How to set up a health surveillance programme

Health surveillance is not a health check for everybody. It is a targeted, repeating set of checks for people whose work exposes them to a specific hazard, and it exists to catch early signs of harm while something can still be done. Getting it right starts well before anyone sees a clinician.

Start with your risk assessment

Surveillance follows exposure, and exposure is established by your risk assessment. Before booking anything, be able to answer: which tasks expose people to noise, vibration, respirable dust, solvents, skin irritants or sensitisers, and at what level. If the assessment is out of date or generic, fix that first. A surveillance programme built on a guessed exposure profile will either miss people who need it or test people who do not.

Decide who needs surveillance, and for what

Build a list of people by hazard, not by department. One person may need three different types of surveillance; their colleague on the next bench may need none. Where control measures already reduce exposure below the level that triggers surveillance, record that reasoning — you may be asked to justify who you left out as well as who you included.

Choose the right tier of assessment

Most hazards have a tiered approach: a questionnaire for everyone in scope, escalating to a clinician where answers indicate it. Hand-arm vibration is the clearest example, moving from an annual questionnaire up to assessment by a qualified person and then a doctor. Choosing the right tier keeps clinical time where it matters and keeps the cost proportionate.

Book the baseline, then the cycle

Every person in scope needs a baseline before, or as soon as possible after, exposure begins — otherwise you have nothing to compare later results against. Then set the repeat interval and hold to it. Annual is common, but the interval should come from the hazard and the exposure level, not from the calendar being convenient.

Act on the results

Surveillance that produces reports nobody reads is a cost with no benefit. Agree in advance who receives outcomes, what happens when someone is flagged, and how quickly. An outcome that suggests early harm is a signal about your controls, not just about that individual — it should trigger a look at the task, not only a restriction on the person.

Keep the records for the full retention period

Health records must be kept for the statutory period, which for COSHH health records is 40 years from the last entry. These are the records of who was exposed to what and what the outcome was — kept separately from confidential clinical notes. If you change provider, the records come with you: make sure that is written into your contract before you need it.

Getting these six steps in order gives you a programme you can defend at audit and, more importantly, one that catches problems while they are still reversible.