The thing most coverage gets backwards
You will read that EPA has no chromium-6 standard, and that this is a gap.
The first half is true. The second is not EPA’s account of it.
EPA sets 0.1 mg/L — 100 parts per billion — for total chromium, and states that this “includes all forms of chromium, including chromium-6”, with systems required to test for total chromium.
Its reason for one limit rather than two:
Chromium-6 and chromium-3 are covered under the total chromium drinking water standard because these forms of chromium can convert back and forth in water and in the human body… Measuring just one form may not capture all of the chromium that is present. In order to ensure that the greatest potential risk is addressed, EPA’s regulation assumes that a measurement of total chromium is 100 percent chromium-6, the more toxic form.
So the single number is not a loophole. It is a conservative assumption: every microgram of chromium in your report is judged as though it were the more toxic species.
What is genuinely dated, and it is not the species
The limit’s basis, not its coverage.
EPA states the MCL was established in 1991, on evidence at the time that continued chromium-6 exposure could result in allergic dermatitis — skin reactions. It began a comprehensive review of chromium-6 health effects in 2008, and is developing an IRIS assessment covering both inhalation and ingestion.
That review is unfinished. No revised standard has been issued.
So the honest position is narrow: the number covers the right thing, conservatively, and rests on a health endpoint set a generation ago while a broader assessment remains in progress. This page takes no view on where a revised limit should land.
What your report can and cannot tell you
It gives total chromium, against 0.1 mg/L. It does not split the two forms, because systems are not required to measure them separately.
And there is no sensory backstop: EPA describes chromium as “an odorless and tasteless metallic element.” Nothing about it will announce itself.
→ How to read your water quality report
Worth holding alongside: chromium-3 is, in EPA’s words, “an essential human dietary element”, found in vegetables, fruits, meats, grains and yeast. Chromium-6 occurs naturally from erosion of chromium deposits and can also be produced by industrial processes. The two are not interchangeable in how they behave, which is exactly why the interconversion point matters.
Where certification is more specific than regulation
This is the inversion. The regulator uses one number for both forms; the certifier splits them into three separate claims.
| Claim | Standard | Distinct models | Category |
|---|---|---|---|
| Chromium (hexavalent) reduction | 58 | 120 | All plumbed-in to separate tap |
| Chromium reduction | 53 | 7 | All pour-through |
| Chromium (trivalent) reduction | 53 | 4 | Mostly plumbed-in to separate tap |
Hexavalent chromium reduction is selectable under NSF/ANSI 58 alone — alongside fluoride, nitrate and TDS — and not under NSF/ANSI 42 or 53.
→ Which claims exist at all, and how populated each is
So a buyer who wants the hexavalent form addressed specifically is buying under-sink reverse osmosis. There is no other certified format in this listing.
And a pitcher marked “chromium reduction” is holding a different claim. Seven pour-through models held it, and it is not the hexavalent claim.
The claim that comes bundled
The 120 hexavalent chromium models are a strict subset of the 127 holding NSF/ANSI 58 fluoride and lead claims. Every hexavalent model holds all three; seven fluoride-and-lead models do not hold hexavalent.
Which gives a shortcut: search for the hexavalent chromium claim and you get the fluoride and lead claims with it. It is the narrowest of the three, so it is the one to filter on.
→ Filters certified for fluoride, and why they are all one format
The decision framework
1. Find your total chromium figure
Compare against 0.1 mg/L. Remember what the figure already assumes — that all of it is chromium-6.
If it is well below, EPA’s conservative assumption is working in your favour and there is no speciation question left to answer.
If it is near the limit, the assumption is still conservative, and the remaining question is whether you are satisfied with a 1991 dermatological basis.
2. Decide whether the dated basis concerns you
This is a judgement, not a fact. EPA’s review has been open since 2008 and its IRIS assessment is unfinished.
Reasonable people land differently here, and “no” is a defensible answer for a supply testing well under the limit.
3. If you want treatment, accept the format
Hexavalent chromium reduction exists under one standard and one product category: under-sink reverse osmosis plumbed to its own tap.
If that fits, 120 models from 34 companies hold the claim.
If it does not, no certified alternative exists in this listing. A pitcher with a chromium claim is not a smaller version of it.
4. Check the claim is listed on the model
Under NSF/ANSI 58, TDS reduction is the only required claim. Everything else, hexavalent chromium included, is optional and listed per model.
