Best Water Filter for Bacteria & Coliform (2026)

We recommend products we believe in. When you buy through links on this page we may earn a commission, which never changes what we recommend or what you pay. See our affiliate disclosure for details.

A lab report that says "total coliform: present" sends most people straight to a search box for a water filter. That instinct leads almost everyone to the wrong product, because the thing sold as a water filter and the thing that deals with bacteria are not the same category.

Here is the short version. Sediment filters, carbon blocks, KDF stages, and filter pitchers are built to catch particles and adsorb dissolved chemicals. Bacteria are roughly 0.5 to 5 microns across, and a typical carbon block is rated at 5 or 10 microns nominal, which means most of them sail through. To deal with coliform you need disinfection, a genuinely absolute membrane, or reverse osmosis, and the first of those is the only one that protects the whole house.

This guide covers the three that work, in the order they make sense for a household, and it spends time on one specific trap we think costs buyers real money: the NSF/ANSI 55 Class A and Class B distinction.

Total coliform and E. coli are not the same finding

Before you buy anything, read your lab report properly, because the two lines people conflate call for different responses.

Total coliform is a broad group of bacteria that live in soil, on vegetation, and in surface water. Most of them will not make you ill. Their value is as an indicator: coliform in your well means a path exists from the surface into your water, and anything else at the surface can use that same path. A positive is a signal about the integrity of your well, not necessarily a toxin in your glass.

E. coli is a subset that lives in the intestines of warm-blooded animals. It does not occur naturally in soil the way other coliforms do. A positive E. coli result means fecal contamination, from a septic field, livestock, or runoff, and it means stop drinking the water now and call your health department.

The EPA recommends testing a private well annually for total coliform bacteria, nitrates, total dissolved solids, and pH, and notes that when contamination is found you should contact your public health department and have the well re-tested to confirm the result. Confirming matters. Sampling technique produces false positives often enough that re-testing before buying a $1,400 system is simply good economics. Our guide to how to test your tap water covers certified labs and how to pull a sample that does not contaminate itself.

What actually inactivates bacteria

Two things fall out of that table. The top three rows are what most people already own and what most people search for, and none of them do anything here. And ultraviolet is the only row that is both effective against all three classes of pathogen and capable of covering every tap in the house, which is why it dominates our picks.

The NSF/ANSI 55 trap worth $435

This is the part we most want you to take away, because it is counterintuitive and it has a price tag.

Ultraviolet systems are certified under NSF/ANSI 55, and that standard has two classes. Class A systems deliver 40 mJ/cm2 and are intended to disinfect water that is microbiologically unsafe. Class B systems deliver 16 mJ/cm2 and are intended only for supplemental bactericidal treatment of water that has already been deemed acceptable by a local health authority. Class B is explicitly not for disinfecting unsafe water.

Now look at what that means on a shelf. Viqua sells both certified and uncertified versions of the same chambers. We pulled every residential model from one authorized specialist retailer on the same day, 5 October 2026:

A buyer comparing those four lines sees two with a third-party certification and two without, and the certified ones are $435 and $494 cheaper. The badge points at the wrong product. The Class B units are genuinely certified, to a standard written for water that is already safe, which by definition is not the water of anyone reading a page about coliform.

The uncertified units are not a lesser product. They are rated by Viqua at the Class A dose of 40 mJ/cm2, and Viqua is explicit that this figure comes from its own internal testing using MS2 phage as a surrogate organism rather than from third-party validation. That is a real limitation and we are not going to paper over it. But a manufacturer-rated 40 mJ/cm2 is a better match to a coliform positive than a third-party-certified 16 mJ/cm2, and if you want genuine Class A validation in this family you are leaving the residential range for commercial equipment at commercial prices.

Our top picks at a glance

If a lab found coliform or E. coli in a house of roughly three to five bathrooms, the Viqua VH410-F12 is the system we would install, because it pairs the 40 mJ/cm2 dose with the pre-filtration ultraviolet actually needs.

If the house has one to three bathrooms, the Viqua VH200-F12 is the same design in the right size, and buying the bigger chamber buys flow you will never draw.

If your contamination risk includes viruses, from a failing septic field or surface flooding, the MSR Guardian is the only portable unit here tested to the military P248 protocol.

If you need safe drinking water this evening while you shock the well and wait on a re-test, the LifeStraw Go bottle is the cheapest honest answer.

And if you want a bacteria barrier for a few dollars that will outlive the problem, the Sawyer Squeeze is rated for 100,000 gallons on a 0.1 micron absolute membrane.

Best Overall for Coliform & E. coli: Viqua VH410-F12

The VH410-F12 is three things bolted to one rack: a 5 micron sediment cartridge, a 20 inch carbon block, and a 304 stainless steel chamber holding a Sterilume-HO low pressure lamp. Water passes the filters first, then the lamp. Viqua rates the chamber at no less than 40 mJ/cm2 for flow up to 14 GPM, 30 mJ/cm2 up to 18 GPM, measured at the end of the lamp's 9,000 hour life and at 95% UV transmittance. It lists at $1,740 and sells for $1,395.

