The most dangerous tap in a research building is usually one the law forced you to fit and then told everyone never to touch. Emergency safety showers and eyewash stations sit plumbed, charged and unused for weeks at a time, then deliver warm, standing water straight at a person’s face and eyes the one moment they are actually needed.
Labs are awkward for a different reason than offices or hotels. The building’s hot and cold water is only half the story. The other half — water baths, deionised and reverse-osmosis loops, equipment cooling, autoclave feeds — is specified, owned and serviced by research groups and suppliers, not facilities, and it rarely appears on anyone’s water register. A laboratory Legionella programme that only covers the building plumbing has already missed half its job.
Why a lab is its own kind of risk
Legionella multiplies fastest in the tepid band that HSE guidance puts at roughly 20–45°C [1], and a surprising amount of lab water sits there by design or by neglect. The features that make a research building distinctive are also the ones that drive risk: outlets that go unused for long periods, water held in storage or recirculating for days, and equipment that turns water into a fine spray — all recognised by HSE as the kinds of systems that create foreseeable risk [3]. The route to illness is inhalation of that spray, not drinking [4], which is precisely why an emergency shower matters more than a sink.
Add the rhythm of the place. A university building empties over the summer and at Christmas. A commercial lab cycles with grant funding and project phases, mothballing benches and whole rooms between programmes. Some sites carry process cooling or even a cooling tower serving equipment, which is a higher-risk system in its own right and worth understanding on its own terms — see Case study: outbreak caused by a poorly maintained cooling tower. Put together, the lab is a patchwork of busy and abandoned water, often under different owners, in one building.
Where lab water programmes go wrong
The failures below are not really plumbing failures. They are ownership and assumption failures, and they repeat across research estates.
The emergency shower nobody owns
On paper a safety shower is safety equipment, so it lives on the EHS or COSHH inventory, gets a periodic function test, and never reaches the water log. The split is the whole problem: environmental health owns the shower, facilities owns the water, and the routine flush that would keep it safe falls down the gap between them. Activating a unit to confirm it works is not the same as flushing it to turn over the standing water in the branch that feeds it. Put every safety shower and eyewash station on the water asset register, flush them at the frequency your risk assessment sets, and record who did it on the same system the rest of your monitoring lives on — so a missed flush is visible rather than invisible. A shared digital log makes that crossover practical when two teams are involved; Real-time logging: using mobile devices for water maintenance covers the recording side.
Stopping the assessment at the wall
A competent assessor walks the incoming main, the calorifier, the tanks and the building outlets, signs it off, and never opens a lab door. But a water bath held warm at around body temperature to incubate reagents is, almost by definition, a Legionella-friendly reservoir if it is topped up rather than emptied and cleaned. Deionised and RO loops grow biofilm like any other pipework. Equipment cooling water can recirculate for weeks. These systems are bought by research groups and maintained by equipment suppliers, which is exactly why a building-led assessment skips them. Map them, decide which ones produce aerosol, and write down who owns each one — the principal investigator, the supplier, or facilities.
Assessing the building you commissioned, not the one you run
Many lab risk assessments are written once, at handover, at full design occupancy. Then reality arrives. Outlets that were self-flushing through constant use turn stagnant the week the students leave, and nobody re-flushes them because the paperwork still assumes they are busy. Tie flushing to the real calendar — term dates, vacation periods, known project voids — and trigger a fresh look when a lab is decommissioned or occupancy drops sharply. Stagnation is the common thread behind most of these lapses; Neglected water systems: the danger of stagnation is worth reading alongside this.
Capping the dead bench instead of removing it
A refit strips a bench, disconnects a rig, takes out a sink — and leaves the branch capped at the wall because pulling it back to the main was out of scope. You have just built a textbook dead leg, and it will feed nothing while quietly incubating everything until the next assessment finds it, or doesn’t. When a lab changes, remove redundant pipework back to the live main rather than capping a stub, and log any dead leg you cannot remove yet so it is at least flushed and on the radar.
Confusing your microbiology with Legionella control
This one is peculiar to research sites: a building full of scientists assumes that because it can culture bacteria, Legionella is covered. Sampling for Legionella has its own method and its own meaning. A negative result describes a few outlets at one moment in time; it verifies that your controls are holding, it does not create control, and it is no substitute for keeping hot water hot, cold water cold and water moving. Sample because the risk assessment calls for it and to check the regime, with frequency following the system and the assessment rather than habit [5] — not to discover a problem you could have designed out.
The one fix worth more than the rest
If you do nothing else this quarter, build a single water asset register that crosses the EHS and facilities line: every outlet, every safety shower and eyewash, every water bath and deionised loop, with a named owner and a flush frequency the assessment supports against each. Almost every failure above is the same failure wearing a different coat — water that belongs to no one. Settle the ownership and the schedule follows from it.
Before you act on any of this
Everything here is general guidance, not a control scheme for your building. The temperatures, flush intervals and sampling frequencies that actually apply come out of a competent, site-specific risk assessment carried out in line with L8 and HSG274 [1][2] — and in a lab that assessment has to reach the specialist water systems, not just the building plumbing. If your premises includes a cooling tower, a recirculating spray rig, an animal facility or anything else that produces a fine mist, treat it as a higher-risk system with controls of its own. Where an outlet is genuinely unusual, get a competent person to look at that specific outlet rather than a generic checklist.
Common questions
Do emergency safety showers and eyewash stations really need flushing?
Generally yes. They are textbook low-use outlets — plumbed for safety, used almost never — which is the exact condition Legionella thrives on. The function test that proves the shower works is separate from the flush that turns over the standing water feeding it, and your risk assessment sets how often each happens. Skipping the flush because the unit is “safety equipment” gets the logic backwards.
Who is responsible for water baths and deionised water — facilities or the research team?
It varies by site, and that ambiguity is itself the risk. The duty holder is accountable for the building’s water safety whoever bought the kit, so the workable answer is to assign each specialist system an explicit owner in writing and make sure the risk assessment knows it exists. Equipment serviced by an outside supplier still needs to appear on your register.
Does our lab’s own testing capability cover Legionella?
Not on its own. General microbiology is not Legionella sampling, which follows a specific approach, and sampling of any kind verifies control rather than producing it [5]. The bacteria are managed first by temperature, water movement and cleanliness; testing only tells you whether that management is working.
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, “Systems most likely to create legionella risk”. https://www.hse.gov.uk/legionnaires/risk-systems.htm [4] CDC, “How Legionella Spreads”. https://www.cdc.gov/legionella/causes/index.html [5] HSE, “Testing and monitoring your water system for legionella”. https://www.hse.gov.uk/legionnaires/testing-monitoring-water-system.htm