The question of filtering fluoride out of drinking water.

While people can elect to use fluoridated toothpaste, people who do not want fluoridated drinking water face added burdens. The biggest burden is on low-income groups. A Chat with Chat.

After reading our last fluoride article, people emailed us to seek advice about fluoride filters. We’re in the middle of AGM prep - so we asked Chat GPT for help.

Discernment is key when working with a large language model (LLM), and critically, always check your references. Today we have posed a simple group of questions to OpenAI’s ChatGPT (GPT-5), and no - we’re not checking them! We’ve told Chat we’re publishing it and requested that Chat go back through the claims and self-verify.

However, like any tool for any solution, people may have personally tried different fluoride filtering methods and arrived at a preferred solution. In the comments section below (particularly for New Zealanders) PSGR welcome advice, personal experience with products, criticism (of products they have previously used, or of Chat’s advice), and different and competing advice. The people can find out the - more readers will know.

• Reverse Osmosis = continuous supply, higher upfront cost, ongoing filter changes, water waste (but convenient once installed).

• Distillation = almost total purity, cheaper upfront, slower, uses electricity, batch production (so less convenient for families). Distillation achieves the highest removal (near 100%).

• Reverse osmosis systems remove 85–95% of fluoride and are more convenient for continuous supply.

• Problem of stripping naturally occurring minerals and salts.

This is a personal/family solutions article. Please kindly refrain from discussing fluoride’s toxicity or government inaction.

We particularly welcome in the thread below - tips about where the public can access clean and safe natural springs on public land close to roadsides and towns that the public can access to refill drinking water.

REVERSE OSMOSIS?

Reverse osmosis pushes water through a semi-permeable membrane with very fine pores. RO membranes have pore sizes in the ~0.0001 µm range, small enough to block most ions including fluoride.

Most dissolved salts, heavy metals, and small molecules (including fluoride ions) are rejected, while water molecules pass through. Fluoride is monovalent but still effectively rejected; rejection rates depend on membrane type and water chemistry.

Fluoride removal efficiency is typically around 85–95%, depending on system design, membrane quality, water pressure, and maintenance. EPA, WHO, and industry literature put fluoride rejection in this band.

Efficiency declines if the membrane is old, fouled, or damaged. Maintenanrtce is key.

RO membranes reject almost all dissolved salts and minerals, including calcium, magnesium, potassium, and trace elements. This includes many beneficial minerals so water may taste “flat” unless remineralised. This is a widely acknowledged issue with RO.

Remineralisation cartridges can be often added to improve taste and restore buffering capacity.

Household RO units usually need pre-filtration (sediment and carbon filters) to protect the membrane. This is standard practice in RO systems.

Household RO units typically waste 2–4 litres of water for every litre of purified water, though newer systems can be more efficient.

⚠️ Needs nuance. This ratio is accurate for older or cheaper under-sink systems. Newer “low-waste” RO systems can be more efficient (1:1 or 1:2). Still, 2–4:1 is a realistic household average.

There seem to be a wide variety of reverse osmosis brands in NZ including Pentair, Ecosoft, Oasis, Puretec, WaterPro, Hydroflux Epco NZ, Pure Hydration, Ecoworld, G-Verse, Zenith Water, NatureZone Water, Taylor Purification, Wellington Water Filters.

(Please check your own sources, not intended to reflect all products that are available.)

---

ALTERNATIVES

Activated alumina filters can also remove fluoride, though performance depends heavily on pH. Activated alumina works best at pH 5–6.

Distillation removes fluoride as well. Near total removal.

• Distillation boils water and re-condenses the steam.

• Minerals and salts do not vaporise with the water, so they are left behind.

• Like RO, distilled water is very low in calcium, magnesium, potassium, etc.

• Some distillers use a small post-filter (activated carbon), but that targets volatile organics, not minerals

Standard (cheaper) carbon filters (like Brita) do not remove (filter/absorb) fluoride.

In short: yes, RO is a reliable method of reducing fluoride in drinking water, but it requires upkeep and comes with trade-offs (mineral loss and water waste).

