A breath analyzer is only as effective as the policy behind it. When workplace alcohol programmes are successfully challenged, the cause is rarely faulty hardware — it is a policy that was unclear, inconsistently applied, or silent on the exact situation that arose. Here is how to structure pre-shift screening so that it protects your people and survives scrutiny.
Start with the purpose, in writing
A policy that reads as punitive invites resistance; one that states a safety rationale gets cooperation. Open by saying plainly why the programme exists — that impaired operation of vehicles, machinery or safety-critical processes endangers the whole workforce — and that testing is the mechanism for preventing it, not a device for catching people out.
This is not window dressing. When a positive result is disputed, the stated purpose shapes how the response is judged.
Define scope precisely
Ambiguity about who is covered is the most common structural flaw. Specify:
Which roles are subject to testing, defined by function rather than by name — anyone operating a vehicle, anyone working with powered machinery, anyone entering a designated area. Function-based definitions survive reorganisations; name-based lists do not.
Whether contractors, agency staff and visiting drivers are included. They usually should be, and their agencies need to know it before they arrive, not at the gate.
Whether managers and supervisors are covered. Exempting them corrodes the programme's credibility faster than almost anything else.
Define the triggers
State every circumstance in which a test occurs, and test only in those circumstances:
Pre-shift screening, applied consistently to everyone in scope rather than at a supervisor's discretion. Discretionary testing is where allegations of targeting take root.
Random testing, if used, with a selection method that is genuinely random and documented as such.
Post-incident testing, with a defined window and a clear description of what qualifies as an incident.
For-cause testing based on observed behaviour, with the observations recorded at the time by a trained supervisor.
Specify the devices and the tiers
Name the equipment classes, because the device determines what the result can support.
Screening is the first tier: non-contact stations or handheld units that give a rapid pass or alert indication at volume. These are the right tool for clearing a shift change quickly, and the wrong tool for a decision with consequences.
Confirmation is the second tier: a fuel cell analyzer that responds specifically to ethanol and produces a printed or logged result. No disciplinary consequence should ever follow from a screening reading alone. Write that rule into the policy explicitly — it is the single most valuable sentence in the document.
State that confirmatory testing happens promptly, on a calibrated device, and that the subject may request the confirmatory test be repeated.
Set out what happens after a positive result
This is the section that gets written vaguely and then relied upon heavily. It needs to answer, without interpretation:
Who is informed, and in what order. Who removes the person from duty, and how they get home safely — a positive result means they should not be driving themselves, and the policy should say who arranges transport.
What the confirmatory process is, and the timeframe for it.
What the consequences are, distinguishing a first occurrence from a repeat, and whether a support or rehabilitation pathway exists alongside the disciplinary one. Programmes that offer a route back see more voluntary disclosure and fewer people concealing a problem until it causes an incident.
What the appeal route is, and who hears it.
Handle refusals explicitly
Decide in advance how a refusal to test is treated, state it, and apply it consistently. Most organisations treat refusal as equivalent to a positive result. Whatever you choose, the policy must say so before the first refusal occurs, because deciding afterwards looks exactly like what it is.
Documentation and calibration
The programme's credibility rests on records that can be produced months later.
Keep the calibration history for every device: dates of checks and calibrations, the certified standard used, readings before and after adjustment, and who performed the work. Keep the test records: date, time, device identity, result, and the operator. Keep the policy acknowledgements showing each covered worker received and understood the document.
If a device is found to have been out of calibration, every result it produced since its last valid calibration becomes contestable. Treat the calibration schedule as a core control, with a named owner and spare devices so a unit going out for service never leaves a gate uncovered.
Privacy and dignity
Test results are personal information about health-adjacent matters and should be handled accordingly: restricted access, defined retention, and no broader disclosure than the process requires. Conduct testing somewhere that is not a spectacle — a screening station positioned so that a flagged result is not announced to a queue of colleagues is a small design decision with a large effect on how the programme is received.
Train the people administering tests. A supervisor who handles a positive result calmly and privately protects both the individual and the organisation.
Review it
Set a review interval and keep it. Roles change, sites change, devices are replaced, and the law does not stand still. A policy that names a device model retired two years ago signals that nobody is maintaining the programme — which is precisely the impression you do not want when it is examined.
Matching equipment to the structure
The tiers in the policy should map onto real hardware. High-traffic gates suit a fixed wall-mount station such as the A-450 or A-450 Pro, wired where appropriate into access control. Supervisors and mobile teams need rugged handhelds. Confirmatory testing needs a fuel cell analyzer with printing — the A-9000, A-100 or A200 class. Get the policy and the instrument tiers right together and the programme does what it was built to do.