Culture Takes Days: UK Legionella Testing Methods, BS 7592 Sampling
For compliance, rely on culture-based testing to BS EN ISO 11731 carried out by a UKAS-accredited laboratory, with samples taken to BS 7592. Use qPCR only as a rapid adjunct, because it can detect DNA from dead organisms. Follow HSG274 for the wider procedures, and expect cooling towers to need periodic Legionella analysis, usually quarterly.
TL;DR:
Treat results above 100 cfu/l as a trigger to review controls and resample; above 1,000 cfu/l, disinfect, investigate the failure, and confirm recovery.
Increase sampling for complex systems, vulnerable occupants, control failures, or recent changes; after remedial work, repeat tests until control is restored.
Select nearest and furthest outlets on each loop, then note whether samples are first draw or post flush and whether a mixing valve is involved.
Use sterile 1 litre containers with 180–200 mg sodium thiosulphate, keep samples cool, and deliver them within the laboratory’s accepted transport window.
Healthcare guidance calls for further steps above 100 cfu/l; repeatedly high results, vulnerable areas, or linked illness warrant prompt investigation and public health advice.
Table of Contents
How do culture, qPCR and field tests differ?
Each method answers a different question, so the choice depends on whether you need a compliance record, a fast signal or a check on your biocide regime. HSE states that the industry standard is the culture-based method following BS EN ISO 11731:2017.
Culture analysis: the lab concentrates the water sample, plates it on selective media, incubates it, counts the colonies and reports a result in colony forming units per litre (cfu/l). It measures organisms that can grow, which is why it underpins compliance decisions. Results take days, because the bacteria grow slowly.
qPCR: this detects Legionella nucleic acid, so it is fast and sensitive. The catch is that DNA from dead organisms also counts. A positive qPCR result tells you something may be present, not that viable bacteria are. HSE and UKAS position qPCR as useful for rapid intervention, but not a replacement for culture unless the method is validated.
Field tests and dip slides: these indicate general microbial activity and whether a biocide is doing its job. They are quick and cheap, and they suit weekly checks on systems such as cooling towers. They do not identify Legionella, so they complement culture and never stand in for it.
A practical way to hold this together: culture gives you the defensible number, qPCR gives you an early warning, and dip slides tell you between lab visits whether your control regime is still holding. If you rely on only one, you may miss other important aspects of control.
Where should you sample, and how do you choose sentinel points?
Good results start with good sample sites. BS 7592:2022 separates routine sampling from incident sampling and guides the selection of sentinel points versus intermediate outlets. Sentinel points are usually the outlets nearest to and furthest from the tank or heater on each loop, since they show how the system behaves at its extremes. If you want the risk assessment thinking behind that choice, our essential guide to water risk assessment for Legionella covers it.
Work through the sampling decisions in this order:
Decide whether the sample is routine or incident-driven. The answer changes how many points you sample and how you read the results.
Pick sentinel points on each principal loop. Include the nearest and furthest outlets, and any dead legs or little-used extremities you know about.
Choose the sample type for the question you are asking. An immediate first-draw sample reflects the outlet itself, a post-flush sample reflects the system behind it, and tank, shower and biofilm or sediment samples target specific reservoirs of risk.
Record where you took each sample. Public health sampling guidance recommends preparing a schematic of the system and noting locations and temperatures.
Use the right container. The same guidance describes sterile 1-litre containers holding 180–200 mg of sodium thiosulphate, which neutralises chlorine so it cannot kill bacteria in transit.
Take duplicate subsamples where legal evidence may be needed, and check your laboratory’s own instructions before you go on site.
If a sample is taken through a thermostatic mixing valve, say so on the paperwork. The same outlet can give very different results with and without the valve in line.
How often should you sample? Let the risk assessment decide
Sampling frequency is not a fixed national number for every building. It comes from your risk assessment, which should follow BS 8580-1 and HSG274. Testing is a verification tool for your control scheme, and the risk assessment tells you how hard to verify. Our guide to reviewing a Legionella risk assessment shows how to keep that document current, and our article on how often to test for Legionella in England gives scheduling examples.
Factors that push frequency up:
Complex system types, such as large hot and cold networks with many loops and long pipe runs
Vulnerable occupants, such as hospital patients or residents in care settings
Any history of control failures, such as temperatures drifting out of range or biocide dosing lapses
Recent changes, such as a new tank, a refurbished area or a period of low occupancy
Typical patterns you can plan around:
Cooling towers: HSG274 Part 1 recommends weekly monitoring of biocide effectiveness, commonly by dip slides, and periodic Legionella analysis, usually quarterly.
Healthcare and other high-risk sites: these may need more frequent sampling, or a specified action algorithm, because the consequences of a missed problem are greater.
