One of the most important lessons from the Covid-19 experience is that absolutes in scientific discourse are the exception. Science evolves as both controlled clinical research and real-world evidence show us how to use health care technologies to better advance the human condition.

Such a tool is fluoride. Extensively studied over decades, fluoride has been scientifically validated for its effectiveness in reducing dental cavities, particularly in children. It is incorporated into municipal water supplies, toothpastes, mouth rinses, and dietary supplements. The benefits of fluoride writ large far outweigh its risks, but no health care intervention is ever 100% risk-free. It’s the Food and Drug Administration’s job to ensure the safety, efficacy, and quality of fluoride-containing drugs and supplements based on robust scientific analysis and regular post-marketing surveillance.

This past May, the FDA announced, in a cart-before-the-horse statement, that it intends to remove fluoride supplements from the market effective October 2025, after a safety review is completed. According to the FDA, this safety review applies to “ingestible fluoride prescription drug products” — not toothpaste, mouth rinses, or other topical products containing fluoride. But what is the point of a safety review if the agency has already made up its mind? Instead of this pale imitation of a scientific inquiry, it should use a new AI tool at its disposal to actually answer any questions about safety.

According to the American Dental Association, “A review of the studies offered as referenced by U.S. Department of Health and Human Services do not in fact demonstrate any harmful effects for the concentrations of fluoride prescribed by physicians and dentists.” But HHS Secretary Robert F. Kennedy Jr. said in a statement from the department, “Ending the use of ingestible fluoride is long overdue. … This decision brings us one step closer to delivering on President Trump’s promise to Make America Healthy Again.” But shouldn’t the decision be made after the scientific review?

On July 23, the Reagan-Udall Foundation, in cooperation with FDA, held a public meeting to gather input on the clinical use and safety concerns associated with orally ingestible unapproved prescription drug products containing fluoride for use in children. The meeting was a reminder of how contentious the issue remains within both the scientific community and the FDA. Both sides of the issue were present, and the panel discussions were, at times, tense.

Susan Winckler, chief executive officer of the Reagan-Udall Foundation, had the difficult task of moderating panels that included both advocates for fluoride supplementation and its detractors. Emotions ran high. During one heated exchange, while trying to maintain a modicum of civil and scientific discourse, Winckler had to remind one speaker, “Sir, let’s illustrate versus interrogate.”

Under debate are a subset of fluoride-containing products, available only through prescription by a licensed health care professional, that are marketed and used without FDA approval. While it is uncommon, some drugs available by prescription are not FDA-approved for safety and efficacy. Often, these “unapproved” drugs are available with a prescription because they were available before current approval processes were in place. Many of these medications were on the market before 1962, when the FDA first started requiring manufacturers to show that a new medication was safe and effective for its intended use.

Prescription fluoride drops and tablets for ingestion were first marketed in the United States in the 1940s to prevent dental caries, particularly in children who lived in areas with low or no water fluoridation. Although these products are still prescribed today, none have been reviewed and approved by FDA for safety, effectiveness, or quality. Over the years, numerous medical and dental organizations in the United States have recommended the use of ingestible fluoride products for those for whom regular topical treatment is difficult to achieve and those for whom topical fluoride only is insufficient to avoid dental decay.

These unapproved drugs present complex challenges for public health, especially for vulnerable populations like children, who are at greater risk for both the benefits and potential harms of fluoride exposure.

Formally withdrawing medical products requires a lengthy rulemaking process that can take years. It is likely that the FDA will ask manufacturers to voluntarily pull their products.

Rather than peremptorily removing these products from the market under the auspices of a predetermined outcome by anonymous reviewers, the FDA should convene a publicly announced blue-ribbon panel of experts. Nominated by both sides of the debate, they should robustly and transparently discuss and deliberate the risks and benefits of these medicines, while simultaneously fielding a comprehensive study of pediatric electronic health records to help determine the real-world outcomes of these unapproved — but often highly effective interventions.

Perhaps parts of this analysis could be augmented by the capabilities of Elsa, the FDA’s highly touted artificial intelligence program. Elsa is a large language model–powered AI tool designed to assist with reading, writing, and summarizing. It can summarize adverse events to support safety profile assessments, perform faster label comparisons, and generate code to help develop databases for nonclinical applications.

In our current age of misinformation, senior leaders of our public health establishment urge us to “do our own research,” leading to emotionally charged anecdotes often trumping legitimate scientific inquiry. Such is certainly the case with fluoride — and particularly with the use of unapproved prescription drug products containing fluoride in pediatric populations. But the plural of anecdote isn’t data.

In the absence of solid evidence to the contrary, public health sentiments — and public health actions — are being driven by stories rather than science.

According to FDA Commissioner Makary, “When it comes to children, we should err on the side of safety.” Certainly. But actions based on anecdotal evidence are actions based on emotion rather than science. It is important to note that no serious or robust data exists to support the contention that these products pose a threat to public health. But, in fairness, absence of evidence is not evidence of absence. And, as more states and municipalities consider removing fluoride from their water supply, alternatives become ever more important to maintain the dental health of America’s children.

Rather than rashly removing fluoride products that are only available through the advice and counsel of learned and licensed expert medical intermediaries, the FDA should seek to advance the science, based not on political pressure, but on sound science. The FDA must allow the science of fluoride to evolve.

As the Yiddish proverb reminds us, “If you can’t bite, don’t show your teeth.”

Peter J. Pitts, a former FDA associate commissioner, is president of the Center for Medicine in the Public Interest.