A cooling tower can make a stranger ill. That is what marks it out from almost every other Legionella risk in a building: the people who fall sick may never have set foot inside, never turned a tap, never known the plant on the roof existed. The bacteria leave in the drift — the fine mist a tower throws off as it sheds heat — and ride the wind down the street.

The reconstruction below is a composite. It is built from the way UK cooling-tower outbreaks tend to unfold, not from any single named incident, and the specifics are illustrative. The failure pattern, though, is real and stubbornly repetitive, which is exactly why it rewards walking through slowly.

How a quiet failure becomes an outbreak

Picture a mixed-use block in a town centre: offices above a supermarket, two evaporative cooling towers on the roof serving the air-conditioning plant. Nothing about it looks dangerous. The towers have been there fifteen years and have never caused trouble.

Late in a warm summer, local GPs and the nearby hospital start seeing more pneumonia than the season explains. Several cases are confirmed as Legionnaires’ disease. UKHSA opens an investigation, interviews patients about where they have been, and plots the cases on a map [5]. They cluster — a loose scatter of homes and workplaces within a few hundred metres, mostly downwind of one block. Environmental health officers and the HSE begin inspecting the cooling towers in that radius. Water samples from the roof of the mixed-use block return Legionella that matches the clinical isolates.

By the time anyone connects the dots, people are seriously unwell. The investigation is now working backwards through records that should have prevented the whole thing.

Where the chain actually broke

No single act caused this. A series of small, individually survivable failures lined up. The duty to assess, control, monitor and review a foreseeable Legionella risk is the spine of the HSE’s Approved Code of Practice, L8 [1] — and every link below is a place that duty quietly came undone.

The tower was not on the local authority’s register. Cooling towers and evaporative condensers must be notified to the local authority, and that notification is part of what lets an outbreak investigation find a suspect device quickly [3]. This one had changed hands when the building was sold, and the paperwork never followed. Investigators found it the slow way.

The dosing had stopped working months earlier. An automatic pump fed biocide into the tower water; when it failed, the reservoir simply ran dry. There was an alarm. Nobody was watching it, and nobody read the dip-slide results that had been creeping upward for weeks. The evidence of lost control existed — it just sat in a folder. That is the precise gap that drives the case for monitoring you cannot ignore; see Wireless data loggers vs manual temperature readings for why a trend nobody reads is barely better than no trend at all.

The drift eliminators were shot. These baffles strip water droplets out of the air the tower discharges, so that even contaminated water does not leave as breathable mist. Years of scale and weather had degraded them. More aerosol escaped, and it escaped further.

The risk assessment was stale and the budget had been trimmed. Contractor visits had quietly dropped from monthly to quarterly to save money. The contractor’s last report had flagged rising bacterial counts. No competent person on the client side owned the job of reading it and acting.

Underneath all of it sat a category error. Evaporative cooling is not a hot-and-cold water system, and it cannot be managed as if it were. The HSE separates it out — HSG274 Part 1 covers cooling systems specifically — because the controls that matter are water treatment, cleanliness, drift control and proper inspection, not tap temperatures [2]. Run a tower on the mental model of “keep it hot, keep it cold” and you miss the controls that actually count.

The reconstruction, in one paragraph

Strip the case to its bones and it reads like this. A high-risk device fell off the corporate memory when the building changed owner; its automatic protection failed silently; the warning signs were recorded but never read; a physical control degraded unnoticed; and cost-cutting removed the visits that might have caught any of it. Each link was ordinary. The chain was lethal. Nothing here needed bad luck — only the absence of anyone whose explicit job was to look.

What this transfers to your site

You do not need an outbreak of your own to learn from one. The transferable checks are short and unglamorous:

  • Confirm the tower is notified to the local authority, and that the notification survived any change of ownership or managing agent [3].
  • Make sure treatment and monitoring evidence is read by a named person who can act, not merely filed. A rising trend nobody acts on is worse than no monitoring, because it looks like control.
  • Treat drift eliminators as a control measure with a condition and a replacement schedule, not as part of the fabric you forget about.
  • Check the risk assessment reflects the system as it is now — BS 8580-1 sets out what a competent assessment covers [6] — and that anyone doing the work is genuinely competent. The Legionella Control Association registration scheme exists to help you judge that [7].
  • Remember that frequency cut to save money is not a saving; it is risk accepted without anyone saying so out loud. That distinction matters most when the HSE later asks why — see Enforcement action: when the HSE comes knocking.

If you want one task off the back of this, pull four things today for every tower you are responsible for: the local-authority notification, the most recent dosing and biocide log, the last dip-slide or sample result, and the date the risk assessment was last reviewed. If any of the four is missing, stale or unread, you have just found the start of your own chain — before it has anywhere to go.

A note on limits

None of this substitutes for a competent assessment of your own tower. Evaporative cooling sits under its own technical regime, and the right biocide programme, drift specification, cleaning intervals and sampling frequency depend on the specific tower, its water supply and how hard it runs. Those are decisions for a competent person working to the Approved Code of Practice [1], not figures to lift from a web page.

FAQ

Do I have to tell anyone the building has a cooling tower?

Yes. Cooling towers and evaporative condensers are notifiable to the local authority, and the duty stays with whoever controls the premises. It is easy to lose track of when a building is sold or a managing agent changes, so confirm the notification is current rather than assuming someone handled it years ago [3].

How far can Legionella travel from a cooling tower?

Further than most people expect. Because the risk is airborne drift rather than water you touch, contaminated mist can be carried on the wind well beyond the building — which is why outbreak investigations map cases across a neighbourhood rather than a single room [4]. The actual reach depends on the tower, the weather and the local layout, so treat any quoted distance as situational.

If the samples come back clear, is the tower in the clear?

No. A sample describes one moment at one point in a system whose counts rise and fall, and a tower losing control can still test low on the day. Sampling supports a control programme; it does not replace temperature, treatment, cleanliness and drift control, and the HSE is explicit that testing frequency should follow the system and the risk assessment rather than a fixed habit [8].

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, “Other duties: RIDDOR and notification of cooling towers or evaporative condensers”. https://www.hse.gov.uk/legionnaires/what-you-must-do/duties.htm [4] CDC, “How Legionella Spreads”. https://www.cdc.gov/legionella/causes/index.html [5] UKHSA, “Investigation of Legionnaires’ disease: cases, clusters and outbreaks”. https://www.gov.uk/government/publications/investigation-of-legionnaires-disease-cases-clusters-and-outbreaks [6] BSI, “BS 8580-1:2019 — Risk assessments for Legionella control. Code of practice”. https://knowledge.bsigroup.com/products/water-quality-risk-assessments-for-legionella-control-code-of-practice-1 [7] Legionella Control Association, “Code of Conduct for Service Providers”. https://www.legionellacontrol.org.uk/ [8] HSE, “Testing and monitoring your water system for legionella”. https://www.hse.gov.uk/legionnaires/testing-monitoring-water-system.htm