Unregulated contaminants — what EPA watches, and what your utility tested anyway

By Andrew Bates, FounderPublished 2026-09-11Updated 2026-09-11

Two EPA mechanisms cover contaminants with no limit: the Contaminant Candidate List, 66 chemicals plus three groups and 12 microbes, and the Unregulated Contaminant Monitoring Rule, which sampled 29 PFAS and lithium from 2023 to 2025. Results reach your annual report — but only above a floor that is not health-based.

There is a formal answer to “what about everything else”

Two mechanisms, both statutory, and most people have heard of neither.

The Contaminant Candidate List is EPA’s register of contaminants “not subject to any proposed or promulgated national primary drinking water regulations but are known or anticipated to occur in public water systems.”

CCL 5, published 14 November 2022, holds 66 chemicals, three chemical groups — PFAS, cyanotoxins and disinfection byproducts — and 12 microbes.

Listing is not a warning. It means EPA treats the contaminant as a priority for regulatory decisions and information gathering.

The second mechanism actually takes samples

The Safe Drinking Water Act requires EPA to publish a list of unregulated contaminants for systems to monitor once every five years.

UCMR 5, published December 2021, required sampling for 30 contaminants — 29 PFAS and lithium — between 2023 and 2025. EPA released the final dataset in August 2026.

Who had to sample	Requirement
Systems serving over 10,000	All of them
Systems serving 3,300 to 10,000	All of them
Systems serving under 3,300	A random sample of 800

So below 3,300 people, most systems have no UCMR 5 result at all. That is missing monitoring, not a clean result.

These results do reach you

The part that surprises people: unregulated monitoring is not secret.

Systems must notify customers of the availability of all UCMR results by Tier 3 public notice within 12 months. Community systems must also report them in the annual Consumer Confidence Report when found, giving the average and the range from the prior year.

→ How to read your water quality report

With one condition that changes everything: results are reported “when contaminants are found (i.e., measured at or above minimum reporting levels).”

The floor is a measurement floor, not a safety floor

This is the sentence to carry away from the whole page.

EPA defines a minimum reporting level as the lowest concentration achievable with 95 percent confidence by at least 75 percent of laboratories. And states plainly that these levels “are not associated with contaminant health effects information.”

Only results at or above it are reported to EPA at all.

	Figure
UCMR 5 reporting floor, 29 PFAS	2 to 20 parts per trillion
UCMR 5 reporting floor, lithium	9 µg/L
EPA’s enforceable limit, PFOA	4.0 parts per trillion
EPA’s enforceable limit, PFOS	4.0 parts per trillion

Part of the reporting floor sits above the enforceable limit. A result can be below what a laboratory will report and above what the law allows, depending on the compound and the method.

So a blank line means “not quantified”, never “not present”.

→ PFAS: the six regulated compounds, and which have a claim

Lithium, as the worked example

UCMR 5’s non-PFAS contaminant shows what unregulated really means.

There is no maximum contaminant level and, EPA says, no health advisory for lithium in drinking water.

The candidate list programme derived a health reference level for screening, from a provisional toxicity value, “based on adverse effects observed in patients administered lithium therapeutically.” EPA states such levels “are not final determinations” and that the science “is still evolving.”

Treatment is equally provisional. EPA states lithium “cannot be removed by heating, boiling, or disinfecting water”, and that bench- and pilot-scale research suggests ion exchange is effective.

NSF offers no lithium claim under any standard. FDA sets no lithium standard for bottled water either.

→ The regulated contaminants with no certified filter

Regulation moves in both directions

Most coverage assumes the list of regulated contaminants only grows.

On 18 May 2026 EPA announced two proposed rules. One would uphold the regulation for PFOA and PFOS, while letting systems request two more years — to 2031 — to comply. The other would rescind the regulations for PFHxS, PFNA, HFPO-DA (GenX) and the Hazard Index mixture of those three plus PFBS.

Both are proposals open for comment. The April 2024 regulation covering all six remains in force until a final rule changes it.

For a household the practical reading is narrow: the two PFAS with the lowest limits and the deepest certified filter coverage are the two the proposal keeps.

