A new report linking fluoridated drinking water to lower IQ scores in children is sure to ratchet up the debate over a practice that’s considered one of the greatest public health achievements of the 20th century.

The report published Monday in JAMA Pediatrics synthesizes the results of dozens of research studies that have been released since 1989. Its overall conclusion is that the more fluoride a child is exposed to, the lower he or she tends to score on intelligence tests.

The analysis was conducted for the U.S. National Toxicology Program, and it has attracted a good deal of criticism over the many years of its development. Among the biggest critiques is that it’s based on data from places where fluoride levels are far higher than the concentration recommended by the U.S. Public Health Service.

Adding fluoride to community drinking water is credited with reducing the average number of teeth with cavities by 44% in adults and 58% in adolescents since the 1960s, the health service says. Yet even with the proliferation of fluoride-containing toothpastes and dental sealants, tooth decay is still the most common chronic disease affecting American children, and the average senior citizen is missing at least 10 permanent teeth.

About 209 million Americans receive fluoridated water in their taps, according to the Centers for Disease Control and Prevention. Robert F. Kennedy Jr., President-elect Donald Trump’s pick to lead the Department of Health and Human Services, has said he’d like to see that number fall to zero, in part due to concerns over “IQ loss.”

The JAMA Pediatrics report is based on work prepared for the National Toxicology Program’s 324-page monograph on fluoride and brain development, which was finalized in August. Here’s a closer look at what it does — and doesn’t — show.

Where do the data come from?

The report combines data from 74 studies on fluoride exposure and children’s IQ. The bulk of them — 45 — were conducted in China, and another 12 were from India. None were from the United States, although three were from Canada and four were from Mexico.

Ten of the studies were designed to follow groups of people over time to see how their differing levels of fluoride exposure affected IQ scores and other outcomes. The rest of the studies assessed a population’s fluoride exposure and IQ at the same time.

IQ scores were usually reported as averages for a group, though sometimes they reflected an individual’s specific level of fluoride exposure.

How much fluoride are we talking about?

Fluoride exposure was measured multiple ways.

Sometimes researchers measured the amount of fluoride in a community’s drinking water, and sometimes they measured the amount of fluoride in participants’ urine. Dental fluorosis — a condition that occurs when teeth get too much fluoride and appear to be stained — was also used to assess exposure. So were environmental factors, such as exposure to pollution from burning coal with a high fluoride content.

The studies were grouped into three categories: those in which exposure was less than 4 milligrams of fluoride per liter of water (the maximum concentration allowed in the U.S. by the Environmental Protection Agency); those in which exposure was less than 2 mg/L (the EPA’s non-enforceable secondary standard to prevent cosmetic problems in places where fluoride levels are naturally high); and those in which exposure was less than 1.5 mg/L (the guideline value set by the World Health Organization).

So what’s the link with IQ?

Of the 65 studies included in the primary analysis, 64 found an inverse relationship between fluoride exposure and children’s IQ — the higher (or lower) fluoride was, the lower (or higher) IQ scores were.

The researchers also pooled together data on nearly 21,000 children from 59 studies that reported average IQ scores. Those data showed that children exposed to higher fluoride levels had lower IQ scores than children exposed to lower fluoride levels.

In addition, the report authors combined data from 38 studies and crunched the numbers themselves to see whether there was an overall dose-response relationship between fluoride and IQ. Sure enough, they wrote, “lower children’s IQ scores were associated with increasing levels of fluoride exposure.”

This sounds bad. Should I be worried?

Not necessarily. The findings are only as strong as the data they’re based on, and the studies in this analysis have some issues.

For starters, 52 of the 74 studies were judged by the report authors to have a “high risk of bias.” That undermines the validity and reliability of their results.

Another issue is that most of the studies considered fluoride exposures far above the target level for the U.S. Since 2015, the Public Health Service has pegged the “optimal” concentration of fluoride at 0.7 mg/L, the equivalent of about 3 drops of fluoride in a 55-gallon barrel. (Prior to that, the target ranged from 0.7 to 1.2 mg/L.)

Only seven of the studies assessed children whose water contained less than 1.5 mg/L of fluoride. When they were considered on their own, there was no relationship between fluoride exposure and IQ.

