**The Fluoride Paradox: When Both Sides Cherry-Pick the Same Science**

Alex: Right, so November 2025, major study comes out saying fluoride in drinking water makes teenagers smarter. American Dental Association celebrates, headlines everywhere. And then almost immediately—like within days—the anti-fluoride activists are citing completely different studies saying fluoride damages children's brains, causes IQ loss. Same chemical. Opposite conclusions. I'm looking at these headlines thinking, how is this even possible?

Bill: Okay, so this one's fascinating because both sides are technically citing real studies.

Alex: Right.

Bill: But they're telling completely different stories with the same evidence. And honestly, when I dug into the actual data, neither side is being straight with people.

Alex: That's what got me. It's not that one side is just making things up entirely—they're both pulling from legitimate research. But somewhere between the studies and the headlines, something's gone very wrong.

Bill: So let's start with the positive claim—the "fluoride makes kids smarter" study. This came from researchers at the University of Minnesota, published in Science Advances. They followed nearly 58,000 students.

Alex: Okay.

Bill: From 1980 through to age 60 in 2021.

Alex: That's quite a long study.

Bill: It is. So they found that students who were exposed to fluoridated water during childhood scored about 7% of a standard deviation higher on reading, math, and vocabulary tests in 12th grade.

Alex: Hang on, 7% of a standard deviation—translate that for normal humans.

Bill: Fair. Think of it this way: if the average kid scores 100 on a test, these kids might score around 102 or 103.

Alex: Okay.

Bill: It's not huge, but it's measurable. Statistically significant. And the headline writes itself: "Fluoride linked to better cognitive performance in adolescents."

Alex: Right, but I'm already skeptical because that headline and what you just described feel like two different things. "Better cognitive performance" sounds like fluoride is actively making kids smarter. Scoring slightly higher on a test could mean a hundred different things.

Bill: Well, but that's the correlation they found. By the time these same students reached age 60, that advantage had completely disappeared. No statistically significant difference in cognitive function at all.

Alex: Hang on, so the benefit just... vanishes? That doesn't sound like a biological effect to me. That sounds like something else was going on.

Bill: The researchers themselves acknowledged this—

Alex: Did they really?

Bill: Yes. They explicitly wrote in their conclusion that they "are unable to fully explain the positive association" and that "the associations may be due to unobserved student- or community-level confounding."

Alex: Translation: "We found a correlation, but we can't prove it was actually the fluoride."

Bill: Exactly. But here's where I think it gets interesting—when you look at how they measured fluoride exposure, they didn't measure individual consumption. They looked at whether students' high schools were located in areas with fluoridated water.

Alex: Wait, I'm sorry, they're assuming that because your school is in a town with fluoridated water, you drank that water your whole childhood?

Bill: Pretty much. And the students exposed to fluoridated water were systematically different from those who weren't. Way more likely to be from urban areas: 27.5% versus 13.1% in non-fluoridated areas.

Alex: Oh brilliant. So you're comparing urban kids with better-funded schools, more resources, different peer groups, to rural kids with completely different educational environments. And you can't just control that away statistically.

Bill: Well, they tried to control for that. They adjusted for socioeconomic status, region, urbanicity—

Alex: But that doesn't work if the confounding is structural. When I was covering education policy, you'd see this all the time—studies that try to adjust for socioeconomic status like it's just a variable you can subtract out. But you can't. The environments are fundamentally different.

Bill: I mean, the researchers did admit their data wasn't complete. They wrote—and I'm quoting here—"Most seriously, we would have preferred more complete information about where panelists lived from conception through late adolescence."

Alex: That's quite an admission. Actually, that's a massive admission. So we've got one observational study with acknowledged confounding issues, incomplete data on exposure, and an effect that disappears by adulthood. And that becomes "fluoride makes kids smarter" in the headlines.

Bill: Right.

Alex: Meanwhile, what are the anti-fluoride people citing?