After incidents, disinfection or remedial work: run an intensified programme with short-term repeat sampling until you can show control has been restored.
Write the schedule into your written control scheme so that sampling is never left to memory or goodwill.
What do laboratory results mean, and what should you do next?
A number on a report only means something when you attach an action to it. Commonly used guidance bands, applied in practical HSG274-style schemes, are these:
Not detected, or up to 100 cfu/l: the system is generally regarded as under control. Keep monitoring and continue your routine programme.
Above 100 and up to 1,000 cfu/l: review your control measures, check temperature and biocide records, and repeat the sampling. Healthcare guidance uses an algorithm in which counts above 100 cfu/l trigger further steps.
Above 1,000 cfu/l: treat this as a prompt for remedial action. That typically means disinfection, a review of why control failed and resampling to confirm recovery.
Context changes how you read any of these. A first-draw sample at a shower is not the same as a post-flush sample, and a result taken through a thermostatic mixing valve needs the valve noted. Temperature and biocide records from the same day often explain a result better than the count alone. Our walk-through on how to interpret Legionella water samples works through examples.
Escalate to a detailed investigation when results are repeatedly high, when they appear in a vulnerable-occupant area, or when anyone linked to the premises is ill. Consider disinfection at once in those cases, and involve your local public health authority rather than waiting for further lab work.
How do you choose a laboratory and read its report?
Ask for a UKAS-accredited laboratory for culture testing. HSE guidance expects such laboratories to apply a low theoretical detection limit for culture methods, so ask what theirs is for your sample type. Confirm that the method follows BS EN ISO 11731:2017, or a validated and documented in-house method. The UKAS technical bulletin on water microbiology notes that labs using “based-on” or in-house variations may need to re-verify methods or adjust the wording of their accreditation schedule. Our UKAS check shows how to confirm this quickly.
Questions worth putting to any lab:
Which media and procedures do you use for different sample matrices, such as clean water, tank sediment or biofilm?
Do you take part in proficiency testing schemes?
How do you handle samples that arrive warm or outside your transport window?
When the report arrives, check these items:
The result in cfu/l, with the detection limit stated
The sample type and location, matching your own records
Any serogrouping or typing performed
The temperature readings recorded at sampling
Any method caveats, such as inhibited samples or delayed receipt
A report that omits these is harder to defend if it is ever challenged.
What should be on your on-site sampling checklist?
A tidy kit and a fixed routine prevent most avoidable errors. Treat the following as a minimum:
Sterile 1-litre containers with sodium thiosulphate (confirm the lab’s requirements first)
A cool box and a calibrated thermometer
Alcohol wipes, gloves and any other PPE your risk assessment calls for
Labels, request forms and chain-of-custody paperwork
The sampling sequence runs like this. Label the containers before you start. Take the immediate first-draw sample, then the post-flush sample where required. Collect tank and shower samples with the appropriate technique, and record the water temperature at the time of each one.
Afterwards, keep the samples cool, complete the request forms accurately, tell the lab about any unusual matrices, and deliver within the laboratory’s accepted transport time. A sound reference on keeping specimens intact is this guide to sample handling procedures for clinical lab professionals.
Pro Tip: Photograph each outlet and its label before you sample. If a result is ever questioned, you can show exactly which tap, and in what state, the sample came from.
When is qPCR useful, and what do reference laboratories do in an outbreak?
qPCR earns its place as a triage tool. It returns results quickly, so it helps you decide whether to act while culture is still incubating. Its weakness, as noted above, is that it can report DNA from organisms that are no longer viable. Pair it with culture, and treat a positive as a reason to investigate rather than as a final verdict.
Outbreaks are a different job from routine monitoring. The national reference service, RVPBRU, provides confirmatory real-time PCR, culture and sequence-based typing for clinical and environmental specimens, and it asks for paired urine and lower respiratory tract specimens where possible. Outbreak investigation guidance describes how NLRL and RVPBRU use molecular and sequence-based typing to link clinical isolates to environmental sources.
Do not hold back public-health action while typing is pending. Act on provisional positive results and follow the advice of your local health protection team.
Are there testing methods beyond culture and qPCR?
Yes, though none has displaced culture for compliance. Immunoassays and enzyme-linked approaches aim to detect Legionella antigens or related markers rather than growing the organism or amplifying its DNA. Their appeal is speed and the potential for simpler on-site use. Their limits are the questions a dutyholder has to ask first: has the method been validated for water matrices, can it quantify in a way comparable to cfu/l, and will it be accepted as evidence in an HSE-style compliance conversation?