→ How to read an NSF listing, footnotes included
Applying the framework: who holds the claim
| Company | Models |
|---|---|
| Microline | 14 |
| Microfilter Co., Ltd. | 12 |
| EcoWater Systems LLC | 10 |
| The LeverEdge | 8 |
| Water Channel Partners | 8 |
| Pentair Residential Filtration | 7 |
| Hellenbrand Inc. | 6 |
| Solventum Purification Inc. | 6 |
| Aqua Systems | 4 |
| Clover Co., Ltd. | 4 |
The remaining 24 companies hold three or fewer each.
Notes tied back to the criteria
- The holder list is close to the fluoride list, which follows from the subset relationship rather than being a separate finding.
- No company dominates: the biggest single holder has 14 of the 120 certified models.
- Several are dealer-channel water treatment firms rather than retail brands, which affects how you would buy one more than whether it is certified.
- The seven pour-through chromium models sit under two companies, so the non-RO chromium option is narrow as well as differently-claimed.
The decision path
| Your situation | What decides it | Where that lands you |
|---|---|---|
| Total chromium well under 0.1 mg/L | The figure already assumes worst case | No purchase indicated |
| Near the limit, comfortable with the basis | Conservative assumption holds | Optional |
| Near the limit, not comfortable | A judgement about a 1991 endpoint | Under-sink RO, 120 models |
| Want hexavalent specifically | One standard, one format | Under-sink RO only |
| Offered a pitcher for chromium-6 | Different claim | Ask which claim is listed |
| Also want fluoride and lead covered | Hexavalent is the narrowest claim | Filter on it and get all three |
| No figure at all | Nothing to size | Report, or a lab test |
Choose under-sink reverse osmosis if
- You want the hexavalent chromium claim. It exists in no other format here.
- You also want fluoride and lead covered, which every one of these 120 models holds.
- You accept the running cost. EPA notes reverse osmosis uses approximately three times as much water as it treats.
Choose to buy nothing if
- Your total chromium figure sits well below the limit and you are content that the figure is already read as all chromium-6.
- You cannot fit an under-sink system. No certified alternative for the hexavalent claim exists in this listing, and a pitcher chromium claim is a different certification.
Mistakes that cost the most money
Treating the single limit as a regulatory failure. EPA’s stated reason is interconversion, and its rule assumes the whole measurement is the more toxic form. That is more protective than a split limit, not less.
Buying a pitcher for chromium-6. Seven pour-through models hold a chromium reduction claim, and it is not the hexavalent claim.
Assuming reverse osmosis certification covers it. TDS reduction is the only required claim under NSF/ANSI 58. Hexavalent chromium is optional and seven fluoride-certified models do not carry it.
Reading a compliant figure as a speciation result. Your report gives total chromium. It cannot tell you the split, and no household test on the report will.
Waiting for the revised standard before acting, or acting because a revision is coming. Both treat an unfinished review as though it had concluded.
Forgetting membrane and cartridge replacement. A certified claim describes a unit inside its rated service life.
Three situations, and what changes
Your report shows chromium well under the limit. The most common case, and the one where EPA’s conservative assumption does the work for you. Nothing here argues for a purchase.
You live near a site with industrial history and want the hexavalent form addressed. The claim exists, in one format, held by 120 models. Filter on the hexavalent claim specifically rather than on “chromium”, because the two are different certifications.
You were already buying reverse osmosis for fluoride or lead. Check whether the model also lists hexavalent chromium. Most do — 120 of 127 — but seven do not, and the listing is where that is recorded.
The objections worth answering
“There is no chromium-6 standard, so nobody is protecting me.” EPA’s total chromium standard covers it and assumes the entire measurement is chromium-6. The criticism worth making is about the 1991 dermatological basis, not about coverage.
“My utility only tests total chromium.” That is what the regulation requires, and EPA’s stated reason is that measuring one form may miss chromium that is present because the forms interconvert.
“Carbon filters remove heavy metals, so mine must handle it.” Hexavalent chromium reduction is not a selectable claim under NSF/ANSI 42 or 53, so no carbon filter is certified for it under those standards.
“These figures will move.” The NSF counts will. NSF dates its listings and asks readers to confirm status at source, and the capture date is printed beside every figure here. EPA’s page was last updated in March 2026.
Who this is for
Anyone who has seen chromium on a water report, or read about chromium-6 and wants to know what the standard actually covers and what can be bought against it.
It takes no position on whether the 1991 limit should change, and names no recommended product.
What to settle first
What is my total chromium figure, and am I content that it is already read as though all of it were chromium-6?
If yes, there is no purchase to make. If no, the certified answer is one format and 120 models.
→ How to read your water quality report → Which claims exist at all, and how populated each is → Filters certified for fluoride, which the same models hold → Nitrate, the narrowest NSF/ANSI 58 claim of the four → Arsenic: another claim split by species, with a condition attached