Bill: They're primarily pointing to a meta-analysis by Taylor and colleagues, also from 2025, published in JAMA Pediatrics. This one reviewed 74 studies from around the world and found that higher fluoride exposure was associated with lower IQ scores. The effect size was actually pretty substantial—a standardized mean difference of -0.45.

Alex: That sounds significant. Is that where RFK Jr. and the anti-fluoride activists get the "fluoride causes IQ loss" claim?

Bill: Yes. But here's what they're not telling you—and this is the critical piece. That -0.45 effect? It's primarily coming from studies in China, India, and Iran where fluoride levels are really high.

Alex: How high?

Bill: Most of those studies looked at fluoride levels above 1.5 milligrams per liter. Some much higher. The recommended level in US drinking water is 0.7 milligrams per liter.

Alex: Wait, so that's more than double?

Bill: More than double. And when you break down the Taylor meta-analysis by country and fluoride level, you get a completely different picture. The 41 studies from China showed that -0.42 effect. But the two studies from Canada? Effect size of 0.01—basically zero.

Alex: Huh.

Bill: Pakistan studies? 0.10, again essentially nothing. New Zealand? 0.01.

Alex: So the studies from places with low fluoride levels, similar to what we use, show no effect.

Bill: Correct. And there's another meta-analysis by Kumar and colleagues from 2023 that specifically looked only at studies with fluoride at or below 1.5 milligrams per liter. Their conclusion was that fluoride exposure relevant to community water fluoridation is not associated with lower IQ scores in children.

Alex: Okay, but hang on. I want to push back on something here.

Bill: Okay.

Alex: You're saying the high-dose studies aren't relevant to US policy because the doses are different. But aren't they still evidence that fluoride can harm brain development? Like, if we know it's neurotoxic at high doses, shouldn't we be more cautious about low-dose exposure, especially in vulnerable populations?

Bill: I mean, that's a fair point, but dose-response relationships don't work that way. Just because something is harmful at one level doesn't mean it's harmful at a tenth of that level. When I was doing A/B testing for marketing campaigns, we'd see this all the time—massive effects at high exposures that completely disappeared below a threshold.

Alex: Right, but brains aren't marketing campaigns. We're talking about developing children.

Bill: True, but the studies at low doses—the ones actually relevant to US water levels—consistently show no effect. It's not like there's ambiguity and we're ignoring it. The signal is pretty clear.

Alex: Okay. Okay, I'll grant you that. But then what do we do with the Minnesota study? Because you're dismissing the high-dose harm studies as irrelevant, but you're also saying the low-dose benefit study is confounded and unreliable. Are we just not supposed to study fluoride?

Bill: No, I'm saying—actually, let me back up. I'm not dismissing the high-dose studies. They're absolutely valid evidence that fluoride is neurotoxic at high concentrations. What I'm saying is that the anti-fluoride activists are taking those valid high-dose studies and misapplying them to argue against low-dose water fluoridation. That's the cherry-picking.

Alex: Right, okay. And the pro-fluoride side is taking one confounded study and holding it up as proof of benefits that probably aren't real.

Bill: Exactly. Both sides are misrepresenting dose-response relationships. This is classic cherry-picking from opposite ends of the same dataset.

Alex: What does the actual scientific consensus say when you look at the complete picture?

Bill: So the National Toxicology Program put out a comprehensive review in August 2024. They concluded that fluoride at levels above 1.5 milligrams per liter is associated with lower IQ in children. But they explicitly stated that more research is needed on effects below 1.5 milligrams per liter—the range that's actually relevant to US water policy.

Alex: So the consensus is: definitely harmful at high doses, unclear at low doses?

Bill: Not quite. When you look at the studies specifically examining low doses close to what we use—0.7 milligrams per liter—they consistently show no effect. The "more research needed" caveat is scientific caution, but the signal is pretty clear: at recommended levels, fluoride isn't affecting IQ.

Alex: But it's also not boosting IQ, despite what that Minnesota study suggests.

