A Legionella certificate that reads “not detected” is one of the most misread lines in water hygiene. It feels like a pass. It is closer to a single photograph of a single bottle of water, drawn from one outlet on one morning, then processed by a method that only counts the cells willing to grow on a plate. Reassuring, often. Proof that your system is under control, no.

If you already run a written scheme and read sample reports, you can skip why we sample at all. The harder question is the useful one: how does a genuinely colonised system hand back a clean result, and what do the alive-but-invisible cells behind that gap mean for the decisions you put your name to?

Why culture is the reference method — and what “reference” quietly leaves out

The standard way to count Legionella in water is culture. A UKAS-accredited lab spreads the sample on selective media, incubates it, and counts the colonies that appear [5]. It has been the reference for decades, and for good reason: a colony is a living organism the lab can confirm and identify.

Its strength is also its blind spot. Culture only finds cells that will multiply into visible colonies under laboratory conditions. Anything alive but unwilling to grow that week is, by definition, not in the count. The method is slow, too — it needs days of incubation before a result is confirmed, so the certificate always describes the water as it was when sampled, never as it is when the report reaches your desk.

How a contaminated system returns a clean result

A false negative rarely means the lab slipped. More often the system, the sample or the timing conspired to hide what was there.

  • The snapshot problem. Legionella lives in the biofilm coating pipe walls, tanks and fittings, and it sheds into the water unevenly [6]. A bottle taken at a sentinel tap can completely miss a colonised dead leg three metres up the run.
  • Residual biocide in the bottle. Sample soon after a chlorine dose or a hot flush and, unless the lab’s neutraliser fully quenches it, leftover biocide keeps knocking the bacteria back all the way to the plate. BS 7592 sets out neutralisation and representative sampling precisely because technique decides the number as much as the water does [4].
  • Overgrowth. Real water carries plenty of other organisms. On a crowded plate, faster-growing competitors can swamp Legionella before it forms countable colonies.
  • Below the line. Low concentrations can sit under the method’s detection limit and report as nothing found — present, but not quite enough to register.
  • Cells that have stopped growing on purpose. The biggest gap, and one that earns its own heading.

Viable but non-culturable: alive, still a hazard, invisible to the plate

Put Legionella under stress — chlorine, heat, starvation, copper-silver dosing — and it can drop into a dormant state in which it stays viable but stops forming colonies on standard media. This is the viable-but-non-culturable, or VBNC, state. Culture scores those cells as absent. They are not.

That distinction matters because dormant is not dead. VBNC cells can later resuscitate, and there is good evidence they do so sheltered inside the free-living amoebae that share the same biofilms. So a hot flush or a slug of biocide can drive the culture count down to “not detected” without clearing the organism. The paperwork looks controlled while a reservoir quietly persists, ready to recover once conditions ease.

PCR doesn’t fix it — it swaps one error for the other

The obvious answer is a faster, more sensitive method, and PCR (qPCR) is it: detect Legionella DNA directly, get a result in hours, and pick up the VBNC and damaged cells culture overlooks. The catch is that DNA outlives the cell. PCR can read strongly positive on a system you have just successfully disinfected, because it is counting genetic material from dead bacteria alongside the living.

So the two methods fail in opposite directions. Culture under-reports by ignoring everything that will not grow; PCR over-reports by counting everything that ever lived. Neither is a verdict on its own, and reading either number as a flat yes or no is how people get this wrong. The two are compared head to head in Detecting Legionella: culture and PCR testing methods.

What a clean certificate doesn’t actually tell you

If you take one thing from these testing limitations, take this: the lab tested the bottle, not the building. A few consequences fall straight out of that, and they are the ones generic guidance tends to skip.

  • Your own controls can manufacture the negative. The cleaner a result looks straight after a flush or a dose, the more you should ask when it was taken. A suppressed count is not the same as a cleared system.
  • “Not detected” is a statement about culturability, not presence. It means nothing grew from that bottle under those conditions. No more, no less.
  • Chain of custody decides as much as water quality. Sample point, time of day, neutraliser, transport temperature and delay to the lab can each turn a true positive into a clean sheet.
  • A result is verification, never the control itself. Sampling supports your scheme; it does not replace temperature management, flushing and cleanliness — and under HSE guidance the frequency is set by your risk assessment, not chosen to produce comfortable numbers [3].

A short caveat, because this is a topic people over-read: the intervals, limits and remedial actions for your site live in your written scheme and risk assessment, decided by a competent person who knows the plant — not in an article. Nothing here changes what your accredited lab reports or how a clinician diagnoses a case. Use it to interpret certificates with appropriate suspicion, then confirm the specifics against L8, HSG274 and BS 7592 for your own system [1][2][4].

What to do with a “clean” result this week

Pull your last few certificates and check three things against the dates in your logbook. Was each sample taken before or after a flush, dose or temperature intervention? Does the sample point genuinely represent the worst-served parts of the system, or the most convenient tap? And does the clean count sit alongside good temperature and flushing evidence, or is it carrying the whole argument on its own?

If a negative result is the only thing standing between you and “controlled”, it is doing a job it was never designed to do. Strengthen the control evidence first, and let sampling do what it is good at: confirming, and occasionally surprising you. For where verification belongs inside a control programme, see Proactive vs reactive: the next step in Legionella control.

FAQ

Does a “not detected” Legionella result mean our water is safe?

No. It means nothing culturable grew from that sample, at that outlet, on that day. It cannot speak for outlets you did not sample, for cells in a viable-but-non-culturable state, or for the system a week later. Read it alongside your temperature and flushing records, not instead of them.

Should we switch from culture to PCR to avoid false negatives?

PCR is faster and catches cells culture misses, but it also detects DNA from dead bacteria, so it tends to over-report live risk — especially soon after disinfection. Use it as a complement for speed or investigation, interpret it with your lab, and do not treat either method as a standalone pass or fail.

Our sample came back clean but a case was linked to the building — how?

It happens, and the reasons are the ones above: the sample point missed the colonised part of the system, the timing suppressed the count, or viable cells simply did not grow on the plate. Treat it as a prompt to investigate properly rather than evidence of a lab error; UKHSA guidance sets out how linked cases are worked through [7].

Sources

[1] HSE, “Legionnaires’ disease. The control of legionella bacteria in water systems - Approved Code of Practice and guidance (L8)”. https://www.hse.gov.uk/pubns/books/l8.htm [2] HSE, “Legionnaires’ disease: Technical guidance (HSG274)”. https://www.hse.gov.uk/pubns/books/hsg274.htm [3] HSE, “Testing and monitoring your water system for legionella”. https://www.hse.gov.uk/legionnaires/testing-monitoring-water-system.htm [4] BSI, “BS 7592:2022 - Sampling for Legionella bacteria in water systems. Code of practice”. https://knowledge.bsigroup.com/products/bs-7592-sampling-for-i-legionella-i-bacteria-in-water-systems-code-of-practice-1 [5] CDC, “Laboratory Testing for Legionella”. https://www.cdc.gov/legionella/php/laboratories/index.html [6] CDC, “How Legionella Spreads”. https://www.cdc.gov/legionella/causes/index.html [7] UKHSA, “Investigation of Legionnaires’ disease: cases, clusters and outbreaks”. https://www.gov.uk/government/publications/investigation-of-legionnaires-disease-cases-clusters-and-outbreaks