Besides, the American Academy of Pediatrics has noted that assessing IQ in children is not a straightforward affair, since measurements can be skewed by “socioeconomic, physical, familial, cultural, genetic, nutritional, and environmental factors.” Comparing IQ scores from multiple studies in multiple countries as if they were the same only compounds the problem, the academy said.

Hmmm. What else should I know about this report?

Plenty. In fact, JAMA Pediatrics published an editorial by Dr. Steven M. Levy, a dental public health expert from the University of Iowa, to enumerate the reasons why the report shouldn’t be taken at face value.

Take the issue of bias. Of the 59 studies that comprised the heart of the analysis, only 12 had a low risk of bias, and eight of them found no inverse connection between fluoride and IQ, Levy wrote.

Then there’s the use of urine to measure fluoride exposure. The report authors touted this as a more precise way to measure an individual’s exposure to fluoride from all sources, not just drinking water. But that reasoning is contrary to the “scientific consensus,” Levy wrote. Urinary fluoride measurements vary significantly over the course of a day and from one day to the next, so there’s no way to know whether any particular sample is indicative of a person’s long-term exposure.

Levy also chided the report authors for cherry-picking the studies they included in their analysis. For instance, given the choice of two publications based on data from the Canadian Maternal-Infant Research on Environmental Chemicals study, the report authors opted against the one that included fluoride exposure after birth. The publication they omitted found no link between “fluoride exposure during pregnancy, infancy, or childhood and full-scale IQ,” he wrote.

Other recent, high-quality studies showing no association got short shrift as well, he added.

Is that all?

There are other critiques about methodology and statistical analysis. But one of Levy’s biggest complaints about the report is the “lack of transparency” about its backstory.

The authors downplayed the report’s link to the controversial monograph they produced for the National Toxicology Program, Levy wrote. The first two drafts of that monograph received harsh peer reviews from the National Academies of Science, Engineering and Medicine. The initial version lacked clear evidence to support the authors’ claim that “fluoride is presumed to be a cognitive neurodevelopmental hazard to humans,” and the second one glossed over the fact that it wasn’t equipped to shed light on the risks posed by the low concentration of fluoride in U.S. water systems, the NASEM reviewers said.

Nor, Levy wrote, does the new report mention that animal studies using fluoride levels that reflect the U.S. standard of 0.7 mg/L found “no exposure-related differences in motor, sensory, or learning and memory performance” in nearly a dozen tests, as researchers reported in 2018.

How can I tell whether fluoride is being added to my water?

State and local authorities decide for themselves whether to fluoridate the water supply. In some places, the water is naturally high in fluoride because there happens to be a lot of it in the soil and bedrock. If the concentration is higher than 2.0 mg/L, the EPA requires officials to notify people who drink that water within 12 months. If the concentration exceeds 4.0 mg/L, officials must notify people within 30 days and take steps to reduce fluoride to safe levels.

Nearly 63% of Americans receive fluoridated water, including the 3.5% whose fluoride levels exceed optimal levels, according to the CDC. If you want to see whether your water system adds fluoride, try looking it up on the CDC website. (Depending on where you live, you may have to contact your water supplier directly.)

If you live in Los Angeles County, you can use this map to see whether you’re among the 62% receiving “optimally fluoridated” water, the 5% whose water is “largely fluoridated,” the 22% whose water is “partially fluoridated,” or the 11% whose water isn’t fluoridated.

That doesn’t mean the water is fluoride-free: According to the Los Angeles Department of Water and Power, the city’s groundwater contains fluoride at concentrations ranging from 0.1 to 0.3 mg/L, and fluoride levels in the water supplied by the Los Angeles Aqueduct range between 0.4 and 0.8 mg/L. All water delivered by the DWP is adjusted to a fluoride concentration of 0.7 mg/L, the agency says.

So what’s the bottom line?

The report authors acknowledged that their analysis “was not designed to address the broader public health implications of water fluoridation in the United States.” Even so, they suggested that their findings “may inform future public health risk-benefit assessments of fluoride.”

A second editorial that accompanies the report said it raises enough questions to warrant a reassessment of “the potential risks of fluoride during early brain development.” The lack of a clear link between IQ scores and fluoride exposure below 1.5 mg/L “does not exonerate fluoride as a potential risk,” the editorial argued.

Levy disagreed. “There is no evidence of an adverse effect at the lower water fluoride levels commonly used” in water systems in the U.S., he wrote. “The widespread use of fluoride for [cavity] prevention should continue.”