The pre-filters are the part people try to economise on, and they are not an accessory. Ultraviolet inactivates an organism by delivering light to it, so anything that shades the organism defeats the lamp. Silt, turbidity, and iron all absorb and scatter UV. A bare UV chamber on raw well water is a lamp burning behind a curtain.

Why it wins: it is the combination that matches the problem. You get the dose that NSF/ANSI 55 Class A asks for, the pre-treatment that makes that dose real at the far end of the chamber, and coverage of every tap in the house rather than one. Viqua states its UV systems inactivate pathogenic E. coli, campylobacter, legionella, salmonella, and shigella, along with Giardia and Cryptosporidium. The Safety-Loc connector cuts power before the lamp can be withdrawn, which matters on a part you will replace annually for a decade.

Tradeoffs: it is not itself certified to NSF/ANSI 55, and Viqua's efficacy data is internal MS2 phage testing rather than third-party validation. Running cost is real at roughly $273 a year, about $160 for the lamp and $113 for the sediment and carbon cartridges. It needs permanent power, so a power cut is an untreated-water event, and most people will pay a plumber. It disinfects and nothing else: iron, hardness, arsenic, and nitrates pass through untouched.

Best for: private wells with a confirmed coliform or E. coli result, in homes of roughly three to five bathrooms, where you want every tap covered rather than a jug on the counter.

Best for 1-3 Bathroom Homes: Viqua VH200-F12

The VH200-F12 is the same architecture on a shorter chamber. Same rack, same 5 micron sediment and carbon block stages, same lamp technology, a chamber 18.4 inches deep instead of 28.2. Viqua rates it at no less than 40 mJ/cm2 up to 7 GPM and 30 mJ/cm2 up to 9 GPM. It lists at $1,540 and we found it at $1,235 and $1,309 at two retailers.

Why it wins: UV dose and flow rate trade against each other directly, because the faster water moves past the lamp the less light each organism receives. A system sized for 18 GPM in a house that peaks at 8 is not safer, it is just larger and more expensive. For one to three bathrooms this delivers the identical 40 mJ/cm2 dose where it counts, runs about $233 a year instead of $273, and fits in places the 28 inch chamber will not.

Tradeoffs: read the price gap carefully before you choose on size alone, because it is smaller than it looks. At the prices we found, the VH200-F12 saves roughly $160 against a system with more than double the rated capacity, so if your household is near the boundary or you expect to add a bathroom, the larger unit is the better buy. 7 GPM at the full dose is genuinely tight if two showers and a washing machine overlap, and when you exceed it the system does not stop, it simply delivers less dose per organism. It carries the same absence of NSF/ANSI 55 certification.

Best for: cottages, seasonal properties, and one to three bathroom houses with a confirmed coliform result and a realistic peak draw under 7 GPM.

Best When Viruses Are a Concern: MSR Guardian

Membrane filters are rated by pore size, and that number decides what they can and cannot stop. Bacteria are large enough that a 0.1 or 0.2 micron membrane blocks them outright. Viruses are one to two orders of magnitude smaller and go straight through. The MSR Guardian closes that gap with a 0.02 micron hollow fiber matrix, built and tested to NSF Protocol P248, the military standard, and it is rated to remove 99.99% of viruses including rotavirus, hepatitis A, and norovirus alongside 99.9999% of bacteria and 99.9% of protozoa.

What makes it usable rather than merely capable is the self-cleaning mechanism. A share of every pump stroke is diverted to backflush the membrane, so it does not clog and never needs scrubbing in the field. It pumps around 2.5 litres a minute and the cartridge is rated to 10,000 litres. It runs about $350.

Why it wins: if your coliform positive came with an E. coli positive, the contamination source is fecal, and fecal contamination carries viruses that no 0.2 micron filter will stop. This is the one portable device here that addresses that. It is also the device that still works when the power is out and your UV lamp is dark.

Tradeoffs: it is heavy, bulky, and expensive for what it treats, which is a pot of water at a time rather than a house. It is a pump, so somebody has to work it. For a household whose only finding is total coliform with no E. coli, it is more purifier than the situation calls for, and the money is better spent on the UV system.

Best for: fecal contamination, flood recovery, and serious preparedness, as a companion to a whole-house system rather than a replacement for one.

Best for Drinking Water Today: LifeStraw Go

There is usually a gap of days or weeks between a positive test and a working UV system, covering the re-test, the shock chlorination, and the install. The LifeStraw Go is what we would put on the kitchen counter during it. It is a 22 ounce bottle with a 0.2 micron hollow fiber membrane and a carbon capsule, tested to NSF Protocol P231, and rated to remove 99.999999% of bacteria including E. coli and salmonella and 99.999% of parasites including Giardia and Cryptosporidium. The membrane is rated for about 1,000 gallons. It costs around $45.

Why it wins: it is immediate, it needs nothing but the bottle, and the claims are backed against a recognised microbiological protocol rather than a nominal micron rating. P231 is the protocol written for exactly this case, water of unknown microbiological quality.