REVERSE OSMOSIS OR DISTILLATION?

Distillation removes nearly all fluoride. Published research and water industry sources confirm that distillation achieves ~99–100% fluoride removal.

Reverse osmosis removes 85–95%. True, though figures vary by membrane and water chemistry. This range is well supported.

Standard carbon filters (e.g. Brita) do not remove fluoride. Correct.

Bench-top distillers are commercially available (Megahome, Springflow, AquaNui, Pure Water, etc.). Correct. These are manufactured for household drinking water, not repurposed lab equipment.

Distillers produce ~4 L over 4–6 hours. Accurate for typical 750–800W domestic units.

Distillers often include a carbon filter to remove volatile organics. Correct.

RO produces wastewater (2–4 L per 1 L clean water). Correct.

Springflow/Megahome: Yes, real brands/models. However, “Springflow” seems to be a reseller brand in NZ for the Megahome unit.

Mitre 10: While some Mitre 10 stores list water distillers, availability is patchy and may change. Don’t imply it is universally stocked.

---

CAN YOU AFFORD A FILTER? QUESTION TO CHAT:

For many households, particularly those on low incomes, accessing fluoride-free drinking water presents both financial and practical barriers. Reverse osmosis and distillation remain the most effective technologies for removing fluoride, but both require significant upfront costs, ongoing maintenance, and, in the case of distillation, additional electricity use. For families already stretched with essentials such as food, housing, and transport, such investment may be unrealistic. This creates a disparity where those with fewer resources are unable to exercise the same level of choice or protection as more affluent households.

Where alternative supplies exist—such as local springs, rainwater tanks, or community bore supplies—the effort required to access them introduces additional burdens. Even when councils or community groups provide fluoride-free refill stations, unless these are located close to where people live and work, the extra travel time can make them effectively inaccessible. For single-car or no-car households, the task of collecting and transporting water can be time-consuming, physically demanding, and often falls to the primary caregiver.

This is especially concerning for low-income families with young children: the very demographic most vulnerable to potential fluoride toxicity due to their smaller body size and higher water consumption relative to weight. These households may face a compound disadvantage: not only higher exposure risk, but also the greatest barriers to mitigation, constrained by childcare responsibilities, limited time, and restricted transport options. The result is a structural inequity where the people least able to bear the risk also have the least capacity to avoid it.

---

RISKS? LONG-TERM CONSUMPTION OF DEMINERALISED WATER

Both methods produce very “pure” but mineral-depleted water: Health agencies note that most people get essential minerals from food, not drinking water — but in places where water is a significant mineral source (hard water areas), switching to RO or distilled water reduces that contribution.

Both reverse osmosis (RO) and distillation produce very pure water, but in doing so they strip away nearly all naturally occurring minerals. RO membranes are designed to reject dissolved salts, metals, and other ions, which means beneficial minerals such as calcium, magnesium, and potassium are removed along with contaminants. Distillation has a similar effect: as water is boiled and condensed, minerals do not vaporise and are left behind in the boiling chamber. The result in both cases is demineralised water that can taste “flat” or slightly acidic.

For most people, this loss of minerals in drinking water does not pose a major nutritional risk, since the majority of essential minerals come from food. However, in regions where water supplies are naturally “hard” and provide a meaningful amount of calcium and magnesium, switching to RO or distilled water removes that contribution. Some studies have suggested a possible link between long-term consumption of very soft or demineralised water and higher risks of cardiovascular issues or mineral imbalance, though evidence is mixed and not definitive.

Practical concerns are often more immediate: pure water can be corrosive to plumbing, and it may leach small amounts of metals from pipes or storage containers. For these reasons, many RO systems now include remineralisation cartridges to add back calcium and magnesium for both taste and stability. Distilled water can also be remineralised by adding trace mineral drops or mixing with mineral-rich water.

In short, both RO and distillation are effective at removing fluoride and other contaminants, but they also strip away minerals. The health implications are debated, but for most households the key issue is taste and balance rather than frank deficiency. Where mineral intake from water is important, remineralisation is advisable.