---
title: "Keeping records of flushing and sampling activities"
source_url: https://legionella.io/articles/keeping-records-of-flushing-and-sampling-activities/
canonical_url: https://legionella.io/articles/keeping-records-of-flushing-and-sampling-activities/
pillar: "Monitoring, Flushing & Sampling"
summary: "What a flushing or sampling entry must capture to prove Legionella control: who did it, when, the actual reading, and how out-of-range results were closed out."
primary_keyword: "Legionella maintenance records"
date_published: 2025-08-16
date_reviewed: 2026-06-26
author: "Legionella.io editorial team (REMOTE TECH LTD)"
reviewed_against: "HSE L8 and HSG274 guidance"
region: "United Kingdom"
license: "(c) REMOTE TECH LTD. Quote freely with attribution and a link to source_url."
---

# Keeping records of flushing and sampling activities

A flushing record that just says "done" proves nothing. Eighteen months later, when an HSE inspector, an insurer or the manager who takes over from you asks how you know the low-use shower on the third floor stayed safe, "done" is worthless. The reading, the date, the name on the entry, and what happened when something looked wrong — that is the evidence.

Legionella control either becomes defensible in the records or it quietly falls apart there. Flushing, temperature checks, cleaning and sampling are only worth as much as the proof that they happened, fell in range, and were put right when they didn't. The work and the record are the same job.

## Who actually reads these records

Three people open your logbook, and never on a good day. An inspector, after a complaint or a notification. An incident investigator, if a case is ever linked to the building. And the next responsible person, at handover, trying to work out what they have just inherited. Write every entry for those three readers and the standard takes care of itself.

That is also the line between a control and its verification. Flushing, temperature management, cleaning and disinfection are the controls — the things that keep the bacteria from multiplying. Sampling verifies the outcome; it does not create control. HSE is explicit that how often you test depends on the system and the risk assessment, and a clean result never licenses you to ignore poor temperatures or stagnation [4]. The same logic governs the routine side: monitoring and inspection intervals follow the risk assessment and the technical guidance behind it, not habit [2]. So the records have to carry both — the routine evidence that the regime ran, and the exceptions that show how fast you reacted when it didn't.

## What every entry has to capture

The checklist below is the payload. If an entry is missing one of these, it is an entry an investigator can pull apart. Treat the first group as the minimum for any task, then add the rows specific to flushing or sampling.

**Every entry, whatever the task**

- Identify the exact asset or outlet by its unique reference — "Outlet 3B-04", not "the shower upstairs".
- Record the date, and for temperature and sampling, the time.
- Name the person who did the work, so competence and holiday cover are both traceable.
- State what was actually done (for example, both hot and cold run to temperature), not just that a visit happened.
- Capture the measured result with its units — a number, not a tick.
- Mark it in or out of range against the limit your written scheme of control sets [1].

**Flushing entries**

- Note why the outlet is on the flushing list (intermittent use, a void room, a redundant fitting awaiting removal). The reason is what justifies the frequency.
- Record that it was run long enough to draw fresh water through, to the duration your scheme specifies.
- Log every miss and its reason — no access, room sold, contractor no-show. A gap with no note reads as a gap in control.

**Sampling entries**

- Match each sample to a point named in the sampling plan and risk assessment; BS 7592 covers how those sampling sites are selected [5].
- Record the method, including whether it was a pre- or post-flush sample, and who took it.
- Reference the laboratory and its report number, and keep the full report, not just the headline figure.
- Record the result against the action levels in your scheme, and the date any re-sample is due.

**Exceptions and close-out**

- For every out-of-range result or missed task, record the action taken, who did it, and the date it was signed off as closed.
- Link the remedial action back to the trigger that started it, so the chain reads start to finish.
- Record the review point — when this control or outlet gets looked at again.

## Putting it to work without drowning in paper

Start by tying the records back to the two documents that govern them: the current risk assessment and the written scheme of control. The responsible person should be able to point at any control and say why it exists, what result is acceptable, and what happens when the result misses. If the records can't answer that, they are activity logs, not control evidence.

