They say that fluoridation is associated with harm to IQ scores, but they do so by relying on studies that use an invalid method of measuring individual long-term fluoride intake or exposure: spot urine sample.
For many years, experts have stated that spot urine is an invalid measure of individual contemporary or chronic fluoride intake or exposure. In fact, the WHO made precisely this point in its 2014 manual setting forth basic methods to measure fluoride exposure. It is important to summarize the WHO advisory document.
First, the opening paragraph of the WHO advisory manual in Section 1.5, page 9, states,
“Ingested fluoride from all sources, whether deliberately or unintentionally ingested, is excreted primarily in the urine. Thus, studies of urinary fluoride levels are ideal for assessing the intake of fluoride in populations.”
Note the word “populations” (not individuals). “Populations” is used again in the next section which covers “areas of uncertainty,” discussing the topics such as “influence of diet on urinary fluoride excretion,” “within-subject variation” and “lack of correlation between urinary fluoride excretion and fluoride intake.” In other words, differences in individual metabolism, diet, how much water they drink, or even how the studies were conducted are important limitations to keep in mind when planning studies and making sense of the results. While measuring fluoride in urine can help us understand general exposure levels in large groups, urine doesn’t accurately show how much fluoride any one person has consumed over the long term. Indeed, the WHO manual states,
“Urinary fluoride excretion is not suitable for predicting fluoride intake for individuals.”
That urine is not suitable to measure individual fluoride intake could not be more clearly stated.
That urine is not suitable to measure individual fluoride intake could not be more clearly stated. The WHO manual says, in effect, “Do not use urine to measure the fluoride intake of an individual person.” Yet, some researchers have done exactly what they have been advised not to do – they have used spot urine to measure individual fluoride intake. They have published analyses of urinary fluoride concentration and child’s IQ – purporting to analyze the individual’s fluoride intake against the individual’s IQ. There is too much intra-subject variation, both within-days and between-days, to allow valid analysis of fluoride in urine on an individual basis.
Consequently, when such researchers decry water fluoridation, they are doing so for invalid reasons. They say that fluoridation is associated with harm to IQ, but they have not provided a scientific basis for their claim. They rely on studies that use an invalid method of measuring individual long-term fluoride intake or exposure: spot urine sample. According to one study, more than 20 publications have used spot urinary fluoride to measure fluoride intake. Consequently, the studies are invalid.
To understand the scientific reasoning behind the WHO advice that urine cannot be used to measure individual fluoride intake or exposure, it is important to examine the standards articulated in the WHO advisory manual. The WHO advisory manual refers to urinary collections, and analysis of fluoride content, over a 24-hour period – in other words, a measure of urinary fluoride excretion during one day (24 hours). Collecting 24-hour urine is quite tedious and requires the subject’s cooperation, so there have been instances of collecting just “spot samples” of urine (a one-time void) rather than collecting over a whole day. This alternative was considered in the WHO advisory manual and rejected as being an invalid measure. The first paragraph in Chapter 8 states:
“Twenty-four-hour urine collection is the most reliable method for estimating urinary excretion of fluoride. However, when it is not feasible to obtain 24-hour urine or shorter time-controlled collections, spot urine samples may be obtained. A spot urine sample is defined as an un-timed “single-void” urine sample. This method is the least informative method [WHO emphasis] for studying fluoride exposure, because the amount of fluoride excreted per day cannot be calculated from the concentration alone.”
Using a single void sample is like seeking to know about a person’s diet by sampling only one meal. Fluoride concentration in a spot sample depends on many factors such as when teeth were brushed, given that some (fluoride) toothpaste is swallowed; whether tea was drunk recently (tea infusion contains about the same fluoride concentration, or more, as fluoridated water); or exercise was performed (which might concentrate the urine). Spot urine is not a measure of individual fluoride intake, yet the studies relating fluoride intake to child’s IQ on an individual basis all use spot urinary samples. This research practice is clearly contrary to WHO conclusions and recommendations.
Part of the evidence considered by the WHO advisory manual was a research article of which I was a co-author. It is an analysis of all data worldwide then available (19 studies) on the relationship between fluoride intake and urinary fluoride excretion. Although the language is rather technical, it is worth reproducing a section:
“… the ample 95 per cent PI [prediction interval] band associated with the regression lines does not allow the use of DUFE [daily urinary fluoride excretion] as a precise estimator of either TDFI [total daily fluoride intake] or DFR [daily fluoride retention] on an individual basis. However, the 95 per cent CI [confidence interval] bands are narrow enough to allow the estimator of average TDFI and DFR in children and adults from average DUFE values, i.e., on a community basis.”
Urine samples are unsuitable for measuring fluoride in individuals but 24-hour samples might be used to estimate fluoride intake in whole populations, as this review article states:
“While fluoride concentrations in plasma, saliva and urine have some ability to predict fluoride exposure, data are, at present, insufficient to recommend fluoride concentration in these body fluids as viable biomarkers of contemporary fluoride exposure for individuals. Fluoride concentration in urine can be considered a useful biomarker of contemporary fluoride exposure for groups of people, and normal values have been published.”
The message is clear: 24-hour urine collection, let alone a spot urine sample, is not a valid way of measuring the fluoride intake of an individual. Spot urine is even more hopeless when used to measure the fluoride intake or exposure of a fetus within an individual. Studies claiming that fluoride exposure affects children’s IQ, where fluoride exposure has been assessed using fluoride concentration in spot samples of urine, are based on unsound science as are 24-hour urine samples to assess individual chronic fluoride exposure.
To determine whether water fluoridation is linked to lower IQ in children, the best available method is to assess the IQ of children in communities with and without community water fluoridation during the developmental phase of the brain and to measure total intake indirectly through an assessment of dental fluorosis, as has been done in an Australian study that also takes possible confounding factors into account. Similar studies have been undertaken and no relation between IQ and community water fluoridation has been found. Consequently, sound science gives us no reason to worry that water fluoridation affects IQ.
This is a very important article because it states plainly that there is a huge problem in US official understandings of water fluoridation. A US government report and a US federal district court decision are incorrect because they have assumed, wrongly, that studies that use spot urine to measure chronic individual fluoride exposure are valid. But the studies are invalid.
The US Health and Human Services Secretary, the State of Florida and of Utah, and several US cities ought to revisit their decisions because of their false understandings. As Dr. Rugg-Gunn concludes, “sound science gives us no reason to worry that water fluoridation affects IQ.”