A lab report lands in your inbox. Somewhere on it is a number — colony-forming units per litre, written cfu/L — a species name, and a sample reference you half recognise. The question that actually matters is not “is that high or low” in the abstract. It is: what does this number, from this outlet, taken this way, on that day, tell me to do next?
That is the whole skill. A Legionella count is a piece of evidence, not a verdict. Read on its own it can panic you or falsely reassure you. Read alongside where the sample came from, how it was taken, and what your control records already say, it points to one of three responses: carry on, look harder, or act now.
What follows takes you from the figure on the page to a defensible decision, in the order a busy responsible person can actually work through.
Before the number means anything
Before you reach for a threshold, the report should let you pin down four things.
- The unit and the system. UK action guidance for Legionella in hot and cold water systems is expressed in cfu/L, and the bands commonly cited in HSG274 are tied to that kind of system [1]. Cooling towers, spa pools and other systems are assessed against different expectations, so confirm which one this sample represents first.
- Where and how it was taken. A pre-flush “first draw” from a tap, a post-flush sample, a hot outlet, a cold tank — each tells a different story. BS 7592 sets out sampling practice precisely because location and method shape the count [2]. A high first-draw figure at one tap is not the same finding as a high count across the cold system.
- The organism. Most reports name what grew. Legionella pneumophila — and serogroup 1 in particular — is responsible for the majority of cases, so a positive for that organism carries more weight than an environmental species [3].
- When it was taken. Culture, still the standard method, has to grow the bacteria before they can be counted, which takes time — often around a fortnight [4]. The result describes the system on sampling day, not today. Anything you changed since sits outside the number.
If you can say in one sentence what was tested, where, how and when, you are ready to interpret. If you cannot, that gap is your first action.
Reading a result, step by step
-
Restate the sample in plain terms. Name the outlet or asset, the system type, the sample type (pre- or post-flush, hot or cold) and the date. Done when you can describe the sample without the lab jargon.
-
Check the organism, not just the count. Record whether it is L. pneumophila (and which serogroup) or another Legionella species. Done when the species sits next to the number in your notes.
-
Place the count against the guidance band. For hot and cold water systems, the bands commonly cited in HSG274 are broadly: under 100 cfu/L (the system is in control), 100 to 1000 cfu/L (review and resample), and over 1000 cfu/L (act promptly) [1]. Treat those as guidance to confirm against the current HSG274 text and your own risk assessment, not as fixed pass/fail lines. Done when the figure is located in a band rather than floating on its own.
-
Match the band to the response it expects. A low or not-detected result supports continuing your existing controls. A mid-range result usually means reviewing the system and resampling to confirm before any heavy intervention. A high result usually means prompt investigation and remedial action, then a resample — and if the count or the setting suggests a genuine exposure concern, that tips into the territory covered in Emergency response planning for Legionella incidents. Done when you have written the specific next action, not just the band.
-
Cross-check against your control evidence. Pull the temperature logs and flushing records for that outlet. Does a raised count fit a story you can already see — a tepid outlet, a missed flush, a dead leg — or does it contradict otherwise good records? Done when you can say whether the result confirms or challenges your control picture.
-
Record the decision and close the loop. Write down the interpretation, the action, who owns it, and the resample date. Done when the report, the decision and the follow-up sit together in the log, not in three different places.
What a count cannot tell you
A single number is one outlet at one moment. The trend across several rounds tells you far more: a stable low baseline that suddenly jumps is a signal even when it technically stays inside a band, and a mid-range count that has been creeping upward deserves more attention than a one-off.
“Not detected” is not the same as zero. It means below the laboratory’s detection limit for that sample, not sterile water.
A clean sample is also not a clean bill of health. Sampling verifies control; it does not create it. Control comes from temperature, water movement and cleanliness, and HSE is clear that how often you sample should follow the system and the risk assessment rather than habit [5][6]. If your counts are good but your temperatures wander, you have a result that flatters a system you are not really controlling.
Finally, the count cannot tell you why. It flags that conditions somewhere allowed the bacteria to grow; finding the cause — the lukewarm return, the rarely-used spray outlet, the fouled tank — is a separate piece of detective work, and a changed finding is one of the triggers for revisiting your assessment (When to review your Legionella risk assessment).
A word on limits and judgement
The bands and responses above are widely used guidance, not law you can apply blindfolded. The figures that matter for your building, what an out-of-range result triggers, and the remedial steps that follow all depend on the system, who uses it and the control scheme you already run — which is why a competent person interpreting against the current HSG274 text and your risk assessment will always beat a memorised rule of thumb. And where a result points to a possible health concern, that is a conversation for your responsible person and, where relevant, public health colleagues, not something to settle alone from a lab sheet.
FAQ
My count is in the middle band but my temperatures look fine — do I still act?
Usually yes, at least to the extent of reviewing the outlet and resampling. A mid-range count that contradicts good temperature records is a flag that something else is going on — a low-use fitting, a problem upstream of where you take readings, or a sampling issue — and a confirmatory resample tells you which before you spend money on remediation.
Does a higher cfu/L always mean a more dangerous system?
Not on its own. The organism matters — L. pneumophila serogroup 1 weighs more heavily than an environmental species — and so does where the figure sits in a trend and whether the outlet actually produces a breathable aerosol. A modest count at a heavily used shower can warrant more attention than a larger count at a tap nobody inhales from.
Why did my result take two weeks to come back?
Standard culture testing has to grow the bacteria from the sample before they can be counted, which simply takes time [4]. That lag is normal, and it is the reason interpretation always describes the system as it was on sampling day. Keep your temperature and flushing controls running while you wait rather than treating the pending report as a pause button.
Sources
[1] HSE, “Legionnaires’ disease: Technical guidance (HSG274)”. https://www.hse.gov.uk/pubns/books/hsg274.htm [2] 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 [3] CDC, “About Legionnaires’ Disease”. https://www.cdc.gov/legionella/about/index.html [4] CDC, “Laboratory Testing for Legionella”. https://www.cdc.gov/legionella/php/laboratories/index.html [5] HSE, “Testing and monitoring your water system for legionella”. https://www.hse.gov.uk/legionnaires/testing-monitoring-water-system.htm [6] 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