Bill: Right. That study is an outlier. It contradicts the broader pattern, it has methodological limitations the authors themselves acknowledge, and the effect disappears by adulthood, which isn't consistent with a real biological benefit.

Alex: So where does that leave us? What's the actual truth here?

Bill: The actual truth is boring: at the levels used in US water fluoridation—0.7 milligrams per liter—fluoride appears to be neither harmful nor beneficial for cognitive development. It does prevent cavities, which is well-established. But brain-boosting or brain-damaging? Neither is supported at recommended doses.

Alex: But that won't fit a headline, will it? "Fluoride: Basically Fine, Helps Your Teeth, Probably Won't Make You Smarter or Dumber." Not exactly clickbait.

Bill: No. And that's exactly why we're seeing these dueling narratives. The dental industry wants to defend water fluoridation, so they amplify the one study showing benefits while downplaying dose-response concerns. Anti-fluoride activists want to end fluoridation, so they show you the Chinese high-dose studies and pretend they're relevant to US policy.

Alex: What's frustrating is that both approaches undermine trust in science. When people see these completely contradictory claims, both supposedly backed by research, they stop believing anyone.

Bill: And the actual science is being misused by both sides. The dose-response relationship is real—fluoride does cause harm at high levels. But conflating high-dose toxicity with low-dose safety policy is dishonest. And using one confounded observational study to claim cognitive benefits is equally misleading.

Alex: Here's what I'm wondering—why is the dose distinction so hard for people to communicate? We understand this with loads of other substances. No one's confused about the fact that a glass of wine is different from a bottle of vodka.

Bill: Actually, hang on. Didn't we talk about this before? The dose-response thing?

Alex: Mmm.

Bill: The plastics episode, I think. Detection versus harm?

Alex: Oh, right. Yeah, that was the same issue—people freaking out because microplastics were detected in brains, but detection doesn't mean harm. You need to know the dose.

Bill: Exactly. And I think it's because admitting "it depends on the dose" requires nuance, and nuance doesn't mobilize people.

Alex: Right.

Bill: If you're campaigning to end fluoridation, "it's toxic at high doses" is less compelling than "it's poisoning our children." If you're defending it, "it's safe at low doses" is less exciting than "it makes kids smarter."

Alex: The problem is that when the truth finally comes out—and it always does—people remember being misled. They remember the headlines that didn't match the studies. And then they become skeptical of legitimate public health measures because they've been burned before.

Bill: Yeah. That's the real cost.

Alex: Okay, so what should people actually take away from this fluoride debate?

Bill: First, dose matters enormously. A chemical that's harmful at one level can be safe or even beneficial at another. Anyone making claims about fluoride without specifying the dose is not being straight with you.

Alex: Second, one study doesn't overturn a pattern. That Minnesota study is interesting, but it's an outlier with limitations the researchers themselves acknowledged. Don't let anyone use it to claim fluoride boosts intelligence.

Bill: Third, if someone's citing studies from China or India about fluoride toxicity and applying them to US water policy, check the doses. Those studies are real, but they're measuring exposure two or three times higher—sometimes more—than what we use.

Alex: Was it three times? I thought you said 1.5 versus 0.7. That's just over double.

Bill: Right, yeah, sorry. At least double. Some of the studies in the meta-analysis looked at much higher levels, but the threshold they used was 1.5.

Alex: Okay.

Bill: And finally, the scientific consensus—when you actually look at all the evidence together rather than cherry-picking—is that fluoride at 0.7 milligrams per liter is safe for cognitive development and effective for cavity prevention. It's not exciting, but it's accurate.

Alex: The real lesson here is about how the same evidence can be weaponized to tell opposite stories. Both sides are using real data. Both are misrepresenting what it actually shows.

Bill: Because the truth doesn't fit anyone's narrative. It's just chemistry doing what chemistry does at different doses.

Alex: Even when it's boring.

Bill: Especially when it's boring.