→ The 72 filters certified for PFAS

What is still outside both mechanisms

Some things people ask about are on neither list.

Contaminant	Status
Microplastics	No maximum contaminant level; NSF/ANSI 401 claim exists
Hydrogen sulfide	No standard; undrinkable before it is harmful
Hardness	Not a contaminant; no federal standard of any kind
Radon in water	Rule proposed, never finalised
Pharmaceuticals	Not regulated; seven have NSF/ANSI 401 claims
Lithium	CCL 5 and UCMR 5; no limit

An absent standard is not a verdict. It usually means the occurrence data, the health assessment, or both, are not yet judged sufficient to set one.

The decision framework

1. Separate “unregulated” from “unmeasured”

They are different. PFAS were unregulated and heavily measured for years before the 2024 rule.

2. Look up your own system before generalising

UCMR 5 results are published per system through EPA’s Data Finder. National summaries tell you nothing about your tap.

3. Read the reporting floor next to the result

A non-detect is bounded by the method, not by health. For UCMR 5 that bound ran as high as 20 parts per trillion.

4. Do not buy against a candidate listing

A CCL entry sets no limit and implies no concentration. There is nothing to measure a product against.

→ Which claims exist at all, and how populated each is

5. Where a certified claim exists, that is the actionable route

PFOA and PFOS have both limits and certified claims. Lithium has neither.

→ How to check a certification claim yourself

The decision path

Your situation	What decides it	Where that lands you
Worried about a CCL contaminant	No limit, no claim, no measurement	Nothing to act on yet
System serves over 3,300	It had to sample for UCMR 5	Look up its results
System serves under 3,300	Probably outside the 800 sample	No data exists; ask the utility
Saw a UCMR non-detect	Below a non-health-based floor	Note the floor, not just the result
PFOA or PFOS detected	Regulated, and proposed to stay so	Certified filters exist; act on it
GenX detected	Regulated now, proposed for rescission	No NSF claim exists either way
Lithium detected	No limit, no advisory, no claim	Nothing certified to buy
Worried about pharmaceuticals	Not regulated, but certified claims exist	No limit; seven NSF/ANSI 401 claims

Choose to act if

A contaminant with an enforceable limit and a certified claim shows up above that limit. That is a short list, and it is the only one where a purchase can be matched to a number.

Choose to watch if

Your system reported a detection of something unregulated. The next UCMR round or a future regulatory determination may change its status.

Choose to do nothing if

The concern is a candidate-list entry with no occurrence data for your system. There is no limit to exceed and no claim to verify.

→ Testing city water: what your utility already measured

Mistakes that cost the most money

Reading a non-detect as zero. It means below the minimum reporting level, which EPA says is not health-based.

Reading a candidate listing as a health warning. It marks a contaminant for study, not for concern at any particular level.

Buying a filter for an unregulated contaminant. Without a limit there is no threshold to treat to, and usually no certified claim to verify.

Assuming absence of a line means absence of the substance. Unregulated results appear only when detected at or above the floor.

Assuming small systems have quieter water. Most systems under 3,300 people were never required to sample at all.

Assuming regulation only expands. Four of the six regulated PFAS are currently proposed for rescission.

Three situations, and what changes

A large city system, UCMR 5 detections published. The most useful case. The data exists per system, the report must carry detections above the floor, and for PFOA and PFOS there are both an enforceable limit and certified filters to match against a real number.

A village system serving 900 people. Probably outside the 800-system sample, so no UCMR 5 data exists. Nothing about that is reassuring or alarming; it is simply unmeasured, and the utility is the only place to ask.

A household worried about pharmaceuticals in tap water. Not regulated and not on UCMR 5, so there is no number to treat to. But certified products do exist: seven of the fifteen named NSF/ANSI 401 claims are for medicines, and 438 distinct models hold at least one of them. A claim without a measurement is a weaker basis for buying than a claim matched to a figure, which is the honest way to put it.

→ Reverse osmosis: what NSF/ANSI 58 actually certifies

The objections worth answering

“If it were dangerous they would regulate it.” Regulation follows occurrence data and health assessment, and both take years. PFAS sat on candidate lists long before 2024.

