A breath analyzer is only as effective as the policy behind it. The most common reason workplace alcohol programs get challenged isn't faulty hardware — it's an unclear, inconsistently applied policy.
A defensible pre-shift screening policy spells out who is tested, when, on what device, and what happens after a positive result. It should require a confirmatory second test, document calibration history, and protect the dignity of the worker throughout. Random and post-incident testing should follow the same documented chain.
Pair the policy with the right equipment tier: a fixed wall-mount station for high-traffic gates, a rugged handheld for supervisors, and a clear escalation path to an evidential fuel cell device when a result needs to stand up to scrutiny.
Who is tested, and the answer that causes trouble
"Everyone" is the answer that holds. Any other answer creates a category, and the category is what gets argued about.
That includes supervisors, managers, visiting engineers and contractors. A programme that exempts the people who run it is the one an employment tribunal finds most interesting, and the exemption is what a challenge is built on. If contractors cannot practically be screened at the same gate, write down how they are screened instead — do not leave it unaddressed.
When, and why the timing is the policy
Pre-shift screening exists to stop an impaired person starting work. That only happens if the test is before the work, at a fixed point people pass through anyway.
Post-incident testing is a different instrument with a different purpose, and it needs its own trigger written down in advance: what counts as an incident, how quickly the test follows, and who decides. Deciding that after an incident is how a test becomes an accusation.
Random testing, where it is used, has to be genuinely random and demonstrably so. A selection method somebody could have influenced is not random, whatever the intention.
What happens after a positive
This is the section that carries the whole policy, and the one most often left vague.
A screening positive is not a finding. It leads to a waiting period — typically fifteen minutes with nothing taken by mouth, which clears mouth alcohol — followed by a confirmatory test on an evidential fuel cell device. Only the confirmed result means anything.
Then write down what follows: who is informed, where the person waits, how they get home, and what the process is afterwards. Deciding this in the moment produces inconsistency, and inconsistency is what turns a safety measure into a dispute.
The records that make it defensible
Keep the calibration history for every device, the training record for every operator, the policy version each person was told about, and the log of screens performed.
In a dispute nobody asks whether alcohol testing is reasonable. They ask what you can show about that device, that day, that person, and whether the same thing was done for everyone else. A programme that can answer those four questions rarely gets further than the question.
Dignity, and why it is not a soft point
Screen in a way that does not single anyone out: same point, same procedure, same for everyone, results shown to the person and not to the queue.
This is partly a matter of treating people properly. It is also the practical reason programmes survive. A workforce that regards screening as fair cooperates with it, and cooperation is what keeps a shift change moving. A workforce that regards it as targeted finds ways around it, and then the programme costs money and delivers nothing.
Get the policy and the instrument right together, and the programme protects both your people and your organisation.