Handing out N95s at the gate feels like a programme. It isn't one, and the gap between the two is exactly where an inspector, or a silica exposure years later, finds you.
Key takeaways
A construction firm doing interior demolition, concrete cutting or spray work is going to have workers in respirators at some point, and the regulatory bar for that isn't “issue a mask.” It's a written programme with a hazard-driven selection step, a fit-test cycle that actually runs, and a review that catches when the programme has stopped matching the work.
British Columbia's OHS Regulation Part 8 requires employers to implement “an effective protective equipment programme” covering — in the regulation's own list — “written procedures for selection, use, inspection, cleaning, maintenance and storage,” “instruction and training in the correct use and maintenance of the equipment,” and “medical assessment of respirator wearers, when required”. That's the full scope: a programme, not a purchase order.
CCOHS's guidance on respirator selection makes the same case from the other direction: it walks through hazard identification, engineering-control feasibility and exposure evaluation as the steps that come before a respirator model gets chosen, and states outright that “without a complete respiratory protection programme, workers may not receive the best protection from a respirator, even if it is the correct choice for a specific job”. A written programme is what makes that true rather than aspirational.
Under s. 8.33, the employer, in consultation with the worker and the joint health and safety committee, must select an appropriate respirator against the applicable CSA standard, matched to the hazard, the exposure limit and the maximum-use-concentration calculation the regulation sets out at ss. 8.32 and 8.34. Skipping the consultation step is a common shortcut on a fast-moving site, and it's the one the regulation names first.
That consultation duty ties directly into the hazard-identification step job hazard analysis already requires under s. 7(1) — the assessment that has to happen before work begins is the same assessment that should be driving what respirator gets specified, not a separate exercise done after the fact.
Section 8.40(2.1) sets four triggers, and any one of them restarts the clock: “before initial use of a respirator,” “at least once a year,” “whenever there is a change in respirator facepiece, including the brand, model, and size,” and “whenever changes to the user's physical condition could affect the respirator fit”.
The annual trigger is the one that gets missed, not because anyone disputes it, but because it doesn't attach to a hire date or a project start — it attaches to the anniversary of the last test, per worker, per facepiece. A crew of thirty respirator wearers means thirty separate anniversaries to track, which is exactly the kind of thing a training matrix exists to catch before it becomes a gap an auditor finds.
Section 8.6 adds a second layer: the employer must review the programme itself, including whether the fit-test programme remains adequate and whether further exposure monitoring is needed. A fit test that passed under last year's dust conditions doesn't automatically mean the programme is still sized to this year's work.
Say a trade contractor runs a fifteen-person crew doing interior concrete cutting and coring, all in half-face respirators against silica dust. Five joined in the last eight months and have never been fit tested; the other ten were tested in a batch fourteen months ago. On the s. 8.40(2.1) rule alone, twelve of the fifteen need testing now — the five new hires under the “before initial use” trigger, and the ten veterans because fourteen months is past the annual mark, not before it.
If three of those ten also switched facepiece models this year for comfort reasons, that's an independent trigger for those three regardless of when their last test fell — meaning the true count due right now isn't “the ones over a year,” it's “every trigger, counted separately, per worker.” That distinction is the entire reason a spreadsheet with a single “last tested” column under-reports what's actually due.
Section 8.5's list includes “medical assessment of respirator wearers, when required,” and the qualifier matters: it isn't a blanket annual physical for every wearer, it's a condition that attaches when the respirator type, the work's physical demands, or a worker's own health history make fitness-to-wear a real question — a negative-pressure respirator under sustained physical exertion is a different medical question than an N95 worn for short, light tasks. A programme that never asks the question isn't meeting s. 8.5; it's assuming the answer.
The same section requires training in correct use and maintenance, which is a separate obligation from fit testing — a worker can pass a fit test and still not know how to inspect a facepiece for the cracking or hardening that would fail it on a day the fit test isn't scheduled. CCOHS's guidance folds this into the same programme list as standard operating procedures “available in written form”, alongside inspection, cleaning and storage procedures kept in writing.
There's no published Canadian figure for how many respirator wearers a firm can safely run per supervisor, or a standard cost per fit test — those vary by provider and aren't part of the regulatory text. What the regulation does fix is the trigger logic above; the operational cost of running it is a business decision the regulation leaves to the employer.
Ontario and Alberta impose the same underlying duty through their general hazard-control hierarchy rather than a dedicated respirator-fit-testing clause with BC's specificity — Alberta's OHS Code s. 9(2)–(5) requires the employer to work through engineering controls, then administrative controls, then personal protective equipment, in that order, before a respirator becomes the answer at all. That's the same “engineering and administrative controls first” discipline BC states directly for respirators, just reached through the general hazard-control provision rather than a PPE-specific clause. Neither province publishes a fit-test interval as specific as BC's s. 8.40(2.1); where your work crosses provinces, build the programme to BC's stricter, published cycle rather than assuming a looser one applies elsewhere.
A documented respiratory protection programme is exactly the kind of safety-management evidence a general contractor's prequalification package is built to surface — not because any single clause names respirators, but because a programme with dated fit-test records and a written selection rationale is the difference between a health-and-safety file that reads as real and one that reads as recently assembled for the submission.
If it's being used as respiratory protection against an assessed hazard rather than worn voluntarily for comfort, BC's OHS Regulation treats it the same as any other respirator for programme purposes — selection, fit testing and training obligations attach to the equipment's function, not its price or disposability.
Not the employer alone. Section 8.33 requires the choice to be made in consultation with the worker and the joint health and safety committee — a legitimate stakeholder step, not a formality to document after the model is already ordered.
At minimum yearly, plus immediately on any of three other triggers: first use, a facepiece brand, model or size change, or a change in the wearer's physical condition that could affect fit. A worker who changes facepiece mid-year is due again regardless of when their last annual test happened.
A 30-minute call is enough to tell you whether AI pays for itself here.