“So the candidate list is just bureaucracy.” It is the mechanism that put PFAS and lithium into national monitoring. UCMR 5 exists because CCL 5 named them.

“My report says non-detect, so there is none.” It says none was quantified at or above the reporting level. EPA states that level is not associated with health effects information.

“Unregulated means nobody is looking.” For 30 contaminants between 2023 and 2025, every system above 3,300 people was looking, and had to publish what it found.

“EPA just regulated six PFAS, so this is settled.” Four of the six are currently proposed for rescission. The two with the lowest limits are proposed to stay.

Who this is for

Anyone who has asked what happens to the contaminants nobody sets a limit for, and anyone reading a UCMR line on their annual report and trying to work out what it means.

It is also for anyone being sold a product on the basis that something is “unregulated” — a claim that, on its own, carries no number at all.

What to settle first

Is there a limit, is there a measurement, and is there a certified claim?

Unregulated contaminants usually fail all three. When they fail all three, there is nothing a purchase can be matched against.

→ How to read your water quality report → The regulated contaminants with no certified filter → What EPA says removes what → PFAS: the six regulated compounds, and which have a claim → City water treatment, and the three gaps a utility cannot close → Cyanotoxins: advisory numbers, no limit, and one certified claim → Radon: proposed in 1999, certified since, never regulated → Legionella: regulated, but with no limit and no measurement → Pharmaceuticals in tap water: unregulated, and certified against anyway

The bottom line

Unregulated does not mean unmeasured. EPA keeps a candidate list of 66 chemicals, three groups and 12 microbes, and requires a fresh round of monitoring every five years. But read the floor before the result: a non-detect means below a reporting level that EPA says is not health-based, and for UCMR 5 that floor ran as high as 20 parts per trillion against enforceable PFAS limits of 4.0. And regulation moves both ways — four of the six regulated PFAS are currently proposed for rescission.

Who this is not for

This explains the machinery, not any particular result. It does not tell you what is in your water, because UCMR data is published per system and only systems above a size threshold had to sample. It is not a health assessment of any listed contaminant, and a candidate list entry is not a warning. Nothing here is a reason to buy a product.

Questions

What is the Contaminant Candidate List?

EPA's list of contaminants not subject to any proposed or promulgated national primary drinking water regulation, but known or anticipated to occur in public water systems. CCL 5, published in November 2022, holds 66 chemicals, three chemical groups and 12 microbes.

Does being on the candidate list mean it is dangerous?

No. It means EPA considers the contaminant a priority for regulatory decision-making and information collection. Listing sets no limit, obliges no utility to act, and says nothing about the concentration in any particular water supply.

What is the Unregulated Contaminant Monitoring Rule?

The Safe Drinking Water Act requires EPA to issue a list of unregulated contaminants for public water systems to monitor once every five years. UCMR 5 covered 29 PFAS and lithium, sampled between 2023 and 2025, with the final dataset released in August 2026.

Will unregulated results appear on my water report?

When they are found. Systems must give Tier 3 notice of UCMR results within 12 months, and community systems must report them in the annual report when measured at or above the minimum reporting level, giving the average and range from the prior year.

What does a non-detect actually mean?

Below the minimum reporting level, not absent. EPA sets that level by what 75 percent of laboratories can achieve with 95 percent confidence, and states it is not associated with contaminant health effects information. It is a measurement floor, not a safety threshold.

Is lithium in drinking water regulated?

No. EPA has no maximum contaminant level and no health advisory for lithium. The candidate list programme derived a health reference level for screening only, based on effects seen in patients given lithium therapeutically, and EPA says the science is still evolving.

Can I filter out lithium?

Not with a certified product. EPA states lithium cannot be removed by heating, boiling or disinfecting, and that bench and pilot research suggests ion exchange is effective. NSF offers no lithium claim, and FDA sets no lithium standard for bottled water.

Can a regulated contaminant become unregulated again?

It can be proposed. On 18 May 2026 EPA proposed rescinding the drinking water regulations for PFHxS, PFNA, HFPO-DA and the Hazard Index mixture, while upholding those for PFOA and PFOS. Those are proposals; the 2024 rule stands until changed.

