NSF 42, 53, 58 and 401: What Water Filter Certifications Actually Cover

“NSF certified” is printed on almost every water filter sold, which is exactly why it tells you so little. Certified against which standard, for which contaminant? Those are the questions the phrase is quietly skipping.

Short answer: NSF/ANSI 42 covers aesthetic effects like chlorine, taste and odor. 53 covers health effects, which is where lead, VOCs and PFAS claims live. 58 covers reverse osmosis systems. 401 covers emerging contaminants such as pharmaceutical residues. Certification is granted claim by claim, not as a blanket, so a filter certified to 53 is certified for the specific contaminants listed on its data sheet and nothing else.

The numbers, and what each one actually buys you

Standard	What it covers	What that means for you
NSF/ANSI 42	Aesthetic effects	Chlorine, taste, odor, some particulate. Nothing health-based.
NSF/ANSI 53	Health effects	Lead, cysts, VOCs, and where PFAS claims are certified for carbon filters
NSF/ANSI 58	Reverse osmosis systems	Whole-system performance for RO, including PFAS and dissolved solids
NSF/ANSI 401	Emerging contaminants	Pharmaceutical residues, pesticides, and similar trace chemicals
NSF/ANSI 372	Lead-free materials	The product doesn’t leach lead. Not a claim it removes any.

That last row catches people out constantly, including plenty of review sites. A filter advertising 42 and 372 has told you it improves taste and that its own housing won’t add lead to your water. It has not told you it removes lead. Those are different standards answering different questions, and a product can hold both while making no certified health claim at all.

The certification is per-contaminant, not per-product

This is the single most misread thing in the category. NSF/ANSI 53 isn’t a badge a filter either has or lacks. It’s a framework, and a manufacturer certifies against specific contaminant reduction claims within it.

So two filters can both say “NSF/ANSI 53 certified” while one is certified for lead and cysts and the other for VOCs and PFAS. The claims are listed on the product’s performance data sheet, which manufacturers are required to make available and which almost nobody reads. That sheet, not the marketing page, is where the real answer lives.

P473 is retired, and boxes still say it

Here’s a live piece of misinformation worth knowing about, because it’s still circulating on retailer pages and review sites in 2026.

NSF P473 was a standalone protocol covering PFOA and PFOS reduction. It was retired and folded into NSF/ANSI 53 and 58 back in 2017. The 2022 revisions then went further, adding reduction criteria for five more PFAS, PFHxS, PFNA, PFHpA, PFBS and PFDA, and introducing a combined “Total PFAS” reduction claim.

So a product page advertising “P473 certified” today is quoting a protocol that no longer exists as a separate thing. It isn’t necessarily a lie, and the underlying certification may be perfectly real, but it’s a sign the copy hasn’t been updated in years. The current, correct thing to look for is a 53 or 58 certification that explicitly names PFAS or carries the Total PFAS claim.

NSF isn’t the only legitimate certifier

The standards are written by NSF and accredited by ANSI, but NSF the organization isn’t the only body that can certify products against them. IAPMO R&T and the Water Quality Association are also ANSI-accredited to do it, and a product certified by IAPMO against NSF/ANSI 53 is properly certified against NSF/ANSI 53.

This trips up shoppers who go looking for a product in NSF’s own listings, don’t find it, and conclude the claim is fake. Check the certifier named on the data sheet, then check that body’s listing database.

What should make you skeptical is different wording entirely. “Tested to NSF standards” and “meets NSF standards” are self-declarations by the manufacturer, not third-party certification. Anyone can run a test and describe the result that way. The distinction between “certified” and “tested to” is doing an enormous amount of work in this category, and it’s the first thing to look for on any product page.

Independent lab testing, and when it’s a fair substitute

Some good filters skip certification for their headline claims and publish independent laboratory results instead. That’s a real and defensible choice, because certification is expensive and slow, especially for a small company adding claims across a wide contaminant list.

Treat it as weaker evidence than certification, not as worthless. What makes lab results credible is specificity: an accredited laboratory named, the contaminants and levels listed, the volume of water the filter had processed when tested, and the full report available rather than a marketing summary. Vague “independently tested to remove 99% of contaminants” language with no report behind it is a claim about nothing.

What to actually do with this

Match the number to your problem, which means knowing your problem first. Find out what’s in your water, then look for the standard that addresses it: 53 for lead and PFAS on a carbon filter, 58 for a reverse osmosis system, 42 if taste is honestly all you care about.

Then read the performance data sheet rather than the product page, and check that the specific contaminant you care about is named. Our comparison of reverse osmosis and carbon filters covers which mechanism suits which contaminant, and our vetted water filter picks note what each one is certified for and where a brand leans on its own lab testing instead.

Frequently asked questions

Is NSF/ANSI 42 enough on its own?

Only if taste is your goal. Standard 42 covers aesthetic effects, meaning chlorine, odor and clarity, and carries no health claims whatsoever. A filter certified to 42 alone is a taste product. That’s a legitimate thing to buy if your water is safe and unpleasant, but it isn’t protection against lead, PFAS or anything else health-related.

Which NSF standard covers PFAS?

NSF/ANSI 53 for carbon and anion exchange filters, and NSF/ANSI 58 for reverse osmosis systems. The old standalone P473 protocol was folded into both in 2017, and 2022 revisions added five further PFAS along with a combined “Total PFAS” claim. Look for PFAS named explicitly in the certified claims rather than assuming a 53 certification includes it.

What’s the difference between “NSF certified” and “tested to NSF standards”?

Certification means an accredited third party such as NSF, IAPMO or WQA verified the product against the standard, audits the manufacturing, and lists it publicly. “Tested to” or “meets” NSF standards is the manufacturer describing its own results, with nobody independent verifying them. The wording gap is deliberate and worth treating as a red flag.

Does a filter need certification to be good?

No, but it needs evidence. Plenty of capable filters publish accredited independent lab results instead of pursuing certification for every claim, which is a reasonable route for a smaller company. What matters is whether the evidence is specific and public: named laboratory, named contaminants, stated test volumes, full report available. A confident claim with no data behind it is the thing to walk away from.

Related reading

Start with what’s actually in your tap water, since the certification only matters relative to your contaminant. Then reverse osmosis versus carbon filters for the mechanism, whether Brita removes PFAS for a worked example of reading a claim, and our vetted picks. For how this pattern of reassuring labels plays out elsewhere, see our greenwashing guide.