The habit that turns compliance into management is writing down the decision, not just the task. "Outlet 3B-04 is flushed weekly because the room is let intermittently; a missed flush escalates to the responsible person; three misses in a quarter trigger a review of the use pattern." That sentence does more for the next reader than fifty clean ticks.

Paper or digital matters less than findability and retention. Whatever the format, the records have to survive long enough to be useful. L8 guidance is commonly cited as keeping records of monitoring, inspection and test for around five years, with the risk assessment and written scheme retained while they remain current [1] — confirm the exact periods against your own scheme rather than this figure.

## The lines people leave blank

The gaps are predictable, which is the good news — you can hunt for them before an auditor does.

The reading behind the tick is the commonest one. "Temp OK" is not a temperature. So is the missing exception: a logbook with only good days is not a record of a system under control, it is a record of someone only writing things down when they go well. Watch the fortnight your usual checker is on leave — a run of blanks with no cover note is a textbook finding. People also forget to record why an outlet is on the flushing list, so the frequency looks arbitrary the moment it is questioned. Lab reports get filed unread against the action levels, which defeats the point of sampling. And at handover, the records live in a spreadsheet only the departing manager could find — see [Common mistakes in Legionella record keeping](https://legionella.io/articles/common-mistakes-in-legionella-record-keeping/) for more of these failure modes laid out in full.

## Where to start this week

Take one recent week of your flushing and temperature records and read them the way an inspector would, cold. For each entry, ask: can I tell what was done, by whom, what the result was, and — if it missed — what happened next? Mark every entry that fails the test. That list of blanks is your real record-keeping job, and it is almost always shorter and more fixable than starting a new system from scratch. If sentinel points keep showing up in the gaps, [Sentinel outlets](https://legionella.io/articles/sentinel-outlets-what-they-are-and-how-to-monitor-them/) is the place to tighten how those few high-value readings are taken and logged.

## What this can and can't decide for you

This is general guidance on the shape of good records, not a ruling on your numbers. The acceptable temperatures, flushing durations, sampling points, action levels and review intervals all come from your own risk assessment and written scheme, set by a competent person for your building and the people in it — not from a template and not from this page. The records make those site-specific decisions visible and auditable; they do not replace the judgement behind them.

## FAQ

### What must a flushing record actually contain?

Enough to stand alone without the person who wrote it. At minimum: the outlet's unique reference, the date, who did it, what was done (and for how long), the temperature or result with units, whether it was in or out of range, and any action if it missed. A bare "done" or a lone tick is the entry investigators pull apart first.

### How long should we keep flushing and sampling records?

Long enough to cover an audit, a handover and an incident review that could come well after the fact. L8 guidance is commonly cited as retaining monitoring, inspection and test records for around five years, with the risk assessment and written scheme kept while current [1]. Confirm the exact periods against your own scheme rather than relying on a single number.

### If a sample comes back clean, do we still need the flushing records?

Yes. Sampling verifies the outcome at one point on one day; flushing and temperature control are what actually keep the system safe, and HSE is clear that a negative result is not proof you can relax those controls [4]. A clean sample never fills a hole in your control records — the two evidence trails do different jobs.

## Related reading

- [Seasonal changes: managing water temperature in summer and winter](https://legionella.io/articles/seasonal-changes-managing-water-temperature-in-summer-and-winter/)
- [Common mistakes in Legionella record keeping](https://legionella.io/articles/common-mistakes-in-legionella-record-keeping/)
- [Sentinel outlets: what they are and how to monitor them](https://legionella.io/articles/sentinel-outlets-what-they-are-and-how-to-monitor-them/)
- [Outsourcing Legionella testing vs doing it in-house](https://legionella.io/articles/outsourcing-legionella-testing-vs-doing-it-in-house/)

## 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
[4] HSE, "Testing and monitoring your water system for legionella". https://www.hse.gov.uk/legionnaires/testing-monitoring-water-system.htm
[5] BSI, "BS 7592:2022 - Sampling for Legionella bacteria in water systems. Code of practice". https://knowledge.bsigroup.com/products/bs-7592-sampling-for-i-legionella-i-bacteria-in-water-systems-code-of-practice-1