Sources — 13 claims

EPA — Contaminant Candidate List 5 (CCL 5), page last updated 2026-02-27
EPA defines the Contaminant Candidate List as contaminants that are not subject to any proposed or promulgated national primary drinking water regulation but are known or anticipated to occur in public water systems, and which may require future regulation. CCL 5, published 14 November 2022, contains 66 chemicals, three chemical groups — PFAS, cyanotoxins and disinfection byproducts — and 12 microbes. (captured 2026-09-11)
Limits: A candidate list, not a regulation and not a warning. Listing means EPA considers a contaminant a priority for regulatory decision-making and information collection. It sets no limit, obliges no utility to do anything, and implies nothing about concentration in any particular supply.
EPA — Fifth Unregulated Contaminant Monitoring Rule (UCMR 5)
The Safe Drinking Water Act requires EPA to issue a list of unregulated contaminants for public water systems to monitor once every five years. UCMR 5, published 27 December 2021, required sampling for 30 chemical contaminants — 29 PFAS and lithium — between 2023 and 2025, and EPA released the final dataset in August 2026. (captured 2026-09-11)
Limits: Monitoring, not regulation. A UCMR result carries no limit to comply with and no violation to declare. The 29 PFAS were included because the National Defense Authorization Act for Fiscal Year 2020 required EPA to include all PFAS with a validated drinking water method that were not already regulated.
EPA — Fifth Unregulated Contaminant Monitoring Rule (UCMR 5)
EPA required UCMR 5 monitoring from all community and non-transient non-community water systems serving more than 3,300 people, plus a representative sample of 800 systems serving fewer than 3,300. Smaller systems outside that sample were not required to monitor at all. (captured 2026-09-11)
Limits: A household on a small system outside the 800-system sample has no UCMR 5 result to look up, which is an absence of monitoring rather than an absence of the contaminant.
EPA — Fifth Unregulated Contaminant Monitoring Rule (UCMR 5)
EPA requires public water systems to notify customers of the availability of all UCMR results through Tier 3 public notification no later than 12 months after the system knows them, and requires community water systems to report UCMR results in the annual Consumer Confidence Report when contaminants are found at or above the minimum reporting level, giving the average and range of the prior year's results. (captured 2026-09-11)
Limits: Unregulated results reach the annual report only when detected at or above the minimum reporting level. A contaminant monitored and not detected produces no line, so its absence from a report is not evidence it was absent from the water.
EPA — Fifth Unregulated Contaminant Monitoring Rule (UCMR 5)
EPA defines a minimum reporting level as the lowest measurable concentration achievable with 95 percent confidence by at least 75 percent of laboratories nationwide using a specified method, states that only results at or above it are reported to EPA, and states that minimum reporting levels are not associated with contaminant health effects information. The UCMR 5 levels are 0.002 to 0.02 µg/L for the 29 PFAS — 2 to 20 parts per trillion — and 9 µg/L for lithium. (captured 2026-09-11)
Limits: A measurement floor set by what laboratories can reliably achieve, not by what is safe. A result below the minimum reporting level means the contaminant was not quantified, not that it was absent, and the floor sits above EPA's own 4.0 part-per-trillion limits for PFOA and PFOS.
EPA — Fifth Unregulated Contaminant Monitoring Rule (UCMR 5)
EPA states it has no health advisory for lithium in drinking water, and that the Contaminant Candidate List programme derived a health reference level for screening purposes from a provisional peer-reviewed toxicity value, based on adverse effects observed in patients given lithium therapeutically. EPA states health reference levels are not final determinations about the level necessary to protect any particular population, and that the science on lithium's effects is still evolving. (captured 2026-09-11)
Limits: A screening value derived from therapeutic dosing, not a drinking water standard. EPA states there has historically been limited information to evaluate health risks at the levels associated with typical drinking water consumption, which is part of why it chose to monitor.
EPA — Fifth Unregulated Contaminant Monitoring Rule (UCMR 5)
