Sampling verifies; it does not control
Temperature control, cleaning, disinfection, removal of dead legs and flushing are the control measures. Sampling tells you whether those measures are working at the points you tested, on the day you tested. It cannot make a system safe, and it cannot substitute for a risk assessment that identifies where the risks are in the first place.
That distinction matters commercially as well as technically. A building can return a negative Legionella result and still have an uninsulated tank at 24°C, a failed TMV and three dead legs — all conditions Part 2 expects you to address regardless of what the laboratory found.
The situations that call for samples
Routine sampling should be defined in the written scheme of control, with sample points and frequency derived from the risk assessment. Beyond the routine round, sampling is expected after cleaning and disinfection, before a new or substantially altered system is returned to service, after a positive result, and after any event that could have introduced contamination.
- Routine sampling as defined in the written scheme of control
- After tank cleaning, disinfection or any remedial work — verification
- New installations and refurbishments, before handover or reoccupation
- After a positive Legionella result, with resampling of the point and its neighbours
- Following flooding, long vacancy, building works or a change of use
- Where control measures have failed or monitoring records have gaps
- Spa pools, whirlpool baths and other aerosol-generating systems, per Part 3
What a sample cannot tell you
Legionella is not distributed evenly through a system. A negative result at one outlet does not mean the tank is clean, and a single positive result does not mean the building is dangerous. Results are a snapshot of the points sampled, interpreted against the system condition, the population using the building and the trend over previous visits.
Low-risk systems with good control may reasonably sample infrequently. Systems serving vulnerable people, systems with little-used outlets, and systems with a history of unsatisfactory results justify more. Anyone quoting a fixed "sample every six months" figure without looking at your building is guessing.
Getting the sample right
A correctly taken sample is worth more than three careless ones. That means the right bottle with the correct neutralising agent, the outlet flushed or not flushed according to the sample type, temperature and residual chlorine measured and recorded at the point of sampling, the sample kept cool and delivered to the laboratory within the required time.
Analysis should be by a laboratory accredited for the method. Interpretation against the alert and action levels in HSG274, and a recommendation on what to do next, is what turns a certificate of analysis into something a facilities manager can act on.
This article is general guidance, not a site-specific risk assessment. Control measures for your building should follow your written scheme of control and the Approved Code of Practice L8 with HSG274.