For now, treat any alternative method as a screening or investigative aid. The compliance benchmark in HSE and UKAS guidance remains culture to BS EN ISO 11731:2017, and the UKAS bulletin shows that regulators expect method claims to be verified. Before you pay for anything newer, ask the supplier for validation data on your type of water, the accreditation status of the method, and what action their result would justify under your control scheme. If they cannot answer those three, keep culture as your record and use the newer tool, if at all, only as a supplement.
How do test results feed into your control programme?
Testing only has value when it changes what you do. Your written control scheme, which HSG274 Part 3 says should be tailored to the specific system, should state purpose, scope, risk assessment, monitoring, inspection and remedial action plans, plus an incident plan for high microbial activity or outbreak. Results are the feedback loop that tells you whether those parts are working.
The method you choose shapes the response you can make. Culture results are slow but defensible, so use them to verify that temperature and biocide controls are delivering and to trigger formal remedial action. Rapid results and dip slides are fast but less specific, so use them to prompt early checks rather than to close an issue. In practice:
Write the action for each result band into the scheme, with a named person responsible
Link every sample result to that day’s temperature and biocide records
Review trends, not just single results, at each risk assessment review
Record what you did after each adverse result and the follow-up sample that confirmed recovery
Risk assessment under BS 8580-1 is the foundation, and testing sits on top of it as verification. A programme that samples diligently but ignores temperature control is only documenting its own failure.
Which testing method suits your system type and water use?
Match the method to the risk the system presents.
Cooling towers and evaporative systems run warm, hold biocide regimes and can spread aerosols widely. Use weekly dip slides to check biocide effectiveness, quarterly culture analysis as HSG274 Part 1 describes, and consider qPCR for fast triage if a dip slide or inspection raises concern.
Potable hot and cold water systems depend mainly on temperature control, so sentinel-point temperature records matter as much as sampling. Take culture samples at sentinel outlets as the risk assessment directs, with tanks, showers and any thermostatic mixing valves included where relevant. Our water analysis and Legionella testing service covers potable testing alongside Legionella work.
Healthcare and care settings need tighter schedules and a defined action algorithm because occupants are more vulnerable, and the healthcare detection guidance shows how sampling type and clinical vulnerability shape the response.
Closed heating and chilled systems are lower risk for Legionella but can still be tested for general microbiology and chemistry where the risk assessment calls for it. When in doubt, begin with culture and let the risk assessment justify any addition.
Author perspective: testing verifies your controls, it does not replace them
The most expensive mistake I see is treating a clean result as proof that a system is safe. Testing confirms that your temperature and biocide regime worked on the day you sampled. It cannot substitute for that regime, so put your effort into an accurate risk assessment first.
The pitfalls are depressingly consistent: wrong containers, samples taken through a thermostatic mixing valve without a note, missing temperature records and unaccredited labs. If you want a sampling plan or training built around your site, ask us.
— Sammi
How we support your sampling, testing and remediation
We turn the guidance above into a plan you can run. We write bespoke Legionella risk assessments, take on-site samples and help you act on the results. Our water analysis and Legionella testing service starts from £55 per sample.
[
Request a bespoke sampling plan for your site
Arrange Legionella awareness training for your team
Book a Legionella risk assessment, from £185
Add system disinfection, tank cleaning and ongoing maintenance where results call for remedial work
Get in touch through the testing page and we will scope the next step with you.
This article is general information, not a substitute for advice from a qualified doctor. Consult a qualified healthcare professional about your own circumstances before acting on anything here.
FAQ
How often should Legionella sampling be done?
Frequency is set by your risk assessment, not by a single fixed interval. For cooling towers, HSG274 Part 1 recommends periodic Legionella analysis, usually quarterly, alongside weekly biocide monitoring. Higher-risk sites and sites that have just had an incident or remedial work are sampled more often.
How is Legionella treated?
In a water system, Legionella is controlled rather than “treated” once: you restore temperature control, correct biocide dosing and, where results demand it, disinfect the system, then resample to confirm recovery. The written control scheme in HSG274 Part 3 sets out the remedial action plan. Illness in a person is a clinical matter for medical professionals.
What is the procedure for testing for Legionella in water?
Select sample points under BS 7592, collect water in sterile 1-litre containers containing sodium thiosulphate, record temperatures and keep samples cool in transit. A UKAS-accredited laboratory then analyses them by the culture method in BS EN ISO 11731:2017, as HSE describes.
What is an acceptable level of Legionella in water?
Practical guidance treats results up to 100 cfu/l as indicating control, with action rising through 100–1,000 cfu/l and above 1,000 cfu/l. Healthcare guidance treats counts above 100 cfu/l as the point for further steps. Always read a result alongside sample type, location and your temperature records.
Sources
Recommended
Comments