EPA states lithium cannot be removed by heating, boiling or disinfecting water, that available literature based largely on bench- and pilot-scale research suggests ion exchange is effective for removing it and that adsorption with novel media can also be effective, and that the Food and Drug Administration has not established a standard for lithium in bottled water. (captured 2026-09-11)
Limits: Bench- and pilot-scale research is not a certified product claim. NSF offers no lithium reduction claim under any standard, so no household device carries listed lithium performance, and bottled water is not an alternative with a standard behind it.
EPA — Per- and Polyfluoroalkyl Substances (PFAS) drinking water page, last updated 2026-05-18
On 18 May 2026 EPA announced two proposed rules on PFAS in drinking water. One would uphold the regulation for PFOA and PFOS while allowing systems to request two additional years, to 2031, to comply. The other would rescind the drinking water regulations for PFHxS, PFNA, HFPO-DA (GenX) and the Hazard Index mixture of those three plus PFBS. Both are proposals for public comment; the 2024 regulation for all six remains in force until a final rule changes it. (captured 2026-09-11)
Limits: Proposed, not final. The April 2024 regulation covering six PFAS stands, and a proposal may be withdrawn, altered or finalised. EPA's stated rationale is quoted verbatim in the source passage and is the agency's own characterisation, not a finding of this site. Nothing here changes what a household should do about PFOA or PFOS, which the proposal would keep.
EPA — Per- and Polyfluoroalkyl Substances (PFAS) National Primary Drinking Water Regulation
EPA set enforceable Maximum Contaminant Levels of 4.0 parts per trillion for PFOA and for PFOS, and 10 parts per trillion each for PFHxS, PFNA and HFPO-DA (GenX chemicals). (captured 2026-09-10)
Limits: These are enforceable limits for PUBLIC water systems. They do not apply to private wells, which nobody monitors against them. EPA set the health-based MCLG for PFOA and PFOS at zero, meaning no level is considered safe; 4.0 ppt is the enforceable level, not a safe level.
EPA — National Primary Drinking Water Regulations
Microplastics does not appear anywhere in EPA National Primary Drinking Water Regulations table of regulated contaminants, so no Maximum Contaminant Level, health goal or treatment technique is set for it there. (captured 2026-09-10)
Limits: A negative check, not a quotation. The full NPDWR page was fetched on 2026-09-10 and the term microplastic occurs zero times in its 26,067 characters, while known entries such as Chromium (total) and Trihalomethanes are present, which confirms the table itself was retrieved. The absence says nothing about health effects, and EPA may address microplastics elsewhere than in this regulation.
Penn State Extension — Hydrogen Sulfide (Rotten Egg Odor) in Water Wells, by Bryan Swistock, Water Resources Coordinator (updated 2024-12-10)
There is no official drinking water standard for hydrogen sulfide, because the gas makes water aesthetically undrinkable long before it reaches harmful concentrations. (captured 2026-09-09)
Limits: Absence of a standard is not absence of a problem — it means the nose is the limiting factor, not toxicity.
EPA — Secondary Drinking Water Standards: Guidance for Nuisance Chemicals
EPA has established National Secondary Drinking Water Standards setting non-mandatory levels for 15 contaminants, states it does not enforce them, and states that because they are not health threatening at the secondary maximum contaminant level, public water systems only need to test for them on a voluntary basis. (captured 2026-09-11)
Limits: Federal position only. States may adopt secondary standards as enforceable requirements of their own, and many public water systems monitor them voluntarily. A figure absent from a Consumer Confidence Report may simply never have been measured.
Crosswalk of EPA National Primary Drinking Water Regulations against the NSF reduction-claim taxonomy (both captured 2026-09-10)
Crossing EPA 94 regulated contaminant entries against NSF 101 selectable reduction claims on 2026-09-10 leaves 33 regulated contaminants with no NSF reduction claim of that name: 19 organic chemicals, 4 microorganisms, 4 inorganic chemicals, 3 disinfection byproducts and 3 radionuclides. (captured 2026-09-10)
Limits: The absence of a claim by name is not evidence that no product treats the contaminant. Reverse osmosis and distillation remove many of these; NSF offers no named claim to certify them against. NSF offers no claims at all under its distillation standard, so distillation is invisible to this crosswalk, and microorganisms are certified through NSF/ANSI 55 disinfection performance classes rather than by contaminant.