How regularly does the HSE recommend sampling for legionella?

The HSE does not recommend routine legionella sampling for ordinary hot and cold water systems. HSG274, its technical guidance, gives sampling a specific and limited role: to validate that a system is under control when it is commissioned or changed, to verify that control has been restored after a problem, and to investigate when monitoring suggests control may have been lost. What the guidance recommends on a regular basis is monitoring — temperatures, flushing and inspections — recorded in a log.

Why the guidance prefers monitoring to sampling

The logic is worth understanding, because it explains the whole shape of the HSE regime. A laboratory sample is retrospective: it tells you whether legionella was present in that volume of water, at that outlet, on the day it was drawn — and the culture takes around ten to fourteen days to grow, so the answer arrives describing a system as it was a fortnight ago. A temperature reading, by contrast, tells you now whether the condition legionella depends on exists in your system. Hot water reaching at least 50°C at outlets within a minute, cold below 20°C within two minutes, storage at 60°C: while those readings hold, growth is suppressed every day, not just on the day someone took a bottle to a tap.

So the guidance builds the routine on weekly flushing of little-used outlets, monthly checks at sentinel outlets (the temperature checker structures exactly this), and periodic tank and calorifier inspections — the full monitoring schedule. Sampling then slots in where monitoring cannot answer the question.

Where the HSE does expect regular sampling

Evaporative cooling systems — cooling towers and condensers — are the clear exception. Because they control legionella with biocides rather than temperature, and because they spray an aerosol over a wide area, HSG274 Part 1 builds routine sampling into their control regime: quarterly as a general baseline, increased to weekly during commissioning, after a positive result, and while a system that has lost control is being brought back into line. Cooling tower legionella control covers the full regime, which sits alongside notification of the tower to the local authority.

Healthcare is the other case: under HTM 04-01 the water safety plan may set a sampling schedule for systems serving augmented-care units and other susceptible patients, again driven by local risk rather than a universal interval. And spa pools have their own regime under HSG282. Outside those settings, "how regularly" is answered by the risk assessment, not by the calendar.

What ACOP L8 actually requires

ACOP L8 sets the legal-shape duties: assess the risk, prepare a written scheme of control, appoint a competent responsible person, implement and monitor the scheme, and keep records for at least five years. Nothing in it imposes a blanket sampling schedule. That cuts both ways: a duty holder who never samples a simple, well-controlled system has done nothing wrong — but a folder full of lab certificates with no monitoring log, no flushing records and no written scheme is not compliance either. The risk assessment decides where sampling belongs, and the sampling and action levels guide explains what to do with the results when they come back.