Tradeoffs: it is 22 ounces at a time, so it is a drinking-water bridge and not a household solution. It is a filter and not a purifier, so viruses pass. It does nothing for lead, arsenic, nitrates, or fluoride, and it does not help you shower. Treat it as the thing that gets you through the fortnight, not the answer.

Best for: drinking and cooking water while you disinfect the well and wait on a re-test, and as the thing you keep for the next boil-water notice.

Best Budget Bacteria Barrier: Sawyer Squeeze

The Sawyer Squeeze is a palm-sized 0.1 micron absolute hollow fiber membrane that threads onto the included pouch, a standard bottle, or an inline hose. It meets NSF Protocol P231 and EPA filtration guidelines, removes 99.99999% of bacteria and 99.9999% of protozoa, and Sawyer rates the membrane to 100,000 gallons behind a lifetime warranty. It costs around $39 and the included syringe backflushes it indefinitely.

Why it wins: 0.1 micron absolute is a finer barrier than the LifeStraw Go at a lower price, and the economics are absurd in the buyer's favour. Roughly $3 a year in practice, for a device you will still own in a decade. If you want one bacteria barrier in a drawer against the next notice, this is the rational purchase.

Tradeoffs: it is a filter, not a purifier, so viruses pass. It removes nothing dissolved, so lead, nitrates, and fluoride are untouched. The included pouches split eventually. Flow drops noticeably if you skip backflushing, and squeezing a pouch for a household's drinking water gets old faster than the filter does.

Best for: the cheapest honest protection against bacteria and protozoa, kept as a standby rather than used daily.

How to choose the best water filter for bacteria

Confirm the result before you spend. Coliform sampling produces false positives when the tap, the bottle, or the handling contaminates the sample. Pull a second sample properly and send it to a state-certified lab. A confirmed positive changes what you buy; an unconfirmed one may change nothing.

Shock chlorinate and re-test before you install anything permanent. A UV lamp treats water flowing past it and reaches nothing behind it, so bacteria living in the casing, the pump, or the pipework survive it. Shock the well, flush it thoroughly, wait, and test again. If it clears and stays clear, you had a one-off. If it goes positive again, you have a structural entry point, and now you know the UV system is justified rather than hoped for.

Find the entry point while you are at it. The usual suspects are a cracked or missing well cap, a casing that does not extend far enough above grade, a septic field too close or failing, and surface water pooling at the wellhead. Treatment that ignores the breach is a subscription to the problem.

Size on your real peak flow, not your bathroom count. Dose and flow trade directly. Work out what actually runs at once, add margin, and pick the chamber that still delivers 40 mJ/cm2 at that number. Over-sizing wastes money and under-sizing quietly lowers your dose without any warning light.

Put UV in the right place in the train. Sediment first, then iron and sulfur treatment if your test calls for it, then softening, then UV, and finally point-of-use reverse osmosis for drinking water. UV goes near the end because it needs water the light can get through. Our roundup of the best water filter for well water covers the full train, and if you are weighing RO for the dissolved contaminants UV cannot touch, how reverse osmosis works explains what the membrane does and does not do.

Plan for the lamp, not just the box. A UV lamp dims long before it dies, which is why manufacturers rate 9,000 hours rather than "until it goes out". A lamp left in for three years is still glowing and no longer disinfecting. Put the replacement on a calendar the day you install it.

Decide what happens when the power goes out. A UV system is dark during an outage and for a while after, until the lamp warms and the standing water is flushed. If you are on a well with a history of positives, keep a membrane filter or a gravity system that needs no power. Our guide to the best gravity water filter covers the no-power options.

How we evaluated these systems

We started from dose and pore size rather than marketing language, because in this category those two numbers are the product. For the ultraviolet systems we took flow rates at 40, 30, and 16 mJ/cm2 from Viqua's own published specifications, and we read the manufacturer's efficacy footnote rather than the headline, which is how we know the pathogen claims rest on internal MS2 phage testing rather than third-party validation. We have said so on every pick where it applies.

We verified prices at two independent water treatment retailers on 5 October 2026 and found matching manufacturer list prices at both, which is also how we assembled the certified-versus-uncertified price table above. We checked stock status rather than assuming it. We do not have Amazon pricing API access, so the figures here are retailer and manufacturer sourced and may differ from what you see at checkout.

For the portable units we used the microbiological protocols, NSF P231 and P248, in preference to micron ratings, because a nominal micron figure is close to meaningless and an absolute one tested against a protocol is not. We did not run our own microbiological assays and we do not claim to.

We deliberately left out the large category of filters marketed to well owners that do nothing about bacteria. Several would have made this page look better stocked. A sediment and carbon system sold against a coliform result is not a budget option, it is the wrong product, and listing it as a cheaper pick would have been the most harmful thing we could do on this page.

A coliform positive is a solvable problem and an urgent one. Confirm the result, shock the well, find how the surface is getting in, and then size disinfection to the water that is left. For the rest of what a well test turns up, browse our full library of best water filter guides, or start with our water testing and treatment guides.